InterLAB tools · Hypotheses · funnel · decision

MVP validation toolkit

Collect data on how potential customers behave — landing page requests, conversion by channel, interview answers — and get a reasoned decision: continue, change course or stop. Thresholds are set before the experiment; conclusions follow published rules.

Free, no sign-up

Loading…

How it works

Hypotheses and thresholds — before the experiment
“We believe that… We will know we are right when…” + a metric, a success threshold, a failure threshold and a minimum sample. Once started, thresholds are locked; changing them requires an explanation that goes into the report.
Landing page and UTM links
A simple page with a request form — an HTML file you host yourself; requests go to your form endpoint or email. UTM links with spelling checks show which channel each person came from.
Funnel and interviews
Numbers by channel and stage: visits → clicks → requests → interviews → payments (stages are editable). Interviews: segment, pain 1–5, current workaround, willingness to pay, quotes, tags.
Report and decision
For each hypothesis — the observed value, a 95% interval and a verdict with a rule ID; the overall decision — continue, change course, stop or gather more data — with reasons and caveats.

Method and rules

The tables are generated from the same code as the calculation — rules v1.0 (13.10.2026). A share is x/n: for the funnel, n is the count at the “from” stage and x at the “to” stage; for interviews, n is everyone who answered the question.

The confidence interval is the 95% Wilson score interval (z = 1.96) without continuity correction. Unlike “±2 standard errors”, it stays within 0–100% and widens honestly on small samples: 0 out of 10 gives 0–27.8%, 5 out of 10 gives 23.7–76.3%.

ID	Verdict	When
MV-H00	Not enough data	The metric cannot be computed: not selected, a stage or channel was removed, the data are inconsistent, or the thresholds are invalid.
MV-H01	Not enough data	The sample (the share's denominator) is below the pre-registered minimum.
MV-H02	Supported	The lower bound of the 95% Wilson interval ≥ the success threshold.
MV-H03	Refuted	The upper bound of the 95% interval < the failure threshold.
MV-H04	Not enough data	The interval spans both thresholds (lower < failure, upper ≥ success): the data cannot yet tell success from failure.
MV-H05	Borderline	All other cases: the interval sits in the grey zone between thresholds or touches only one of them.

MV-H00

Verdict: Not enough data
When: The metric cannot be computed: not selected, a stage or channel was removed, the data are inconsistent, or the thresholds are invalid.

MV-H01

Verdict: Not enough data
When: The sample (the share's denominator) is below the pre-registered minimum.

MV-H02

Verdict: Supported
When: The lower bound of the 95% Wilson interval ≥ the success threshold.

MV-H03

Verdict: Refuted
When: The upper bound of the 95% interval < the failure threshold.

MV-H04

Verdict: Not enough data
When: The interval spans both thresholds (lower < failure, upper ≥ success): the data cannot yet tell success from failure.

MV-H05

Verdict: Borderline
When: All other cases: the interval sits in the grey zone between thresholds or touches only one of them.

Hypothesis verdict

ID	Decision	When
MV-D01	Stop	A problem hypothesis is refuted and no value or willingness-to-pay hypothesis is supported.
MV-D02	Change course (pivot)	At least one hypothesis is refuted (other than MV-D01): change what was refuted — segment, value, price or channel.
MV-D03	Continue	Nothing is refuted, and every problem, value and willingness-to-pay hypothesis is supported (at least one exists).
MV-D04	Gather more data	All other cases: key hypotheses are borderline or lack data.
MV-D05	caveat	Thresholds were changed after the experiment started — the report shows what and why; trust the conclusion less.
MV-D06	caveat	A hypothesis is past its deadline without enough data — decide whether to extend the experiment or change how you test it.

MV-D01

Decision: Stop
When: A problem hypothesis is refuted and no value or willingness-to-pay hypothesis is supported.

MV-D02

Decision: Change course (pivot)
When: At least one hypothesis is refuted (other than MV-D01): change what was refuted — segment, value, price or channel.

MV-D03

Decision: Continue
When: Nothing is refuted, and every problem, value and willingness-to-pay hypothesis is supported (at least one exists).

MV-D04

Decision: Gather more data
When: All other cases: key hypotheses are borderline or lack data.

MV-D05

Decision: caveat
When: Thresholds were changed after the experiment started — the report shows what and why; trust the conclusion less.

MV-D06

Decision: caveat
When: A hypothesis is past its deadline without enough data — decide whether to extend the experiment or change how you test it.

Overall decision (checked top to bottom)

F-01 “small sample”: a funnel step has fewer than 30 observations or fewer than 5 “successes” or “failures” — the interval is wide.
F-02: a stage exceeds the previous one — most likely an input error; that share is not computed.
F-03 channel comparison: if the best and worst channels' intervals overlap, the difference may be due to chance; if not, it is unlikely to be chance (a cautious check, not a formal test).
Landing page: a headline, button text, a contact field label and an endpoint (https:// only) or a valid email are required; 3–5 benefits work best; a privacy line is recommended.

ID	Level	Rule
U-01	error	The page address is a full URL with http:// or https://.
U-02	warning	The address uses http:// — prefer https://.
U-03	error	utm_source, utm_medium and utm_campaign are required.
U-04	warning	Uppercase letters: analytics tools are case-sensitive, “Instagram” and “instagram” are different sources.
U-05	warning	Spaces in values — replace them with _ or -.
U-06	hint	Characters other than a–z, 0–9, dot, _ and - get percent-encoded — reports become harder to read.
U-07	warning	The address already has utm tags — they will be replaced.
U-08	warning	The same value is written differently (case or separators) across rows.
U-09	hint	Values mix the _ and - separators.
U-10	warning	Two rows produce the same link — they cannot be told apart in reports.

U-01

Level: error
Rule: The page address is a full URL with http:// or https://.

U-02

Level: warning
Rule: The address uses http:// — prefer https://.

U-03

Level: error
Rule: utm_source, utm_medium and utm_campaign are required.

U-04

Level: warning
Rule: Uppercase letters: analytics tools are case-sensitive, “Instagram” and “instagram” are different sources.

U-05

Level: warning
Rule: Spaces in values — replace them with _ or -.

U-06

Level: hint
Rule: Characters other than a–z, 0–9, dot, _ and - get percent-encoded — reports become harder to read.

U-07

Level: warning
Rule: The address already has utm tags — they will be replaced.

U-08

Level: warning
Rule: The same value is written differently (case or separators) across rows.

U-09

Level: hint
Rule: Values mix the _ and - separators.

U-10

Level: warning
Rule: Two rows produce the same link — they cannot be told apart in reports.

UTM link checks

Limitations

The tool itself does not publish, host or send anything. The landing page is a file you host yourself; requests are collected by your form service or email.
Funnel numbers are entered by hand from your analytics: the tool does not connect to ad accounts or counters.
The intervals do not account for systematic errors: an unrepresentative sample, leading questions, bots and repeat visits.
Saying “I would pay” is a weak signal; a prepayment or pre-order is stronger. That is why the funnel can extend to payment.
The rule-based decision supports a discussion; it is not a verdict: the final call belongs to a person who knows the context.

Privacy

Everything is computed in your browser: computing never sends hypotheses, numbers, interviews or quotes to a server or puts them in the page address.
Data and versions are stored only in this browser's localStorage; “Clear all” deletes them. Exports are files: a Markdown or PDF report, JSON, landing page HTML, a CSV of links.
The landing page preview runs in a sandbox without scripts: the preview form does not submit anywhere.
Analytics receives only the fact of a report or export with the decision or format category — none of your data.
The result reaches us only if you send a request yourself and keep the “Attach the result” box ticked — the form shows exactly what will be sent: the decision, hypotheses with verdicts and intervals, funnel totals and interview aggregates. Quotes, notes and individual interview records, the landing page and UTM links are never included.

Related tools

All tools · 40

Proposal builderProposal templates for IT projects: scope, phases, estimate, timeline, assumptions and exclusions, versions with number control, approval and PDF.

Change request calculatorAn approved scope + new requirements → affected modules, extra cost, time and risk, a before/after comparison and a change order that leaves the original agreement intact.

Idea → Product blueprintAn interview about your idea → roles, features, three MVP scope options, priorities and risks. You get a Product Blueprint to start development from.

MVP complexity estimateProduct type, platforms, modules and integrations → a complexity band, workstreams and the questions you need answered before any honest price.

Scope and cost calculatorModules, effort, your rates, complexity factors, dependencies and a risk reserve → a cost and timeline range with every number explained, plus estimate comparison.

Technical brief generatorBuild a brief step by step: goal, roles, scenarios, integrations, data, acceptance criteria. You get Markdown to copy or download.

Tell us about your task — we'll review your request and get back to you

Demand confirmed — what next?

MVP development: a minimal product version for the people who already left a request. The cost is set after a brief and an agreed scope; the first conversation about the format is free.

The first short conversation is free — we choose the format
NDA — yours or ours
Pricing after a brief and an agreed scope

Message on WhatsApp