Help

FA Qs

Help.FAQs History

Hide minor edits - Show changes to output

Added lines 1-10:
'''What is Ninja Metrics? What makes you different/unique?'''  Ninja Metrics is a group of scientists (http://vwobservatory.org/) who've been working on "big data" and gaming for the past 7 years. We're half social scientists and half computer scientists. Our group has built the field of big data game analysis from scratch, having now published over 80 journal articles on gamer psychology, algorithms and development. A unique expertise is our background in social network analysis. There is simply no other group like us. We've done work for the CIA, the National Science Foundation, and others. The result is a powerful set of techniques and algorithms that reveal more than has ever been possible about gamers, their communities, their value, and their future actions.

'''How are you different than Google Analytics?  Kontagent, KISS Metrics, Mix Panel?'''  There are several basic metrics that any analytics company can deliver--things like ARPU, DAU, k-factor, etc. If you think about it, these are all essentially variations on '''what has happened?''' It's not complex. Some of the more advanced approaches also bring in funnel analysis, segmentation, marketing support, or offer consulting services to data mine. All of these firms collect data via APIs and process them in the cloud. They do a good job, and we do these things too.

'''What's different about us are two key abilities?'''

# We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

# We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.

Deleted lines 102-111:

'''How are you different than Google Analytics?  Kontagent, KISS Metrics, Mix Panel?'''  There are several basic metrics that any analytics company can deliver--things like ARPU, DAU, k-factor, etc. If you think about it, these are all essentially variations on '''what has happened?''' It's not complex. Some of the more advanced approaches also bring in funnel analysis, segmentation, marketing support, or offer consulting services to data mine. All of these firms collect data via APIs and process them in the cloud. They do a good job, and we do these things too.

'''What's different about us are two key abilities?'''

# We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

# We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.

'''What is Ninja Metrics? What makes you different/unique?'''  Ninja Metrics is a group of scientists (http://vwobservatory.org/) who've been working on "big data" and gaming for the past 7 years. We're half social scientists and half computer scientists. Our group has built the field of big data game analysis from scratch, having now published over 80 journal articles on gamer psychology, algorithms and development. A unique expertise is our background in social network analysis. There is simply no other group like us. We've done work for the CIA, the National Science Foundation, and others. The result is a powerful set of techniques and algorithms that reveal more than has ever been possible about gamers, their communities, their value, and their future actions.
Changed line 96 from:
What's different about us are two key abilities?
to:
'''What's different about us are two key abilities?'''
Changed lines 98-100 from:
#, We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

#, We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.
to:
# We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

# We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.
Changed lines 98-100 from:
#1, We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

#2, We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.
to:
#, We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

#, We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.
Changed lines 96-100 from:
->What's different about us are two key abilities?

->#1, We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

->#2, We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.
to:
What's different about us are two key abilities?

#1, We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

#2, We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.
Changed line 95 from:
\\
to:
Changed line 97 from:
\\
to:
Changed line 99 from:
[[<<]]
to:
Changed line 95 from:
//
to:
\\
Changed line 97 from:
//
to:
\\
Changed line 95 from:
[[<<]]
to:
//
Changed line 97 from:
[[<<]]
to:
//
Changed line 32 from:
\\
to:
[[<<]]
Changed lines 1-55 from:
'''Do my timestamps need to be formatted a specific way when sending data?'''

->Yes - Here is an example of a properly formatted timestamp, "2013-11-15T19:07:30.221Z".

'''Are there any required events in order to integrate with Ninja Metrics?'''

->Yes - Required events for any data to display within Katana include Event #1 (Login) and Event #9 (Account Creation).  Event #1 and Event #9 must then have matching account IDs.

'''What timeout parameter does Ninja Metrics use for my users?'''

->Katana's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease.

'''My application has a lot of items.  Do I need to send Katana all of the items in my application?'''

->Within your data there will be some very commonly used items. These make up the bulk of the item use count. The very common actions and items can be ignored if the volume of events is very large. You may feel there is no reason to send the data to Ninja Metrics because the data is so commonly used and you know the impacts to your application. However, there are probably items within your application that are not commonly used or might be new to your application. These uncommon and new items should be sent Ninja Metrics so you will be able to learn more about them. Ultimately it is your decision to send the data that you want more information about.

'''I see you have several Enter/Exit and Begin/End events. Do I need to send both events when sending you data?'''

->Yes, you do need to send both events. An example might be the Recruitment Send and Receive events (EventID #33 & #34). Katana can not calculate K-Factor within both #33 & #34 events. Another example is with Level Begin and Level End (EventID #27 & #28). Katana will not be able to calculate time at level unless both events are received.

'''Do you allow for the use of custom events to track data?'''

->Yes.  Event #54 is the Custom Slot event.  You are permitted to use as many Custom Slot events as you require.  There is no limit. 

'''How can you help us with the integration of your service?'''

->We have a full-time integrations specialist who will help you with the entire process.

'''Where do you store the data?'''

->Securely in the cloud, AWS.

'''Where do I generate the AppPassword / Access Token string?'''

->The Access Token can be generated from the Add Game/App functionality. 

'''How do I know the data I am sending you is valid and accepted by Ninja Metrics?'''

->With every account we provide a link to a Logs page. The page shows the last 12000 events you sent. Whatever events you see there are being logged and they are valid.

'''What about events that may not be valid?'''

->Within the Logs page you can switch to a view showing the last 12000 events received by Ninja Metrics with errors.  These errors are specific and will help you fix any data quality issues happening during the sending of the events.

'''How do I set up and use my development data and production data at the same time?'''

->Simply add another game/application and treat one as your production game and the other as your development environment game.

'''An example of a currency transaction where a user is buying in game gold by spending real world money.'''

->The game features gold as one of its currencies and gold can either be earned by completing certain in-game actions or quests, or gold can be purchased by paying for it using real world currency. A player would decide how much gold they want to purchase and then complete the transaction.  The Currency Transaction Event (EventID #18) would be called and the required parameters would then be sent to Ninja Metrics.  Be sure to set the transaction type to zero in order to indicate the transaction is using real money.

'''An example of a PvP game using Custom Slots (EventID #54).'''

->The image below displays the following scenery:
to:
'''Do my timestamps need to be formatted a specific way when sending data?'''  Yes - Here is an example of a properly formatted timestamp, "2013-11-15T19:07:30.221Z".

'''When do I trigger an event?'''  When an action occurs, you should trigger the corresponding event. I.e.: if a user logs in, then fire the login event.

'''What is an event?'''  An event is an action such as login or logout, creating an account, interacting with a character, buying a sword, sending an invite (or accepting one), etc.

'''Are there any required events in order to integrate with Ninja Metrics?'''  Yes - Required events for any data to display within Katana include Event #1 (Login) and Event #9 (Account Creation).  Event #1 and Event #9 must then have matching account IDs.

'''What timeout parameter does Ninja Metrics use for my users?'''  Katana's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease.

'''My application has a lot of items.  Do I need to send Katana all of the items in my application?'''  Within your data there will be some very commonly used items. These make up the bulk of the item use count. The very common actions and items can be ignored if the volume of events is very large. You may feel there is no reason to send the data to Ninja Metrics because the data is so commonly used and you know the impacts to your application. However, there are probably items within your application that are not commonly used or might be new to your application. These uncommon and new items should be sent Ninja Metrics so you will be able to learn more about them. Ultimately it is your decision to send the data that you want more information about.

'''I see you have several Enter/Exit and Begin/End events. Do I need to send both events when sending you data?'''  Yes, you do need to send both events. An example might be the Recruitment Send and Receive events (EventID #33 & #34). Katana can not calculate K-Factor within both #33 & #34 events. Another example is with Level Begin and Level End (EventID #27 & #28). Katana will not be able to calculate time at level unless both events are received.

'''Do you allow for the use of custom events to track data?'''  Yes.  Event #54 is the Custom Slot event.  You are permitted to use as many Custom Slot events as you require.  There is no limit. 

'''How can you help us with the integration of your service?'''  We have a full-time integrations specialist who will help you with the entire process.

'''Where do you store the data?'''  Securely in the cloud, AWS.

'''Where do I generate the AppPassword / Access Token string?'''  The Access Token can be generated from the Add Game/App functionality. 

'''How do I know the data I am sending you is valid and accepted by Ninja Metrics?'''  With every account we provide a link to a Logs page. The page shows the last 12000 events you sent. Whatever events you see there are being logged and they are valid.

'''What about events that may not be valid?'''  Within the Logs page you can switch to a view showing the last 12000 events received by Ninja Metrics with errors.  These errors are specific and will help you fix any data quality issues happening during the sending of the events.

'''How do I set up and use my development data and production data at the same time?'''  Simply add another game/application and treat one as your production game and the other as your development environment game.

'''An example of a currency transaction where a user is buying in game gold by spending real world money.'''  The game features gold as one of its currencies and gold can either be earned by completing certain in-game actions or quests, or gold can be purchased by paying for it using real world currency. A player would decide how much gold they want to purchase and then complete the transaction.  The Currency Transaction Event (EventID #18) would be called and the required parameters would then be sent to Ninja Metrics.  Be sure to set the transaction type to zero in order to indicate the transaction is using real money.

'''An example of a PvP game using Custom Slots (EventID #54).'''  The image below displays the following scenery:
Changed lines 42-158 from:
'''When do I trigger an event?'''

-> When an action occurs, you should trigger the corresponding event. I.e.: if a user logs in, then fire the login event.

'''What is an event?'''

-> An event is an action such as login or logout, creating an account, interacting with a character, buying a sword, sending an invite (or accepting one), etc.

'''When will the data I send to Katana be available in the interface?'''

->Your data will be ready for review in the Katana Analytics Engine within 24 hours after you sent it.

'''Why does it take 24 hours to process my data?'''

->Your data is being processed through a complicated process that requires intense machine learning algorithms to determine the predictive analytics displayed within Katana. The accuracy of the process is vital for you to have viable information and therefore it can take up to 24 hours.

'''How do submit a ticket to Ninja Metrics?'''

->You can submit a ticket to the Ninja Metrics helpdesk by sending an email to support@ninjametrics.com.

'''What is an active user? '''

->A user who has logged in once in a calendar month.

'''How can I check my current active user count? '''

->In the Katana interface, the Monthly Active Users count can be found in the KPIs under the main dashboard.

'''Social Networks - How does Ninja Metrics determine the ranking of a user within Network Power?'''

->Ninja Metrics looks for trends in your data. Trends which describe how the behavior from a user is impacted by their neighbors in the social network. These trends are then used to quantify/measure '''influence''' of users over others in the network.

->The next step is to breakdown spending from each user and attribute some of it to the social influence from neighbors. Based upon this attribution/breakdown of each users personal spending into social/non-social components we can then figure out the three different metrics in the table. Spending with no influence, influence exerted by me, and influence exerted on me.

'''How do you collect the data from the users about their: country, sex, age and language?'''

->This type of information can be sent to Ninja Metrics within the Account Creation event, Event #9.  We analyze data that you collect - we don't collect data from your users.  This is why we are unique - our analysis is specific to your individual properties.

'''What are the other segmentation possibilities you can provide us with?'''

->We can provide segmentation and cohort analysis based on any data or events that you can make available. This includes: Acquisition source, account start date, specific in-game actions, and even custom A/B testing.

'''Do you track revenue total?'''

->Yes

'''Do you track revenue by application/game? '''

->Yes

'''Do you track revenue by categories (OFD, IAP, Advertising Networks, Country Age, Sex)? '''

->We can break down revenue by most variables. If there is something specific that we don't have in the basic dashboard, you can always export your own data and create your own custom report.

'''Can your system track User Acquisition?'''

->Yes.  Event #59 is the Katana Traffic/Attribution event. 

'''Can we directly acquire users through your platform or need to integrate a third party SDK?'''

->No.  You need to work with third party in order to acquire users.  Ninja Metrics can work with those third parties to directly obtain user acquisition information.

'''Are we able to track all the behavior of the acquired cohorts/users with customized events, segmentation and other parameters available in your platform.'''

->Yes

'''Do you track ARPU, ARPPU, APRDAU (including the revenue from IAPs, OFD and ad network revenue)?'''

->Yes

'''Do you track DAU, MAU etc.?'''

->Yes

'''Do you track New users, unique users?'''

->Yes

'''Do you track Number of sessions, session length, frequencies of use?'''

->Yes

'''Do you track Retention?'''

->Yes

'''Do you track Churn rate?'''

->Yes. We also provide a predictive model for churn so you can take action to keep users who are about to leave.

'''Do you track LTV – what are the possibilities of counting the LTV in your platform?'''

->We can track not only spending of a user over their lifetime, but what they cause other players to spend. We call this Social Value, and it helps identify who your biggest in-game influencers really are (and, combined with our churn predictions, enables you to take action to retain them).

'''Do you track  top devices?'''

->Yes. If you are tracking device version, operating system, or other variables, we can make that segmentation information available to you.

'''Do you have tools for sending and targetting push-up notifications?'''

->We do not - we identify the customers for you to take action in any way you want to.

'''Does your system provide real-time data? What data can we track in real time?'''

->Calculations are displayed within 24 hours. We can provide real-time data but that would be an upcharge

'''Does your system provide us with any A/B testing functionalities?'''

->Yes - we have extensive A/B testing mechanisms as part of our standard package.

'''What`s the time of the data processing after we create funnel, customized event?'''

->Customized events are processed at the same speed as all other events.

'''How are you different than Google Analytics?  Kontagent, KISS Metrics, Mix Panel?'''

->There are several basic metrics that any analytics company can deliver--things like ARPU, DAU, k-factor, etc. If you think about it, these are all essentially variations on '''what has happened?''' It's not complex. Some of the more advanced approaches also bring in funnel analysis, segmentation, marketing support, or offer consulting services to data mine. All of these firms collect data via APIs and process them in the cloud. They do a good job, and we do these things too.
to:
'''When will the data I send to Katana be available in the interface?'''  Your data will be ready for review in the Katana Analytics Engine within 24 hours after you sent it.

'''Why does it take 24 hours to process my data?''' Your data is being processed through a complicated process that requires intense machine learning algorithms to determine the predictive analytics displayed within Katana. The accuracy of the process is vital for you to have viable information and therefore it can take up to 24 hours.

'''How do submit a ticket to Ninja Metrics?'''  You can submit a ticket to the Ninja Metrics helpdesk by sending an email to support@ninjametrics.com.

'''What is an active user? '''  A user who has logged in once in a calendar month.

'''How can I check my current active user count? '''  In the Katana interface, the Monthly Active Users count can be found in the KPIs under the main dashboard.

'''Social Networks - How does Ninja Metrics determine the ranking of a user within Network Power?'''  Ninja Metrics looks for trends in your data. Trends which describe how the behavior from a user is impacted by their neighbors in the social network. These trends are then used to quantify/measure '''influence''' of users over others in the network.  The next step is to breakdown spending from each user and attribute some of it to the social influence from neighbors. Based upon this attribution/breakdown of each users personal spending into social/non-social components we can then figure out the three different metrics in the table. Spending with no influence, influence exerted by me, and influence exerted on me.

'''How do you collect the data from the users about their: country, sex, age and language?'''  This type of information can be sent to Ninja Metrics within the Account Creation event, Event #9.  We analyze data that you collect - we don't collect data from your users.  This is why we are unique - our analysis is specific to your individual properties.

'''What are the other segmentation possibilities you can provide us with?'''  We can provide segmentation and cohort analysis based on any data or events that you can make available. This includes: Acquisition source, account start date, specific in-game actions, and even custom A/B testing.

'''Do you track revenue total?'''  Yes

'''Do you track revenue by application/game? '''  Yes

'''Do you track revenue by categories (OFD, IAP, Advertising Networks, Country Age, Sex)? '''  We can break down revenue by most variables. If there is something specific that we don't have in the basic dashboard, you can always export your own data and create your own custom report.

'''Can your system track User Acquisition?'''  Yes.  Event #59 is the Katana Traffic/Attribution event. 

'''Can we directly acquire users through your platform or need to integrate a third party SDK?'''  No.  You need to work with third party in order to acquire users.  Ninja Metrics can work with those third parties to directly obtain user acquisition information.

'''Are we able to track all the behavior of the acquired cohorts/users with customized events, segmentation and other parameters available in your platform.'''  Yes

'''Do you track ARPU, ARPPU, APRDAU (including the revenue from IAPs, OFD and ad network revenue)?'''  Yes

'''Do you track DAU, MAU etc.?'''  Yes

'''Do you track New users, unique users?'''  Yes

'''Do you track Number of sessions, session length, frequencies of use?'''  Yes

'''Do you track Retention?'''  Yes

'''Do you track Churn rate?'''  Yes. We also provide a predictive model for churn so you can take action to keep users who are about to leave.

'''Do you track LTV – what are the possibilities of counting the LTV in your platform?'''  We can track not only spending of a user over their lifetime, but what they cause other players to spend. We call this Social Value, and it helps identify who your biggest in-game influencers really are (and, combined with our churn predictions, enables you to take action to retain them).

'''Do you track  top devices?'''  Yes. If you are tracking device version, operating system, or other variables, we can make that segmentation information available to you.

'''Do you have tools for sending and targetting push-up notifications?'''  We do not - we identify the customers for you to take action in any way you want to.

'''Does your system provide real-time data? What data can we track in real time?'''  Calculations are displayed within 24 hours. We can provide real-time data but that would be an upcharge

'''Does your system provide us with any A/B testing functionalities?'''  Yes - we have extensive A/B testing mechanisms as part of our standard package.

'''What`s the time of the data processing after we create funnel, customized event?'''  Customized events are processed at the same speed as all other events.

'''How are you different than Google Analytics?  Kontagent, KISS Metrics, Mix Panel?'''  There are several basic metrics that any analytics company can deliver--things like ARPU, DAU, k-factor, etc. If you think about it, these are all essentially variations on '''what has happened?''' It's not complex. Some of the more advanced approaches also bring in funnel analysis, segmentation, marketing support, or offer consulting services to data mine. All of these firms collect data via APIs and process them in the cloud. They do a good job, and we do these things too.
Changed line 96 from:
->What's different about us are two key abilities.
to:
->What's different about us are two key abilities?
Changed lines 102-131 from:
'''What is Ninja Metrics? What makes you different/unique?'''

->Ninja Metrics is a group of scientists
(http://vwobservatory.org/) who've been working on "big data" and gaming for the past 7 years. We're half social scientists and half computer scientists. Our group has built the field of big data game analysis from scratch, having now published over 80 journal articles on gamer psychology, algorithms and development. A unique expertise is our background in social network analysis. There is simply no other group like us. We've done work for the CIA, the National Science Foundation, and others. The result is a powerful set of techniques and algorithms that reveal more than has ever been possible about gamers, their communities, their value, and their future actions.

'''Do you share the data I send you?'''

->No. We do not share your data with anybody outside of Ninja
Metrics. Internally, our data scientist may analyze your data in order to improve our algorithms.

'''What are you doing with the data I am sending you?'''

->The data you send is processed by or modeling software in order to derive your
results, all of which are showcased in Katana.

'''How long do you keep my data?'''

->We keep your data indefinitely and it is available in Katana for you to compare different metrics through
different periods of time.

'''Is my data secure?'''

->Yes. Ninja Metrics conducts continuously internal security audits at all
levels. From the moment you make a call we safeguard your data.

'''Can I use Google Analytics and Katana at the same time?'''

->Yes. There is no conflict in using
both. Some of the instrumentation will be similar, so you can leverage what you might have already done with GA in favor of Katana.

'''Why should I track all this data and information?'''

->By tracking the activity in your game you can gain a great deal of insight into what’s
working, what’s not, where to improve and where to leave things alone. Furthermore, by tracking activity you can see what your users are doing, how they interact with the game, what they buy, who influences buying and how groups of users influence each other.

'''Which browsers do you support?'''
to:
'''What is Ninja Metrics? What makes you different/unique?'''  Ninja Metrics is a group of scientists (http://vwobservatory.org/) who've been working on "big data" and gaming for the past 7 years. We're half social scientists and half computer scientists. Our group has built the field of big data game analysis from scratch, having now published over 80 journal articles on gamer psychology, algorithms and development. A unique expertise is our background in social network analysis. There is simply no other group like us. We've done work for the CIA, the National Science Foundation, and others. The result is a powerful set of techniques and algorithms that reveal more than has ever been possible about gamers, their communities, their value, and their future actions.

'''Do you share the data I send you?'''  No. We do not share your data with anybody outside of Ninja Metrics. Internally, our data scientist may analyze your data in order to improve our algorithms.

'''What are you doing with the data I am sending you?'''  The data you send is processed by or modeling software in order to derive your results, all of which are showcased in Katana.

'''How long do you keep my data?'''  We keep your data indefinitely and it is available in Katana for you to compare different metrics through different periods of time.

'''Is my data secure?'''  Yes. Ninja Metrics conducts continuously internal security audits at all levels. From the moment you make a call we safeguard your data.

'''Can I use Google Analytics and Katana at the same time?'''  Yes. There is no conflict in using both. Some of the instrumentation will be similar, so you can leverage what you might have already done with GA in favor of Katana.

'''Why should I track all this data and information?'''  By tracking the activity in your game you can gain a great deal of insight into what’s working, what’s not, where to improve and where to leave things alone. Furthermore, by tracking activity you can see what your users are doing, how they interact with the game, what they buy, who influences buying and how groups of users influence each other.

'''Which browsers do you support?''' 
Changed lines 125-136 from:
'''How do I sign into Katana?'''

->You can securely log into Katana
here:  https://accounts.ninjametrics.com/user/login

'''Link to the Ninja Metrics Privacy Statement.'''

->Privacy Statement
:  http://www.ninjametrics.com/privacy.html

'''Link to the Ninja Metrics Terms of Service.'''

->Link to the TOS:
  https://accounts.ninjametrics.com/terms-and-conditions
to:
'''How do I sign into Katana?'''  You can securely log into Katana here:  https://accounts.ninjametrics.com/user/login

'''Link to the Ninja Metrics Privacy Statement.'''  http://www.ninjametrics.com/privacy.html

'''Link to the Ninja Metrics Terms of Service.'''  https://accounts.ninjametrics.com/terms-and-conditions
Changed lines 143-149 from:
'''How do I make sure the items I use in my application are displayed within Katana? '''

->You need to initialize all items used within your application that you want Katana to analyze and
track.  Use Event #15 to initialize each item and then trigger Event #16 (Item Use) to record the use of each item.  You do not need to use Event #16 for items being purchased using Event #18 but the items being purchased must be initialized before the purchase.

'''An example of buying items using virtual or in-game currency.'''

->
Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
'''How do I make sure the items I use in my application are displayed within Katana? '''  You need to initialize all items used within your application that you want Katana to analyze and track.  Use Event #15 to initialize each item and then trigger Event #16 (Item Use) to record the use of each item.  You do not need to use Event #16 for items being purchased using Event #18 but the items being purchased must be initialized before the purchase.

'''An example of buying items using virtual or in-game currency.'''  Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
Changed line 63 from:
https://developer.ninjametrics.com/uploads/Help/PvP2.png
to:
%width=100pct%https://developer.ninjametrics.com/uploads/Help/PvP2.png
Changed lines 249-271 from:
->What's nice about machine learning models is that they deal with all of the details and permutations and simply give you the likelihood. You can then think of this likelihood like a %, though it's actually slightly more powerful than that. In many cases, our confidence numbers are ""F-scores,"" which are more reliable than a %. An F-score rules out both false positives and false negatives. This makes it an extremely ""conservative"" number, and more trustworthy. It's also going to be lower than a %. For example, F-scores in the phone industry for user churn tend to hit around .35 or .4. Numbers over .6 are very strong."

'''Individual & Social Values'''

->Katana's super-power is a combination of predictive analytics and social network analysis that lets it tell you who your most important users are. Unlike other approaches, Katana splits every user's value into two distinct components:
[[<<]]
->#1: Their Lifetime Value (LTV), or how much they will spend before they leave someday
->#2: Their Social Value, which is how much they drive other users to spend.
[[<<]]
->Put together, these two values tell you how much any one user contributes to your bottom line. If that user were to disappear tomorrow, this is how much future revenue would disappear.
[[<<]]
->The users who have a LTV are commonly known as ""whales."" We call the users with high Social Value ""Social Whales."" They are usually a minority, and usually not the same people as the regular whales.
[[<<]]
->Katana lets you take advantage of these users in three ways:
[[<<]]
->#1: Be smart about finding more of them.
->#2: Monetize them when you have them.
->#3: Focus on retaining them over others.
[[<<]]
->When you've identified your whales or Social Whales, you should consider doing something with them. Katana lets you sort and export custom lists of these users from the dashboard so that you can run custom promotions, messaging or experiences on them. Our testing system is under construction, and will enable you to sample these users, test interventions on them, then see the results on both them and their friends.
[[<<]]
->When we look at social networks we use the metaphor of a '''ripple on the pond.'''  If you start the intervention at the right person and do it the right way, you may see viral effects. Katana will quantify these for you."
to:
->What's nice about machine learning models is that they deal with all of the details and permutations and simply give you the likelihood. You can then think of this likelihood like a %, though it's actually slightly more powerful than that. In many cases, our confidence numbers are ""F-scores,"" which are more reliable than a %. An F-score rules out both false positives and false negatives. This makes it an extremely ""conservative"" number, and more trustworthy. It's also going to be lower than a %. For example, F-scores in the phone industry for user churn tend to hit around .35 or .4. Numbers over .6 are very strong."
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 183 from:
to:
[[<<]]
Changed line 185 from:
to:
[[<<]]
Changed line 187 from:
to:
[[<<]]
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 244 from:
\\
to:
[[<<]]
Changed line 246 from:
\\
to:
[[<<]]
Changed line 248 from:
\\
to:
[[<<]]
Changed line 254 from:
\\
to:
[[<<]]
Changed line 257 from:
\\
to:
[[<<]]
Changed line 259 from:
\\
to:
[[<<]]
Changed line 261 from:
\\
to:
[[<<]]
Changed line 263 from:
\\
to:
[[<<]]
Changed line 267 from:
\\
to:
[[<<]]
Changed line 269 from:
\\
to:
[[<<]]
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 242 from:
to:
[[<<]]
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 241-242 from:
->A model is a way of building up a pattern. For example, a model may say that '''men over 40 are more likely to visit the doctor, '''which would require knowing gender, age and doctor visits for a lot of men. A confidence number indicates how well the model is performing over time, i.e. how accurate that model really is. When you're looking into the future, the models will never be 100%, but they can become very strong.
to:
->A model is a way of building up a pattern. For example, a model may say that men over 40 are more likely to visit the doctor, which would require knowing gender, age and doctor visits for a lot of men. A confidence number indicates how well the model is performing over time, i.e. how accurate that model really is. When you're looking into the future, the models will never be 100%, but they can become very strong.
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Added lines 242-243:

->Our projection models are known as '''machine learning models,''' which means that the software looks for patterns and gets better at recognizing and forecasting them over time. Here's a simple way of understanding them. Say that there are four events called A, B, C, and D. The software notices that they often happen in order: A, then B, then C, then D. So, when it sees A, B and C occur, it gives a likelihood that D is going to happen next. When you think that '''D''' could be Churn or Conversion or Virality or the spending driven socially by a person (Social Value), or any number of things, you can appreciate the power of these models.
Deleted lines 244-245:
->Our projection models are known as '''machine learning models,''' which means that the software looks for patterns and gets better at recognizing and forecasting them over time. Here's a simple way of understanding them. Say that there are four events called A, B, C, and D. The software notices that they often happen in order: A, then B, then C, then D. So, when it sees A, B and C occur, it gives a likelihood that D is going to happen next. When you think that '''D''' could be Churn or Conversion or Virality or the spending driven socially by a person (Social Value), or any number of things, you can appreciate the power of these models.
\\
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 241 from:
->A model is a way of building up a pattern. For example, a model may say that ""men over 40 are more likely to visit the doctor,"" which would require knowing gender, age and doctor visits for a lot of men. A confidence number indicates how well the model is performing over time, i.e. how accurate that model really is. When you're looking into the future, the models will never be 100%, but they can become very strong.
to:
->A model is a way of building up a pattern. For example, a model may say that '''men over 40 are more likely to visit the doctor, '''which would require knowing gender, age and doctor visits for a lot of men. A confidence number indicates how well the model is performing over time, i.e. how accurate that model really is. When you're looking into the future, the models will never be 100%, but they can become very strong.
Changed line 243 from:
->Our projection models are known as ""machine learning models,"" which means that the software looks for patterns and gets better at recognizing and forecasting them over time. Here's a simple way of understanding them. Say that there are four events called A, B, C, and D. The software notices that they often happen in order: A, then B, then C, then D. So, when it sees A, B and C occur, it gives a likelihood that D is going to happen next. When you think that ""D"" could be Churn or Conversion or Virality or the spending driven socially by a person (Social Value), or any number of things, you can appreciate the power of these models.
to:
->Our projection models are known as '''machine learning models,''' which means that the software looks for patterns and gets better at recognizing and forecasting them over time. Here's a simple way of understanding them. Say that there are four events called A, B, C, and D. The software notices that they often happen in order: A, then B, then C, then D. So, when it sees A, B and C occur, it gives a likelihood that D is going to happen next. When you think that '''D''' could be Churn or Conversion or Virality or the spending driven socially by a person (Social Value), or any number of things, you can appreciate the power of these models.
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 242 from:
to:
\\
Changed line 244 from:
to:
\\
Changed line 246 from:
to:
\\
Changed line 248 from:
to:
\\
Changed line 254 from:
to:
\\
Changed line 257 from:
to:
\\
Changed line 259 from:
to:
\\
Changed line 261 from:
to:
\\
Changed line 263 from:
to:
\\
Changed line 267 from:
to:
\\
Changed line 269 from:
to:
\\
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 231-234 from:
'''Link to the Ninja Metrics Privacy Statement. http://www.ninjametrics.com/privacy.html

'''Link to the Ninja Metrics Terms of Service. https://accounts.ninjametrics.com/terms-and-conditions
to:
'''Link to the Ninja Metrics Privacy Statement.'''

->Privacy Statement
:  http://www.ninjametrics.com/privacy.html

'''Link to the Ninja Metrics Terms of Service.'''

->Link to the TOS: 
https://accounts.ninjametrics.com/terms-and-conditions
Changed line 279 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 221-226 from:
\\
->
The latest version of Internet Explorer and versions 8, 9, 10.
->Firefox, latest version.
->Chrome, latest version.
->Safari, latest version.
to:

*The latest version of Internet Explorer and versions 8, 9, 10.
*Firefox, latest version.
*Chrome, latest version.
*Safari, latest version.
Changed line 275 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 221 from:
to:
\\
Changed line 275 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 220-221 from:
->Supported Browsers:
\\
to:
->Supported Browsers -
Changed line 275 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 225-226 from:
->Safari, latest version."
to:
->Safari, latest version.
Changed line 275 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 221-222 from:

to:
\\
Changed line 275 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Added line 222:
Changed line 276 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 221 from:
\\
to:
Changed line 275 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 220 from:
->"Supported Browsers:
to:
->Supported Browsers:
Changed line 275 from:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
->Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 214-226 from:
'''Why should I track all this data and information? By tracking the activity in your game you can gain a great deal of insight into what’s working, what’s not, where to improve and where to leave things alone. Furthermore, by tracking activity you can see what your users are doing, how they interact with the game, what they buy, who influences buying and how groups of users influence each other.

'''Which browsers do you support? "Supported Browsers:

• The latest version of Internet Explorer and versions 8, 9, 10.
Firefox, latest version.
Chrome, latest version.
Safari, latest version."

'''How do I sign into Katana? "You can securely log into Katana here:

https://accounts.ninjametrics.com/user/login "
to:
'''Why should I track all this data and information?'''

->
By tracking the activity in your game you can gain a great deal of insight into what’s working, what’s not, where to improve and where to leave things alone. Furthermore, by tracking activity you can see what your users are doing, how they interact with the game, what they buy, who influences buying and how groups of users influence each other.

'''Which browsers do you support?'''

->
"Supported Browsers:
\\
->The
latest version of Internet Explorer and versions 8, 9, 10.
->Firefox, latest version.
->Chrome, latest version.
->Safari, latest version."

'''How do I sign into Katana?'''

->
You can securely log into Katana here:  https://accounts.ninjametrics.com/user/login
Changed lines 235-262 from:
'''Please provide some details about Confidence Scores, Projections, and Machine Learning. "A model is a way of building up a pattern. For example, a model may say that ""men over 40 are more likely to visit the doctor,"" which would require knowing gender, age and doctor visits for a lot of men. A confidence number indicates how well the model is performing over time, i.e. how accurate that model really is. When you're looking into the future, the models will never be 100%, but they can become very strong.

Our projection models are known as ""machine learning models,"" which means that the software looks for patterns and gets better at recognizing and forecasting them over time. Here's a simple way of understanding them. Say that there are four events called A, B, C, and D. The software notices that they often happen in order: A, then B, then C, then D. So, when it sees A, B and C occur, it gives a likelihood that D is going to happen next. When you think that ""D"" could be Churn or Conversion or Virality or the spending driven socially by a person (Social Value), or any number of things, you can appreciate the power of these models.

They're also quite complex. It may be that D will come next if it's a Friday, or if C is an action taken by a woman and not a man, or any of a million other possibilities. For the system to reach these conclusions, it needs to see many actions. So, on day 1, the confidence scores are going to be quite low. Over time--sometimes a few days and sometimes a week or two--they will improve to a point where they'll level off because the patterns the system is seeing aren't changing much--though the individuals in the system may be. If there's some kind of change in the system, e.g. activity drops due to a seasonal pattern or some large event, the model will adapt, take a slight hit in confidence, then climb back to the peak.

This need for seeing actions leads many to either wait for the models to get strong, or to implement the analytics as early as possible (betas are great), or to load in historical data to get good results immediately.

What's nice about machine learning models is that they deal with all of the details and permutations and simply give you the likelihood. You can then think of this likelihood like a %, though it's actually slightly more powerful than that. In many cases, our confidence numbers are ""F-scores,"" which are more reliable than a %. An F-score rules out both false positives and false negatives. This makes it an extremely ""conservative"" number, and more trustworthy. It's also going to be lower than a %. For example, F-scores in the phone industry for user churn tend to hit around .35 or .4. Numbers over .6 are very strong."

'''Individual & Social Values Katana's super-power is a combination of predictive analytics and social network analysis that lets it tell you who your most important users are. Unlike other approaches, Katana splits every user's value into two distinct components:

#1: Their Lifetime Value (LTV), or how much they will spend before they leave someday
#2: Their Social Value, which is how much they drive other users to spend.

Put together, these two values tell you how much any one user contributes to your bottom line. If that user were to disappear tomorrow, this is how much future revenue would disappear.

The users who have a LTV are commonly known as ""whales."" We call the users with high Social Value ""Social Whales."" They are usually a minority, and usually not the same people as the regular whales.

Katana lets you take advantage of these users in three ways:

#1: Be smart about finding more of them.
#2: Monetize them when you have them.
#3: Focus on retaining them over others.

When you've identified your whales or Social Whales, you should consider doing something with them. Katana lets you sort and export custom lists of these users from the dashboard so that you can run custom promotions, messaging or experiences on them. Our testing system is under construction, and will enable you to sample these users, test interventions on them, then see the results on both them and their friends.

When we look at social networks we use the metaphor of a ""ripple on the pond."" If you start the intervention at the right person and do it the right way, you may see viral effects. Katana will quantify these for you."
to:
'''Please provide some details about Confidence Scores, Projections, and Machine Learning.'''

->
A model is a way of building up a pattern. For example, a model may say that ""men over 40 are more likely to visit the doctor,"" which would require knowing gender, age and doctor visits for a lot of men. A confidence number indicates how well the model is performing over time, i.e. how accurate that model really is. When you're looking into the future, the models will never be 100%, but they can become very strong.

->Our projection models are known as ""machine learning models,"" which means that the software looks for patterns and gets better at recognizing and forecasting them over time. Here's a simple way of understanding them. Say that there are four events called A, B, C, and D. The software notices that they often happen in order: A, then B, then C, then D. So, when it sees A, B and C occur, it gives a likelihood that D is going to happen next. When you think that ""D"" could be Churn or Conversion or Virality or the spending driven socially by a person (Social Value), or any number of things, you can appreciate the power of these models.

->They're also quite complex. It may be that D will come next if it's a Friday, or if C is an action taken by a woman and not a man, or any of a million other possibilities. For the system to reach these conclusions, it needs to see many actions. So, on day 1, the confidence scores are going to be quite low. Over time--sometimes a few days and sometimes a week or two--they will improve to a point where they'll level off because the patterns the system is seeing aren't changing much--though the individuals in the system may be. If there's some kind of change in the system, e.g. activity drops due to a seasonal pattern or some large event, the model will adapt, take a slight hit in confidence, then climb back to the peak.

->This need for seeing actions leads many to either wait for the models to get strong, or to implement the analytics as early as possible (betas are great), or to load in historical data to get good results immediately.

->What's nice about machine learning models is that they deal with all of the details and permutations and simply give you the likelihood. You can then think of this likelihood like a %, though it's actually slightly more powerful than that. In many cases, our confidence numbers are ""F-scores,"" which are more reliable than a %. An F-score rules out both false positives and false negatives. This makes it an extremely ""conservative"" number, and more trustworthy. It's also going to be lower than a %. For example, F-scores in the phone industry for user churn tend to hit around .35 or .4. Numbers over .6 are very strong."

'''Individual & Social Values'''

->Katana's super-power is a combination of predictive analytics and social network analysis that lets it tell you who your most important users are. Unlike other approaches, Katana splits every user's value into two distinct components:

->#1: Their Lifetime Value (LTV), or how much they will spend before they leave someday
->#2: Their Social Value, which is how much they drive other users to spend.

->Put together, these two values tell you how much any one user contributes to your bottom line. If that user were to disappear tomorrow, this is how much future revenue would disappear.

->The users who have a LTV are commonly known as ""whales."" We call the users with high Social Value ""Social Whales."" They are usually a minority, and usually not the same people as the regular whales.

->Katana lets you take advantage of these users in three ways:

->#1: Be smart about finding more of them.
->#2: Monetize them when you have them.
->#3: Focus on retaining them over others.

->When you've identified your whales or Social Whales, you should consider doing something with them. Katana lets you sort and export custom lists of these users from the dashboard so that you can run custom promotions, messaging or experiences on them. Our testing system is under construction, and will enable you to sample these users, test interventions on them, then see the results on both them and their friends.

->When we look at social networks we use the metaphor of a '''ripple on the pond.''' If you start the intervention at the right person and do it the right way, you may see viral effects. Katana will quantify these for you."
Changed lines 269-271 from:
'''How do I make sure the items I use in my application are displayed within Katana? You need to initialize all items used within your application that you want Katana to analyze and track.  Use Event #15 to initialize each item and then trigger Event #16 (Item Use) to record the use of each item.  You do not need to use Event #16 for items being purchased using Event #18 but the items being purchased must be initialized before the purchase.

'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
'''How do I make sure the items I use in my application are displayed within Katana? '''

->
You need to initialize all items used within your application that you want Katana to analyze and track.  Use Event #15 to initialize each item and then trigger Event #16 (Item Use) to record the use of each item.  You do not need to use Event #16 for items being purchased using Event #18 but the items being purchased must be initialized before the purchase.

'''An example of buying items using virtual or in-game currency.'''

->
Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 56 from:
to:
\\
Changed line 263 from:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 56-58 from:
\\
https://developer.ninjametrics.com/uploads/Help/PvP2.png
\\
to:
Changed lines 62-65 from:
to:
\\
https://developer.ninjametrics.com/uploads/Help/PvP2.png
\\

Changed line 263 from:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed line 56 from:
to:
\\
Changed lines 58-60 from:

 [[(Attach:)pvp54.png]]
to:
\\
Changed line 262 from:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Added lines 57-58:
https://developer.ninjametrics.com/uploads/Help/PvP2.png
Changed line 264 from:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
to:
'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency. 
Changed lines 63-262 from:
->Total number of times/account 9 has killed minion 2."
to:
->Total number of times/account 9 has killed minion 2."

'''When do I trigger an event?'''

-> When an action occurs, you should trigger the corresponding event. I.e.: if a user logs in, then fire the login event.

'''What is an event?'''

-> An event is an action such as login or logout, creating an account, interacting with a character, buying a sword, sending an invite (or accepting one), etc.

'''When will the data I send to Katana be available in the interface?'''

->Your data will be ready for review in the Katana Analytics Engine within 24 hours after you sent it.

'''Why does it take 24 hours to process my data?'''

->Your data is being processed through a complicated process that requires intense machine learning algorithms to determine the predictive analytics displayed within Katana. The accuracy of the process is vital for you to have viable information and therefore it can take up to 24 hours.

'''How do submit a ticket to Ninja Metrics?'''

->You can submit a ticket to the Ninja Metrics helpdesk by sending an email to support@ninjametrics.com.

'''What is an active user? '''

->A user who has logged in once in a calendar month.

'''How can I check my current active user count? '''

->In the Katana interface, the Monthly Active Users count can be found in the KPIs under the main dashboard.

'''Social Networks - How does Ninja Metrics determine the ranking of a user within Network Power?'''

->Ninja Metrics looks for trends in your data. Trends which describe how the behavior from a user is impacted by their neighbors in the social network. These trends are then used to quantify/measure '''influence''' of users over others in the network.

->The next step is to breakdown spending from each user and attribute some of it to the social influence from neighbors. Based upon this attribution/breakdown of each users personal spending into social/non-social components we can then figure out the three different metrics in the table. Spending with no influence, influence exerted by me, and influence exerted on me.

'''How do you collect the data from the users about their: country, sex, age and language?'''

->This type of information can be sent to Ninja Metrics within the Account Creation event, Event #9.  We analyze data that you collect - we don't collect data from your users.  This is why we are unique - our analysis is specific to your individual properties.

'''What are the other segmentation possibilities you can provide us with?'''

->We can provide segmentation and cohort analysis based on any data or events that you can make available. This includes: Acquisition source, account start date, specific in-game actions, and even custom A/B testing.

'''Do you track revenue total?'''

->Yes

'''Do you track revenue by application/game? '''

->Yes

'''Do you track revenue by categories (OFD, IAP, Advertising Networks, Country Age, Sex)? '''

->We can break down revenue by most variables. If there is something specific that we don't have in the basic dashboard, you can always export your own data and create your own custom report.

'''Can your system track User Acquisition?'''

->Yes.  Event #59 is the Katana Traffic/Attribution event. 

'''Can we directly acquire users through your platform or need to integrate a third party SDK?'''

->No.  You need to work with third party in order to acquire users.  Ninja Metrics can work with those third parties to directly obtain user acquisition information.

'''Are we able to track all the behavior of the acquired cohorts/users with customized events, segmentation and other parameters available in your platform.'''

->Yes

'''Do you track ARPU, ARPPU, APRDAU (including the revenue from IAPs, OFD and ad network revenue)?'''

->Yes

'''Do you track DAU, MAU etc.?'''

->Yes

'''Do you track New users, unique users?'''

->Yes

'''Do you track Number of sessions, session length, frequencies of use?'''

->Yes

'''Do you track Retention?'''

->Yes

'''Do you track Churn rate?'''

->Yes. We also provide a predictive model for churn so you can take action to keep users who are about to leave.

'''Do you track LTV – what are the possibilities of counting the LTV in your platform?'''

->We can track not only spending of a user over their lifetime, but what they cause other players to spend. We call this Social Value, and it helps identify who your biggest in-game influencers really are (and, combined with our churn predictions, enables you to take action to retain them).

'''Do you track  top devices?'''

->Yes. If you are tracking device version, operating system, or other variables, we can make that segmentation information available to you.

'''Do you have tools for sending and targetting push-up notifications?'''

->We do not - we identify the customers for you to take action in any way you want to.

'''Does your system provide real-time data? What data can we track in real time?'''

->Calculations are displayed within 24 hours. We can provide real-time data but that would be an upcharge

'''Does your system provide us with any A/B testing functionalities?'''

->Yes - we have extensive A/B testing mechanisms as part of our standard package.

'''What`s the time of the data processing after we create funnel, customized event?'''

->Customized events are processed at the same speed as all other events.

'''How are you different than Google Analytics?  Kontagent, KISS Metrics, Mix Panel?'''

->There are several basic metrics that any analytics company can deliver--things like ARPU, DAU, k-factor, etc. If you think about it, these are all essentially variations on '''what has happened?''' It's not complex. Some of the more advanced approaches also bring in funnel analysis, segmentation, marketing support, or offer consulting services to data mine. All of these firms collect data via APIs and process them in the cloud. They do a good job, and we do these things too.

->What's different about us are two key abilities.

->#1, We calculate Social Value. This is unique, new, and powerful, telling developers not just the lifetime value of a user, but their impact--in dollars--on others.

->#2, We've built a system that allows for predictive modeling. That is, we tell you not just what happened, but what will, on a user-by-user basis. For example, this player is worth $10 and is likely to quit this week. Some other companies offer the ability to hire their experts to build these kinds of advanced data models on a one-off basis. Our system does this automatically, daily, at no extra cost.

'''What is Ninja Metrics? What makes you different/unique?'''

->Ninja Metrics is a group of scientists (http://vwobservatory.org/) who've been working on "big data" and gaming for the past 7 years. We're half social scientists and half computer scientists. Our group has built the field of big data game analysis from scratch, having now published over 80 journal articles on gamer psychology, algorithms and development. A unique expertise is our background in social network analysis. There is simply no other group like us. We've done work for the CIA, the National Science Foundation, and others. The result is a powerful set of techniques and algorithms that reveal more than has ever been possible about gamers, their communities, their value, and their future actions.

'''Do you share the data I send you?'''

->No. We do not share your data with anybody outside of Ninja Metrics. Internally, our data scientist may analyze your data in order to improve our algorithms.

'''What are you doing with the data I am sending you?'''

->The data you send is processed by or modeling software in order to derive your results, all of which are showcased in Katana.

'''How long do you keep my data?'''

->We keep your data indefinitely and it is available in Katana for you to compare different metrics through different periods of time.

'''Is my data secure?'''

->Yes. Ninja Metrics conducts continuously internal security audits at all levels. From the moment you make a call we safeguard your data.

'''Can I use Google Analytics and Katana at the same time?'''

->Yes. There is no conflict in using both. Some of the instrumentation will be similar, so you can leverage what you might have already done with GA in favor of Katana.

'''Why should I track all this data and information? By tracking the activity in your game you can gain a great deal of insight into what’s working, what’s not, where to improve and where to leave things alone. Furthermore, by tracking activity you can see what your users are doing, how they interact with the game, what they buy, who influences buying and how groups of users influence each other.

'''Which browsers do you support? "Supported Browsers:

• The latest version of Internet Explorer and versions 8, 9, 10.
• Firefox, latest version.
• Chrome, latest version.
• Safari, latest version."

'''How do I sign into Katana? "You can securely log into Katana here:

https://accounts.ninjametrics.com/user/login "

'''Link to the Ninja Metrics Privacy Statement. http://www.ninjametrics.com/privacy.html

'''Link to the Ninja Metrics Terms of Service. https://accounts.ninjametrics.com/terms-and-conditions

'''Please provide some details about Confidence Scores, Projections, and Machine Learning. "A model is a way of building up a pattern. For example, a model may say that ""men over 40 are more likely to visit the doctor,"" which would require knowing gender, age and doctor visits for a lot of men. A confidence number indicates how well the model is performing over time, i.e. how accurate that model really is. When you're looking into the future, the models will never be 100%, but they can become very strong.

Our projection models are known as ""machine learning models,"" which means that the software looks for patterns and gets better at recognizing and forecasting them over time. Here's a simple way of understanding them. Say that there are four events called A, B, C, and D. The software notices that they often happen in order: A, then B, then C, then D. So, when it sees A, B and C occur, it gives a likelihood that D is going to happen next. When you think that ""D"" could be Churn or Conversion or Virality or the spending driven socially by a person (Social Value), or any number of things, you can appreciate the power of these models.

They're also quite complex. It may be that D will come next if it's a Friday, or if C is an action taken by a woman and not a man, or any of a million other possibilities. For the system to reach these conclusions, it needs to see many actions. So, on day 1, the confidence scores are going to be quite low. Over time--sometimes a few days and sometimes a week or two--they will improve to a point where they'll level off because the patterns the system is seeing aren't changing much--though the individuals in the system may be. If there's some kind of change in the system, e.g. activity drops due to a seasonal pattern or some large event, the model will adapt, take a slight hit in confidence, then climb back to the peak.

This need for seeing actions leads many to either wait for the models to get strong, or to implement the analytics as early as possible (betas are great), or to load in historical data to get good results immediately.

What's nice about machine learning models is that they deal with all of the details and permutations and simply give you the likelihood. You can then think of this likelihood like a %, though it's actually slightly more powerful than that. In many cases, our confidence numbers are ""F-scores,"" which are more reliable than a %. An F-score rules out both false positives and false negatives. This makes it an extremely ""conservative"" number, and more trustworthy. It's also going to be lower than a %. For example, F-scores in the phone industry for user churn tend to hit around .35 or .4. Numbers over .6 are very strong."

'''Individual & Social Values Katana's super-power is a combination of predictive analytics and social network analysis that lets it tell you who your most important users are. Unlike other approaches, Katana splits every user's value into two distinct components:

#1: Their Lifetime Value (LTV), or how much they will spend before they leave someday
#2: Their Social Value, which is how much they drive other users to spend.

Put together, these two values tell you how much any one user contributes to your bottom line. If that user were to disappear tomorrow, this is how much future revenue would disappear.

The users who have a LTV are commonly known as ""whales."" We call the users with high Social Value ""Social Whales."" They are usually a minority, and usually not the same people as the regular whales.

Katana lets you take advantage of these users in three ways:

#1: Be smart about finding more of them.
#2: Monetize them when you have them.
#3: Focus on retaining them over others.

When you've identified your whales or Social Whales, you should consider doing something with them. Katana lets you sort and export custom lists of these users from the dashboard so that you can run custom promotions, messaging or experiences on them. Our testing system is under construction, and will enable you to sample these users, test interventions on them, then see the results on both them and their friends.

When we look at social networks we use the metaphor of a ""ripple on the pond."" If you start the intervention at the right person and do it the right way, you may see viral effects. Katana will quantify these for you."


'''How do I make sure the items I use in my application are displayed within Katana? You need to initialize all items used within your application that you want Katana to analyze and track.  Use Event #15 to initialize each item and then trigger Event #16 (Item Use) to record the use of each item.  You do not need to use Event #16 for items being purchased using Event #18 but the items being purchased must be initialized before the purchase.

'''An example of buying items using virtual or in-game currency. Make sure you have initialized each item using Event #15 and then use Event #18 to record the currency transaction.  Make sure to set the transaction type to one, which represents virtual currency.
Changed line 57 from:
 [[(Attach:)PvP.png]]
to:
 [[(Attach:)pvp54.png]]
Changed line 57 from:
 [[(Attach):PvP.png]]
to:
 [[(Attach:)PvP.png]]
Changed line 57 from:
 [[C(Attach):PvP.png]]
to:
 [[(Attach):PvP.png]]
Changed line 57 from:
 [[Attach:PvP.png]]
to:
 [[C(Attach):PvP.png]]
Changed line 57 from:
 [[Attach:PvP]]
to:
 [[Attach:PvP.png]]
Changed line 57 from:
 [[Attach:PvP.png]]
to:
 [[Attach:PvP]]
Changed line 57 from:
 Attach:PvP.png
to:
 [[Attach:PvP.png]]
Added lines 56-57:

 Attach:PvP.png
Changed lines 57-61 from:
Total number of times/account 1 has killed account 2.
Total number of times/account 3 faced death by the actions of account 4.
Total number of times/account 5 has assisted account 6.
Total number of times/account 7 has summoned minion 1.
Total number of times/account 9 has killed minion 2."
to:
->Total number of times/account 1 has killed account 2.
->Total number of times/account 3 faced death by the actions of account 4.
->Total number of times/account 5 has assisted account 6.
->Total number of times/account 7 has summoned minion 1.
->Total number of times/account 9 has killed minion 2."
Changed lines 11-61 from:
-> Katana's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease.
to:
->Katana's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease.

'''My application has a lot of items.  Do I need to send Katana all of the items in my application?'''

->Within your data there will be some very commonly used items. These make up the bulk of the item use count. The very common actions and items can be ignored if the volume of events is very large. You may feel there is no reason to send the data to Ninja Metrics because the data is so commonly used and you know the impacts to your application. However, there are probably items within your application that are not commonly used or might be new to your application. These uncommon and new items should be sent Ninja Metrics so you will be able to learn more about them. Ultimately it is your decision to send the data that you want more information about.

'''I see you have several Enter/Exit and Begin/End events. Do I need to send both events when sending you data?'''

->Yes, you do need to send both events. An example might be the Recruitment Send and Receive events (EventID #33 & #34). Katana can not calculate K-Factor within both #33 & #34 events. Another example is with Level Begin and Level End (EventID #27 & #28). Katana will not be able to calculate time at level unless both events are received.

'''Do you allow for the use of custom events to track data?'''

->Yes.  Event #54 is the Custom Slot event.  You are permitted to use as many Custom Slot events as you require.  There is no limit. 

'''How can you help us with the integration of your service?'''

->We have a full-time integrations specialist who will help you with the entire process.

'''Where do you store the data?'''

->Securely in the cloud, AWS.

'''Where do I generate the AppPassword / Access Token string?'''

->The Access Token can be generated from the Add Game/App functionality. 

'''How do I know the data I am sending you is valid and accepted by Ninja Metrics?'''

->With every account we provide a link to a Logs page. The page shows the last 12000 events you sent. Whatever events you see there are being logged and they are valid.

'''What about events that may not be valid?'''

->Within the Logs page you can switch to a view showing the last 12000 events received by Ninja Metrics with errors.  These errors are specific and will help you fix any data quality issues happening during the sending of the events.

'''How do I set up and use my development data and production data at the same time?'''

->Simply add another game/application and treat one as your production game and the other as your development environment game.

'''An example of a currency transaction where a user is buying in game gold by spending real world money.'''

->The game features gold as one of its currencies and gold can either be earned by completing certain in-game actions or quests, or gold can be purchased by paying for it using real world currency. A player would decide how much gold they want to purchase and then complete the transaction.  The Currency Transaction Event (EventID #18) would be called and the required parameters would then be sent to Ninja Metrics.  Be sure to set the transaction type to zero in order to indicate the transaction is using real money.

'''An example of a PvP game using Custom Slots (EventID #54).'''

->The image below displays the following scenery:

Total number of times/account 1 has killed account 2.
Total number of times/account 3 faced death by the actions of account 4.
Total number of times/account 5 has assisted account 6.
Total number of times/account 7 has summoned minion 1.
Total number of times/account 9 has killed minion 2."
Changed lines 9-11 from:
'"What timeout parameter does Ninja Metrics use for my users?'"

-> Katana
's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease.
to:
'''What timeout parameter does Ninja Metrics use for my users?'''

-> Katana's
automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease. 
Changed lines 9-11 from:
"'What timeout parameter does Ninja Metrics use for my users?"'

-> Katana's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease.
to:
'"What timeout parameter does Ninja Metrics use for my users?'"

-> Katana's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease. 
Changed lines 7-11 from:
->Yes - Required events for any data to display within Katana include Event #1 (Login) and Event #9 (Account Creation).
to:
->Yes - Required events for any data to display within Katana include Event #1 (Login) and Event #9 (Account Creation).  Event #1 and Event #9 must then have matching account IDs.

"'What timeout parameter does Ninja Metrics use for my users?"'

-> Katana's automatic logout of users is set to 12 minutes.  The 12  minute auto logout will change based on the behavior of your users.  The auto logout can increase and decrease.
Changed lines 5-7 from:
'''Do my timestamps need to be formatted a specific way when sending data?'''

->Yes - Here is an example of a properly formatted timestamp, "2013-11-15T19:07:30.221Z".
to:
'''Are there any required events in order to integrate with Ninja Metrics?'''

->Yes - Required events for any data to display within Katana include Event #1 (Login) and Event #9 (Account Creation).
Changed lines 1-2 from:
'''''Do my timestamps need to be formatted a specific way when sending data?'''''
to:
'''Do my timestamps need to be formatted a specific way when sending data?'''
Changed line 5 from:
'''''Do my timestamps need to be formatted a specific way when sending data?'''''
to:
'''Do my timestamps need to be formatted a specific way when sending data?'''
Changed lines 3-7 from:
Yes - Here is an example of a properly formatted timestamp, "2013-11-15T19:07:30.221Z".
to:
->Yes - Here is an example of a properly formatted timestamp, "2013-11-15T19:07:30.221Z".

'''''Do my timestamps need to be formatted a specific way when sending data?'''''

->
Yes - Here is an example of a properly formatted timestamp, "2013-11-15T19:07:30.221Z".
Changed line 1 from:
""Do my timestamps need to be formatted a specific way when sending data?'''
to:
'''''Do my timestamps need to be formatted a specific way when sending data?'''''
Changed lines 1-3 from:
!!!!''' Login (Event ID#1)'''[[#Login]]

This event records the time at which the user logged into the system (website, game, software, etc.). To log a login event, you need to pass the user's account ID, character ID, platform, and timestamp
. Optionally, you may pass the shard ID, and login area name and ID.
to:
""Do my timestamps need to be formatted a specific way when sending data?'''

Yes - Here is an example of a properly formatted timestamp, "2013-11-15T19:07:30
.221Z".
Added lines 1-3:
!!!!''' Login (Event ID#1)'''[[#Login]]

This event records the time at which the user logged into the system (website, game, software, etc.). To log a login event, you need to pass the user's account ID, character ID, platform, and timestamp. Optionally, you may pass the shard ID, and login area name and ID.