Analytics is define as a systematic computation analysis of data or statistics. Analytics can be used for many things and are a great way to determine different things, such as what kind of content the player is enjoying, how easy or hard a level is, and seeing what content is more popular than others and why. They can be used to map your users and determine what is working, and what isn't.
Steel Seal was intended to use analytics to tracks statistics such as how many play through's per session, average score, what IAP purchases people are making the most and how long a players average play session. All of these bits of information could have been useful in improving the gameplay experience, but due to over scope and lack of time, we couldn't get our analytics set up in time for launch. This data could have helped refine the gameplay and make it a better experience for the player. If we did use analytics, this is the type of data we would have collected and how we would have used it.
If we were able to note how many play through's per session, and how long an average session was, we could use that to determine whether or not the difficulty of the game was okay. If a players play session would only last a minute or two, and they had multiple play through's of the game, we could gather from this that the game may be too difficult in the early stages and could be frustrating to the player. If a player is getting frustrated with the game easily because it's too hard, this could cause them to uninstall the game, which means less active users. This also relates to the average score metric. If a lot of players scores were really high, it could indicate that the games difficulty ramps up too slowly, or if their scores were all very similar it could indicate that the game is fairly easy, but then suddenly gets really difficult without any ramping. This again could be frustrating for the player, and could cause them to uninstall or not play the game.
The information we could get from IAP data would have to do with future updates. If a large group of players own a certain skin, for example the Santa suit, we could create more holiday event skins, such as Easter and Halloween. This would help us identify customers and take the guessing game out of creating new content. Another bit of IAP data that is useful would be what power-up's people are using. If nobody is buying a certain type of power-up, we know it's either over priced, or not effective enough to be worth a purchase at all. We can use this data to balance out the abilities so that each power-up is a fair price and the player is getting a worthy product for their currency.
Unfortunately due to over scoping and just being really busy with the final release of the game, we couldn't incorporate analytics in the final product, but if some time in the future, that could be implemented, it could be used to make future content more appealing to players, and help balance the game.