Foundational Principles of Strike

Over the past 5 months, while building and iterating on ideas and implementations that add up to what Strike is, we’ve also been understanding what makes up a good product analytics SaaS through tons of user research and market analysis. ML and the pace at which advancements are being made in the field of AI add to the complexity. It’s a privilege to be a consumer of the rapidly progressing and competing AI advancements.

We are building Strike to be the product analytics SaaS for product builders, nailing down the UX and leveraging data from both quantitative and qualitative sources. Understanding features and tracking their individual metrics in silos has been done to death; we want to understand how your product is doing holistically and how we can act on problems and opportunities within.

1. Data Assimilation

   Building that holistic understanding requires combining data from different sources and iteratively building an understanding of the state of your product(s). Through integrations, our customers can integrate their data into our platform, both quantitative and qualitative. With tools like Notion and Figma making it ever so easy to write and collaborate and creating heaps of content, we have a lot of text and graph-like data scattered in tools within a company. Individually, these sources contain a lot of context, conversation and decisions. The issue has always been a lack of complete tool-belt to combine bits of information and relevant data points from these sources and to build a shared understanding that helps product builders get a clearer pictures of metrics and act on things.

2. Insights and Signals

   We want to go from data to generating insights to combining insights to build signals that further combine to tell us a story about your data. The story within your data can be:

   • Something problematic in your data and what it might be affecting and how to address it

   • An opportunity: As a product builder, with all the things going on, we sometimes miss opportunities that are industry standard or a common practice.

   These are just some examples of how Strike aims to help product builders stay ahead of their data through opportunities and problem detection.

   We are building ML tools and technologies to iteratively understand what the data is telling us, focusing on metrics and segments that are industry standard first (our user research and tons of industry benchmark data we have been collecting since March have been helping a lot here).

3. Data lineage

   Traceability and attribution of data are crucial for product builders to have a healthy discussion and build confidence in decision-making. This is even more significant when we consider several integrations already in place within Strike (Mixpanel, CSV, Notion, Slack) and much more to come soon.

4. Auto-Detection and Assistance Capabilities

   A more long-term goal for us is to build systems and models that go beyond comprehending insights and signals to detecting problems and opportunities with minimal supervision. This closely aligns with what’s commonly referred to as an agent in the AI world.