<services>the 3-week validation sprint</services>
Have an idea? Test it before you build it.
We turn it into working prototypes, put them in front of your actual users, and hand you a clear build, change, or stop call. One fixed price, three weeks, no guesswork.
Some of the teams we’ve worked with
The 3-Week Validation Sprint
Find out if your idea is worth building, before you invest the money, time, and focus to build it.
Most products don’t fail because they were built badly. They fail because no one checked whether people actually wanted them.
- of failed startups cite poor product-market fit. They built something the market did not want.CB Insights, 2026
- 43%
- of features in the average software product are rarely or never used.Pendo, 2019
- 80%
- of U.S. adults are more concerned than excited about AI in daily life. Just 10% are more excited.Pew Research, 2025
- 50%
An AI can hand you a hundred plausible answers in seconds. Knowing which to trust is the craft, and running the tests that get answers from real people is what product designers like us do.
Two ways to answer the same question
Do people actually want this?
Both ways start the same, with market research and workshops together. Then they split.
The traditional way
Months
and counting
- Questionnaires and interviews
- Still gathering insights, months in
Ends with opinions.
Our AI-native way
3 wks
start to finish
- Real prototypes, wired with analytics
- Several concepts, side by side
- Ads driving real traffic
- Launched to real users
Evidence you can bet on, in three weeks.
How it works
Everything above, in three weeks. Not three months.
Think of it as the traditional validation playbook on steroids. Being fully AI-native lets us test assumptions and ideas far faster, so you go to market with a focused, evidence-backed direction instead of throwing ideas at the wall to see what sticks.
- 1
Week 1 · Days 1 to 5
Frame and build
Agree on the real problem, then build what we test.
- Workshops to lock in your brand and positioning
- Up to three realistic three-to-five page concepts, with on-brand AI visuals
- Three concepts, in a fixed three-day build window
Ends with · Real, testable prototypes
- 2
Week 2 · Days 6 to 10
Test and refine
Put it in front of real people, then fix what they trip on.
- Moderated interviews with real prospects
- Five-second, headline, and price tests
- A refinement pass on whatever the sessions surface
- Qualitative first, no paid traffic yet
Ends with · Prototypes ready to earn real traffic
- 3
Week 3 · Days 11 to 15
Traffic and decide
Send real strangers to the pages, then give you a straight answer.
- A thousand dollars of ad spend, aimed at the channel your audience is actually on, from Google, Meta, Reddit, TikTok, or LinkedIn
- A price test, the same page shown at two prices
- Ad spend released only after the refinement pass
Ends with · Evidence and a clear recommendation
How we run it
Inside a remote workshop
It opens with live sessions online, on your calendar. We align on your brand, your customer, and the real problem first. From there we design a few light concepts in Figma, then move straight into coded versions with Claude Code, so you test real working sites, not static mockups.
Rushing to AI before that alignment is how teams build the wrong thing quickly.
What the workshops fill in
- Brand strategy
We fill in the fundamentals together
Across the first two days we work through the brand exercises live, from purpose and positioning to voice, personality, and tagline. Even when you feel sure of the answers, doing it as a team surfaces the gaps.
- 9 brand exercises, start to finish
- 2 days of live working sessions
- 2 to 4 stakeholders, a few hours each
- AI-agentic research
Then our AI does the digging
While we align as a team, agentic research pulls the market together fast, so the concepts we build are grounded in what is really out there, not in guesses.
- A SWOT analysis for your idea and its market
- A visual competitor teardown, with homepages, pricing, and onboarding captured side by side
- How rivals price, package, and position, pulled into one view
- The category landscape, and the gaps worth aiming at
- Audience, demand, and search signals worth testing
- Objection and messaging patterns pulled from real reviews
Where we fit
We come first, then hand your brand team real evidence.
A brand strategy agency shapes how you look and sound. We prove what people actually want first, so that work is built on evidence instead of assumptions.
QuestionA brand strategy agencyVisualBoston
Where we focusA brand strategy agency · Identity, messaging, and the long-term brandVisualBoston · Validating the product idea with real users
When we come inA brand strategy agency · Usually once there is a product to brandVisualBoston · First, before you commit to building
How we workA brand strategy agency · Positioning, naming, and visual identityVisualBoston · Real prototypes, moderated and unmoderated testing, live traffic
What you walk away withA brand strategy agency · A brand platform and guidelinesVisualBoston · Evidence: what people want, will pay for, and where it breaks
We recommend holding off on a product name or a full visual identity until you know people want the thing, because it is easy to fall for a logo on a product the market never asked for. Once the evidence is in, we can take on the visual identity ourselves, or point you to a brand strategy partner we trust.
Is this you?
You don’t have to guess your way through moments like these.
- You have a new idea and want evidence before building it
- You’re deciding how to price or package an offer
- You’re about to launch and aren’t sure your messaging lands
- You’re preparing a redesign and want feedback early
- Your team disagrees on what the problem is
The questions we help you answer
A sprint is built around the questions that decide whether your idea works, whatever stage you’re at.
- New idea
Will people want this?
Put a concept, prototype, or landing page in front of your target audience before you build, and see whether it resonates.
- New idea
What might they be willing to pay?
Test subscription tiers, which features belong in each, and whether people even understand your pricing model, like AI credits, before you set it.
- Any stage
Does our message land?
Test headlines, positioning, and value props with your target buyers before you spend on a launch or campaign.
- Any stage
What’s holding people back?
Uncover the questions and objections people have about your product or service, so you can answer them upfront.
- Live product
Can people actually use it?
Watch real people move through your product on desktop or mobile, and find exactly where they get stuck.
- Live product
Is the new version better?
Test a redesign side by side with your current version to measure whether it actually improves the experience.
The Sprint is new; the research behind it is not
VisualBoston gave us a solid design strategy backed by a wealth of research into our market and what would help us stand out. Truly experts in their field.
How we test
Every sprint is remote. We run moderated and unmoderated sessions, and recommend the right mix for your question and your budget.
Moderated
Live, one-on-one video sessions guided by a researcher.
- The why behind people’s reactions, not just what they clicked
- Follow-up questions asked in the moment
- Recordings you can rewatch and share with your team
Best for early concepts, complex products, and understanding motivations.
Unmoderated
Self-serve tasks people complete on their own, with screen and voice recorded.
- Patterns across more people, faster
- Clean side-by-side comparisons of options
- Quick, honest reads on pricing and messaging
Best for pricing and messaging tests, comparing designs, and quick answers.
What you get
You leave with a decision, and the evidence behind it.
Not a deck of opinions. A clear recommendation, the research that backs it, and everything we built along the way.
- A discovery report with a clear build, change, or stop recommendation
- The evidence behind it: what people want, will pay for, and where it breaks
- The actual coded prototypes we built, real working sites wired with analytics
- Session recordings and highlight clips to share with your team
- Your assumptions rescored: confirmed, changed, or dropped
- A recommended audience and offer to focus on next
Every sprint ends one of three ways
Even a “stop” is a win. Knowing early is the whole point.
- 01
Build
The evidence backs it. Move forward with a validated direction and the prototypes to build from.
- 02
Change
People want something adjacent. Pivot the audience, offer, or positioning before you sink months in.
- 03
Stop
The demand is not there. Save the months and budget, and put them toward a better idea.
Pricing
One fixed price, everything included.
3-Week Validation Sprint
$19,500fixed, all in
Remote, for websites, web apps, and mobile apps.
- Strategy session and brand exercises
- Up to three realistic three-to-five page site concepts
- Up to 12 moderated and 30 unmoderated participants
- We recruit, schedule, and pay every incentive
- $1,000 of ad spend to real traffic
- A report with a build, change, or stop call
Half to get started, half at hand-off. Payment methods are in the questions below.
Common questions
How many participants do we need?
The Sprint includes up to 12 moderated sessions and 30 unmoderated participants in one audience, which covers most single-segment tests well. If you are testing several segments, or want a larger read on pricing and messaging, we will recommend and scope the right number up front.
What happens after I book a call?
The first step is a short call, not a contract. We make sure the Sprint is a fit, confirm we can actually reach and recruit your audience before you commit, and agree on the one question worth answering first. If it is a fit, we scope a start date and send everything over. Most sprints kick off within a week or two of that call.
How fast can we get results?
Most sprints go from kickoff to a final recommendation in about three weeks, roughly 15 working days, start to finish.
Why is it three weeks, and not faster?
The order is what makes the evidence trustworthy. Week 1 we frame the problem and build up to three real concepts. Week 2 we test them with real people, then take a couple of days to refine what those sessions surface. Only in Week 3 do we spend the $1,000 of ad spend to drive real strangers to the pages. Buying traffic to a prototype before you have fixed the obvious problems just means paying to measure your own bugs, so we always test and refine first, then send paid traffic to pages we already know hold up.
What does a sprint cost, and what is included?
A validation sprint is a fixed $19,500 for three weeks of remote testing, agreed before we start. It covers a strategy session and brand exercises, building up to three realistic three-to-five page site concepts, up to 12 moderated sessions and 30 unmoderated participants in one audience, a refinement pass before any paid traffic, and a discovery report with a clear build, change, or stop recommendation. Recruiting, all participant incentives, and $1,000 of ad spend are included, which is roughly $2,500 of testing costs you would otherwise source and manage yourself, so there is no hourly billing and no surprises. Extra audiences or products, more participants, additional ad spend, and in-person sessions, which usually run two to four weeks around participant schedules, are scoped and quoted up front.
How does payment work?
We split it into two: 50% to get started and 50% at hand-off, when we deliver the report and the evidence. You can pay by card, bank transfer, PayPal, or Venmo. If you would rather spread the cost, you can pay over time with Affirm in monthly installments while we get started right away. Financing through Affirm is subject to eligibility, and checking your options is a soft credit check that does not affect your credit.
What kinds of products do you test?
We focus on digital products, so websites, web apps, and mobile apps. If you are validating an idea in one of those, we can build something real enough to put in front of people and test within three weeks.
What should we share with you before we start?
Anything that gives us a head start on your thinking. Investor or pitch decks, any market research you have already done, notes on your customers, and the competitors you are watching are all helpful, and we are happy to review them with you. If some of that came from an AI like ChatGPT or Claude, share the original prompt too, so we can check that what it produced is valid and accurate enough to build on rather than taking it at face value.
Do you handle recruiting the right people?
Yes, we take that off your plate. We recruit, schedule, and pay incentives for participants that match your target audience, and for unmoderated tests we run ads to bring real strangers to your pages. We buy that traffic on whichever proven channel fits your audience, which the Week 1 exercises help us pin down, from Google, Meta, Reddit, and TikTok, plus LinkedIn when your audience is B2B. We stay off brand-new ad platforms like ChatGPT ads, which are too unproven to trust for a real signal. That way the response reflects the market rather than just people you already know.
Can we test an idea that isn’t built yet?
Yes, it’s our bread and butter. Because our team is fully AI-native, we can spin up realistic three-to-five page sites, complete with on-brand AI illustrations, imagery, and video, so people react to something that feels real. These are built to get honest early reactions, not to be your final go-to-market assets, and that speed is how the whole sprint fits into three weeks.
Do you run in-person sessions?
Every sprint is remote by default, which is what keeps it fast and affordable. When a question can only be answered in the real world, we run in-person sessions too. Those usually take two to four weeks around participants’ schedules, so we scope them separately from the remote sprint.
Can testing prove product-market fit?
No single study proves it, but the right mix gets you strong early signals. We start with foundational research to reduce bias in who we recruit and in what we ask and why. From there, moderated sessions surface the motivations and objections behind a decision, while unmoderated tests scale that across more people to spot patterns in demand, pricing, and messaging, so you learn what’s real before you invest heavily.
Why we do this
I have been part of too many projects where we shipped a beautiful, well-built product that turned out to be a vitamin, not a painkiller. AI has made it easy to get instant validation from an LLM and build something huge without ever stepping back to ask the questions that matter. Who is this really for, do they actually need it, and why now? Testing is how you answer that with real people instead of a guess, and being AI-native lets us do it in far less time and cost, so you move forward knowing you are building the right thing.
Alec Harrison, CEO and Creative Director, VisualBoston
Have an idea? Let’s find out if it’s worth building.
In three weeks, you’ll know what people want, what they’ll pay for, and where the idea falls apart, before you spend months on it.
$19,500 · 3-week Validation Sprint · Everything included, remote, one fixed price
Schedule a call