DEV.co
Resource · Technical Discovery Worksheet

Surface the hidden complexity before it surprises you.

Estimates blow up because of what nobody asked about up front. This worksheet covers the questions — users, data, integrations, constraints, risks — that reveal the real shape of a project.

Users · data · integrations · constraints · compliance · risks · unknowns

The expensive surprises are always discoverable.

The integration with the legacy system. The compliance requirement. The data that isn't as clean as everyone assumed. None of these are unknowable — they're just unasked.

Good discovery is structured curiosity: a deliberate sweep across the dimensions where complexity hides. This worksheet is the sweep we run before estimating any non-trivial build, so the estimate reflects reality instead of optimism.

The discovery worksheet.

Answer these honestly and most estimate-killers show up early.

  • Users + roles — who uses this, what permissions exist, how many, internal or external?
  • Core workflows — the actual day-to-day paths, including the messy edge cases people forget.
  • Data — what data, where does it live, how clean is it, who owns it, how much?
  • Integrations — which external systems, what APIs, are they documented, do you have access?
  • Constraints — performance, scale, uptime, browser/device, offline, and budget limits.
  • Compliance + security — GDPR, HIPAA, SOC 2, PCI? Auth requirements? Data residency?
  • Existing systems — what's already built, what stack, what's the migration story?
  • Risks + unknowns — what worries the team, and what's never been done before here?
Show, don't tell

Discovery turns vague asks into a real estimate.

The same feature can be a week or a quarter depending on the answers. Discovery is how you find out which.

discovery.yamlyaml
feature: "Let customers export their data"users: "external · ~12k accounts"data: "spread across 3 services, inconsistent schemas"   # ← complexityintegrations: [stripe, legacy-billing (undocumented)]    # ← riskcompliance: [gdpr]                                        # ← must-haveunknowns: "does legacy-billing expose an API at all?"     # ← de-risk first
What it reveals
“export data” is not a 1-week task here
the legacy API is the real risk
de-risk that before quoting the rest

A flagged unknown is cheap; the same unknown discovered mid-build is the thing that wrecks the timeline.

Discovery as a service

Let us run discovery before you commit.

A paid discovery engagement produces a clear technical plan, a realistic estimate, and a de-risking path for the unknowns — usable with us or any other team.

It's the cheapest insurance against a project that balloons after it starts.

Book a discovery

Common questions.

Why pay for discovery before building?
Because discovery is where you find the things that 10x an estimate. A small upfront investment prevents large, late surprises — and the output is yours to use with anyone.
How long does discovery take?
For most projects, a few days to two weeks depending on complexity and how many unknowns need de-risking.
What do we get out of it?
A technical plan, a realistic estimate with assumptions stated, and a prioritized list of risks with a plan to retire them.
Can I just use the worksheet myself?
Absolutely — the checklist above is the core. The full worksheet adds prompts and examples for each area.

Get the full technical discovery worksheet.

We'll send the complete worksheet — and can run a discovery engagement to produce a real plan and estimate.