This is part two of the FeatureCat Diaries-series. FeatureCat is a feature- and feedback management SaaS I am building in public. This is a sort of Captain's log, where I am rambling about my progress.

April 2021

Day 15

10x features and target customers

While I’m working towards the closed beta of FeatureCat, including user registration and multi-tenancy, I am also thinking about new features, as one does. That is still a little ways down the road1, but I had some thoughts about the general direction of FeatureCat, inspired by an episode of the always excellent IndieHackers podcast where Courtland Allen interviewed Aleem Mawani of Streak.

Aleem’s company Streak is a CRM built right into Gmail. Aleem talked a lot about their initial growth spurt precisely because of that Gmail integration: customers loved the fact that they had a CRM which lived in the same place as their email. Even though Streak was missing a lot of what one might deem essential CRM-features, the fact that was built right into Gmail was so valuable to customers they used it anyway.

For sales people that live in their Gmail all-day-every-day, Streak wasn’t just a slightly better CRM, but it was ten times better than using plain Gmail without Streak, or hopping back-and-forth from another CRM. Streak had found their target customer (sales people using Gmail) and their customer-winning feature (a CRM right in Gmail).

I have to figure out who FeatureCat’s target customer and what its customer-winning feature is. I do have a hypothesis, the foundation FeatureCat is built on, which is me: I develop mobile apps and I know that getting feedback is difficult. I also know that organizing and keeping track of feedback and requirements is cumbersome, and developing and shipping the right features one of the biggest challenges.

That forms my current working hypothesis:

Target customer: Mobile app developers (it’s-a me)
10x Feature A: Receiving more and better feedback by making the feedback process effortless and delightful for customer’s users, with a feedback board right in the app.
10x Feature B: Keeping track of feedback (requirements/artifacts) and prioritizing the right features (build what your customers want) with an artifact graph.

The target customer base might shift over time, but for now mobile app developers are the one I know, can support, and empathize with (because it’s me). Mobile app developers are under-served by the current offerings in the market, since most available tools focus on web apps, and mobile app developers struggle with knowing anything about their users, since Apple/Google handle the distribution and a large part of the user interactions.

That feedback feature is my direct sell: get more and better feedback, both qualitative (i.e. suggestions/ideas) and quantitative (voting on proposed features). That is being validated as it’s being used in my own apps, but just because users use it does not mean that customers actually want it. Figuring that out will be the focus of the beta phases.

The feature tracking/prioritization/organization is an entirely different beast. I do a lot of product management in my day job (i.e. requirements engineering), and while the user stories for the dev team are the structured visible output, the “collect-and-organize inputs from various stakeholders and translate into well-defined requirements” is a messy amalgamation of emails, documents, meeting notes, PowerPoints and Excel sheets, in dozens of different places, tools and file systems.

The proposed solution in FeatureCat is an artifact graph, a place for structuring loose pieces of information (artifacts), which are then linked to features. It’s currently in an even rougher state than the feedback systems, but it’s the unique twist on the feedback part I can imagine being the better selling feature in the end — if it can become an essential part of the product management process.

Where does FeatureCat fit into the product management/software development process? The V-model can be a useful framework for thinking about the initial-idea-to-shipped-product-increment. It has gone somewhat out of style at the macro level (entire product) with all the agile hype in the recent years, but it is a great description of the development process at the micro level (for a feature or product increment).

V-model with FeatureCat positioning
The V-model for software development, with FeatureCat right at the top left. Adapted from Wikipedia.

In the V-model, the Y-axis represents the inverse detail level of requirements, and the X-axis is time. It starts at the top-left with a shallow concept, and goes down the left side of the V with increasingly detailed requirements, until you reach a detailed design, which then goes into implementation. On the right side of the V, you go up from implementation, continuously verifying and validating the product increment against the requirements.

FeatureCat is located on the top-left half of the V: initial concepts, requirements (i.e. user feedback) and roadmaps all live in there, while detailed implementation design moves to an issue tracker like Jira or Github. This is a hint which future integrations might be important. I’m not sure yet how to support the verification and validation phase, apart from high-level reconciliation of product increment with feature requirements, but I feel like at least the top-left half of the V-model is grossly underserved and usually solved by makeshift file-system dumps.

That is where I would see the artifact graph going: a better solution for requirements engineering in an agile world. An interesting challenge is that the target customer for an artifact graph in all its glory likely won’t be the indie iOS developer that I’ll target at first, but the ‘bigger dev shop/department’ product manager. But that’ll be a problem for another day.

Thanks to Courtland and Aleem for inspiring me to think what the customer winning (10x) features of FeatureCat could be. Validating these hypotheses will be one of the most important challenges for me in the next couple of months, as will be readjusting or changing them, and pivoting the entire product as a consequence.

That is it for this part. It was a more product than dev-focused, but development will return with full force in the next part. Coming up: closed beta.

  1. Memo to me: don’t. I know you love building new stuff, but FeatureCat’s most basic premise does not even work yet. Focus.