New Customer Acquisition System, and Seasons


I haven’t posted anything for the past two weeks, for good reason: I have been travelling to Sydney, Perth and Singapore. It was some great recharging time but things have still been moving forward for Startup Freak. Here is a quick breakdown of what has been happening:

Performance and Security:

I already mentioned this in my last post that security and performance are no longer just tasks that you level up, but rather new feature types. Security features protect features/components within a radius, whereas performance features are directional and boost the speed of components in a given direction. I have also added “heat maps” which let you see where the security holes and performance bottlenecks are at a glance.


Most Wanted Features

You can now build “most wanted” or “most requested” features. Think of them as blueprints of a “good feature”. As you perform market research and your visibility of the product/market space increases you will be able to see their outline. Building them will give you special boosts to signups and user retention (more on that later). Ideally I want to balance the game such that it’s not always possible to build them (because of conflicting priorities), but doing so has rewards.

A last note on this: you can now drag the map and build out features as far as you want.


Customer Acquisition

Given the above two changes and also based on feedback from the pre-alpha, I really wanted to streamline the customer acquisition system. I just didn’t like the way it worked. It wasn’t very clea what action as the player would improve each segment of the funnel. Here are the changes (much testing is still needed):

Removed Trials

  • As much as I wanted to keep it around, there was just too much overlap with how signups work. It muddied the water. So the funnel now is visitors => signups => churn.

Removed Cohesion

  • Gone

Visitor Calculation

  • Continues to be tied to your marketing spending.
  • Activities now use an S curve rather than a linear model.
  • This means that there is an upper limit to monthly users an particular activity can bring in.
  • Encourages figuring out the best activities for marketing rather than just pumping money into all activities.

Signup Calculation

  • Signups only depend on number of completed features.
  • There is a value representing “feature breadth expectation” in the market. Over time this value increases, meaning if  you don’t keep building new features your signup rate will fall.

Churn Calculation

  • Depends on average size of features (depth). The market expectation for this increases over time.
  • Depends on average performance, The market expectation for this increases over time.
  • Depends on bug rate
  • Depends on design/tech ratio

Most Wanted Features

  • Building most wanted features will give you a direct boost to signups and reduce churn. I’m still working out the specifics.

There are still a lot of numbers here and maybe more simplification is needed. But I think at least the decision points are a lot more clear:

  • To boost signups, build a lot more features (which translates to smaller features)
  • To reduce churn, build larger features (which can adversely affect signups), add more performance features, fix bugs, and get your design/tech ratio right.

This new “expectation” number has an added benefit of containing growth over time. Exponential growth was a big problem in the pre-alpha version and something I’m looking for interesting / meaningful ways to deal with.


How About the Weather?

I have just been dealing with too many numbers in the game and it gets a little tiresome. So I decided to work on a simple season simulation which I have been intending for the game:

  • I have 4 tint colors for the sky and the background tiles (the city). I transition between these around the shoulder seasons. It works quite nicely actually though I need an artists help with picking the colors.
  • During winter snow particle systems randomly kick in (added some parallexing to this which looks nice as you scroll the level).
  • During spring rain particle systems randomly kick in.
  • Clouds/Haze roll through randomly in all months except summer.

The art definitely needs work but I’m quite happy with how it looks so far.



Scope Reduction

I have been thinking more and more about how I can wrap up the game. The first step is to get very clear about what big features I should cut. So far:

  • Logo design feature: I just can’t do this justice without spending a ton of time on it. Unfortunately that means I don’t currently have any meaningful tasks for designers besides features.
  • Competitor companies and AI: Again doing poor job of AI will just make the game really bad. As much as it would be a great addition, it just won’t fit in a reasonable timeline.


What’s Next?

Trello knows.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s