Experiment design assistant
Pre-registered decision logic with explicit assumptions and measurement limits.
- For
- Growth and product teams testing new features
- Solves
- Experiments begin without clear success and stopping criteria.
- Delivers
- Reviewed experiment design brief
- Built in
- about 3 weeks of creation time, MVP in 3 days
- Investment
- $6,000 for the MVP, $19,000 for the full product
- Run it
- Inside your business, or as part of your offer to clients
What it does
For growth and product teams testing new features, turn business hypothesis, baseline metrics and testing constraints into reviewed experiment design brief.
- Clarify proposed mechanism.
- Define primary measures.
- Specify guardrails.
- Document assignment approach.
- Flag measurement gaps.
- Draft analysis plans.
What goes in, what comes out
- Business hypothesis
- Baseline metrics
- Testing constraints
AI drafts, people review. Assumption-driven planning and decision workspace.
- Reviewed experiment design brief
How it works
The workflow
- InStart with
Business hypothesis, baseline metrics and testing constraints
- 1
Validate baseline inputs
- 2
Confirm definitions and constraints
- 3
Select editable assumptions
- 4
Calculate feasible alternatives
- 5
Inspect sensitivities
- 6
Let the responsible person approve a plan
- 7
Compare later actuals with the recorded assumptions
- OutFinish with
Reviewed experiment design brief
AI does the heavy lifting, people stay in charge
Extract input context and explain scenario differences. Use deterministic calculations or explicit optimization for quantities, compatibility, dates and prices. Show uncertain assumptions. Never let generated prose silently change the calculation rules.
What your team sees
Key screens: Hypothesis canvas, measure definitions, decision rules. Place editable drivers and constraints beside a clearly labeled scenario output. Include a baseline view, comparison chart or schedule, and an assumptions history. Let users trace a proposed quantity or date back to its inputs. Keep forecasts distinct from actual results. In this product, the first view is hypothesis canvas, followed by measure definitions and decision rules.
Accounts and administration
Scenario versions, baseline reconciliation, constraint checks, assumption ownership, reviewer approvals, plan exports and actual-versus-plan tracking.
Integrations and data access
Product feedback, authorized interviews, usage exports and requirement records. Read-only operational exports, calendars and finance or inventory records as relevant. Start with plan exports and retain human approval for execution. These are candidate integration categories, not verified supported connectors.
How we build it
We build with our own AI software development factory, so most implementations take days to a few weeks of creation time, not months. You see working software at every step, and exact timing depends on availability.
- 1
Scoping call
Day 1Thirty minutes on your process, your data and how you want to run it: for your own team, or for your clients. You get a fixed scope and price for the MVP.
- 2
MVP
3 daysOne buyer segment, one recurring use case; first modules: clarify proposed mechanism; define primary measures. Manual review in the loop. Built by our AI software factory.
- 3
Paid pilot
4 daysAccounts, roles, review states, audit trail and the first integration, hardened for two to three paying pilot customers.
- 4
Full product
6 daysRemaining modules: document assignment approach; flag measurement gaps; draft analysis plans. Self-serve onboarding, billing, monitoring and the wider integration set.
- 5
Run and improve
MonthlyWe host, monitor and improve it for a fixed monthly fee, or hand it over to your team. How the retainer works.
Why we start with an MVP
An MVP, or minimum viable product, is the smallest version that your users can actually work with. It is not a cheap version of the full solution. It is a test, built to answer the questions that decide whether the rest is worth building.
- Pick the riskiest assumption. Here: will growth and product teams testing new features use it to solve "experiments begin without clear success and stopping criteria"?
- Build only what tests it. One team, one use case, a few core modules. People do the rest by hand for now.
- Run a paid pilot. Reproduce a known historical plan, test missing inputs and boundary constraints, then run a new scenario.
- Measure, then decide. Track protocol completeness and interpretable results. Then expand, change course or stop, with evidence instead of opinions.
MVP scope for this solution. Begin with growth and product teams testing new features and one recurring use case. Build the first two modules: clarify proposed mechanism; define primary measures. Provide operator assistance for the third module: specify guardrails. Deliver reviewed experiment design brief through a manual review queue. Perform other necessary full-scope functions manually during the pilot. Include all applicable access, accuracy and professional-review controls from the start.
After the MVP. After paid pilots establish value, automate the remaining modules: document assignment approach; flag measurement gaps; draft analysis plans. Add one validated source integration, reusable customer configuration and recurring delivery. Expand to additional teams, document formats or languages only after testing the new scope.
What the build depends on. A defensible calculation model, explicit units, constraint validation and representative boundary tests. Advanced forecasting or optimization needs adequate historical data.
Investment
A planning range to start the conversation, not a quote. You pay per phase, so you can stop after the MVP.
- Phase 1
MVP
One buyer segment, one recurring use case; first modules: clarify proposed mechanism; define primary measures. Manual review in the loop.
- Phase 2
Paid pilot
Accounts, roles, review states, audit trail and the first integration, hardened for two to three paying pilot customers.
- Phase 3
Full product
Remaining modules: document assignment approach; flag measurement gaps; draft analysis plans. Self-serve onboarding, billing, monitoring and the wider integration set.
Indicative total, MVP to full product$19,000about 3 weeks of creation time · start with the MVP from $6,000
Running costs per month
A rough indication of monthly hosting and AI model costs once it is live, not tested. Real costs depend on usage, file sizes and the models chosen.
| Stage | Hosting and infrastructure | AI usage | Total per month |
|---|---|---|---|
| MVP and paid pilotabout 3 customers | $30–$60 | $50–$100 | $80–$160 |
| Full productabout 50 customers | $110–$210 | $350–$700 | $460–$910 |
Run it or resell it
For your own team
Growth and product teams testing new features run it inside the business: business hypothesis, baseline metrics and testing constraints in, reviewed experiment design brief out, reviewed by your people.
As part of your offer
Agencies, consultancies and software companies can offer it to their own clients under their brand. We build and maintain it; you sell and deliver it.
Your brand, or this one
Run it under your own brand, or start from this concept style.
- primary
#8d2791 - accent
#54c956 - surface
#f0e4f1 - ink
#22201e
- Headings
- Archivo
- Text
- Lora
- Voice
- Curious, rigorous, user-led
Selling it to your own clients: the go-to-market playbook
Pricing to test
Test USD 750-3,000 for a scoped planning setup and review, then USD 200-900 monthly for refreshes within agreed complexity. Data integration and optimization are separately scoped. All ranges are hypotheses.
Message to test
Experiment design assistant for growth and product teams testing new features. Pre-registered decision logic with explicit assumptions and measurement limits. Demonstrate the claim through a complete experiment specification for one hypothesis.
Where to find buyers
Product experimentation consultants
Lead magnet
A complete experiment specification for one hypothesis
The first 30 days
- Week 1: interview five prospective buyers in this segment: growth and product teams testing new features. Ask to see a recent example of the problem and their current process.
- Week 2: prepare this demonstration using authorized or synthetic material: a complete experiment specification for one hypothesis.
- Week 3: present it through product experimentation consultants and seek one narrowly scoped paid pilot.
- Week 4: review protocol completeness, interpretable results, total delivery effort and a concrete renewal decision before increasing scope.
Paid pilot
Reproduce a known historical plan, test missing inputs and boundary constraints, then run a new scenario. Compare feasibility, reconciliation and observed error rather than judging the quality of the explanation alone. For this solution, use business hypothesis, baseline metrics and testing constraints and evaluate reviewed experiment design brief. Agree success thresholds with the buyer before starting; collect a baseline for protocol completeness, interpretable results. A positive signal is payment and repeat use with acceptable quality and delivery cost, not a favorable demo reaction alone.
Success metrics
Protocol completeness, interpretable results
Retention and expansion
Refresh inputs, compare recorded assumptions with actual outcomes and refine validated constraints. Expand scenario complexity only when the buyer uses it for a decision.
Why clients would pick it
A validated domain model, customer-approved constraints and forecast or decision history that improves practical planning. For this solution, build around pre-registered decision logic with explicit assumptions and measurement limits. This advantage requires execution and accumulated customer trust; the base model alone is not a defensible asset.
Alternatives and positioning
Spreadsheets, planners, specialist forecasting tools and existing scheduling or configuration software. Differentiate on this specific proposed advantage: pre-registered decision logic with explicit assumptions and measurement limits. Test it against the buyer's current method on the same task. Competitor coverage and uniqueness have not been established.
Main delivery costs
Data preparation, domain modeling, validation, scenario computation, reviewer support and ongoing assumption maintenance.
Safeguards
Use consented research and preserve contradictory evidence. Separate observed user behavior, proposed explanations and untested product assumptions. Validate source access and reviewer availability during the pilot. Maintain customer-level access, data deletion controls and a record of final approvals.