01 · Discovery Sprint
You have an idea and domain expertise — but no clear answer to: should this be built, what would it cost, who should build it. A Discovery Sprint is structured research that ends with a complete product brief, a realistic budget and timeline, and a shortlist of the right teams to deliver.
↓ Expand a section to read more
Why you can trust me
Acting CEO of TreximAI · 8 years BA across AI products, logistics, fintech, B2B SaaS
Cases on this site are illustrative patterns drawn from 8 years inside IT — not invented for the page. No published clients yet.
You have deep knowledge of a specific industry — logistics, fintech, healthcare admin, education. You see a problem others don't, because you live with it daily. The product idea has been turning in your head for a year or two. But — how do you get from "idea" to a plan you can actually build against? Discovery brings structure: where's the hypothesis, where's the fact, what's worth building, what isn't.
You have an offline business that needs digitalizing, or an idea for a product on top of your existing one. The question is — what to build in v1, what's a realistic scope, what it'll cost, who should build it. Discovery gives you those answers as a document you can take to vendors — or even to investors.
A corporation, agency, or startup that wants to launch a new line — either a B2B service built on existing expertise, or an internal tool that automates a routine. Discovery gives you objective research on the idea and a concrete product brief the team can start building from.
Documents your team (or the vendor you hire) can start working from immediately — no "what else needs figuring out?" overhead.
What exactly we're building in v1: key features, user flows, what's NOT in v1 (explicitly stated — often more important). The brief works the same with vendors, investors, and the team — everyone reads it the same way.
We test critical assumptions: does the market need this product in this shape, are customers ready to pay, can it realistically be built in sensible timelines. Confirmed — moves into the plan. Rejected — out, with reasoning.
Not "plus or minus a few tens of thousands." Concrete ranges with reasoning, where the key uncertainty points are, how they affect price. Show it to a CFO or board — no surprise questions.
3-5 teams or contractors that genuinely fit your project — by tech stack, domain experience, and price range. With a brief summary of each one's strengths and weaknesses, plus specific questions to ask on the discovery call.
Arguments, numbers, technical details — everything you'll need during talks with the chosen vendor. So you walk into the conversation with data, not feelings.
The end value isn't the documents themselves — it's knowing exactly what to do next:
Deep interviews — with you, with potential clients (4-6 conversations), with industry reps. I gather the real picture: what hurts, what's been tried, what the market actually lacks. End-of-phase artifact: a "Context & Hypotheses" doc with the key assumptions to test.
I test the critical hypotheses: are clients ready to pay, is it technically feasible, is there market space. In parallel — I form the product brief: v1 features, user flows, technical assumptions. End-of-phase: complete product brief + validated hypotheses.
I pick 3-5 vendors that genuinely fit. Ask them for a short estimate based on the brief. Assemble a negotiation pack — arguments, numbers, red flags. End-of-phase: a ready-to-negotiate package.
A spec describes what needs to be done. Discovery gives you not only "what" but also "why" and "how confident." The output isn't a dry spec — it's a package you use to decide whether to commit capital at all.
If the MVP already works and has first users — probably not. Discovery is for the pre-code stage. If the MVP exists but the result isn't satisfying — we should talk about Audit, not Discovery.
4-6 deep conversations with potential or existing clients in your industry — that's the minimum needed to form objective hypotheses. If your industry is very narrow or clients are hard to reach — we discuss alternatives (secondary sources, expert conversations, analytics).