Minimum viable test

Start with your core assumption. What do you believe will happen if you execute this idea? Write that assumption down in a single sentence. "If we create a webinar about X, 15% of registered attendees will become marketing-qualified leads." That's your hypothesis.

Design the minimum test that would validate or disprove that assumption. Don't build more than you need. If you're testing whether customers want a feature, ask them in a survey rather than building it. If you're testing a new messaging angle, test it with email to a small segment rather than launching a full campaign. If you're testing a new pricing model, offer it to 5 customers manually before building billing infrastructure.

Run the test with a clear decision rule. Before testing, decide what result constitutes success. If 15% of webinar registrants should become MQLs and you only achieve 8%, is that enough to move forward or should you change the approach? Decide this threshold before running the test so results don't bias interpretation.

A consulting firm was considering launching a new service line. Before investing in hiring and infrastructure, they created a landing page describing the service with a call-to-action to request more information. They drove traffic through organic search and paid ads. Within two weeks, they had 50 inquiries. This validated that market demand existed before they committed to building the service. The landing page test cost less than 5,000 pounds and provided clear evidence of demand.

A minimum viable test is the smallest, cheapest thing you can do to find out whether an idea actually works, before you pour real time and money into building it. Instead of building the whole feature, running the full campaign, or waiting for perfect conditions, you isolate the single assumption everything rests on and test just that. The question it answers is blunt: "will this work?" , and it answers it for a fraction of the cost of finding out the hard way.

It's not a pilot and it's not an MVP. A pilot is a full-scale run of a finished solution. An MVP is a real, working product with its core features. A minimum viable test is leaner than both , often just a survey question, a single landing page, an email to a small list, or a manual process you'd never want to keep. The whole point is speed: a test that costs a few hundred pounds and a week teaches you far more than three months of building something nobody wanted.

The trick is to test the assumption without building the thing:

• Will customers want this feature? Don't build it. Say you're collecting interest with a Google Sheets survey sent to 30 customers, asking what they'd pay , you get a signal in two days, not two months.
• Will the market buy this new service? Don't hire for it. Say you stand up a single landing page in Unbounce describing the offer with a "request more info" button, drive a little traffic, and count the inbound. Fifty enquiries in a fortnight is your answer.
• Will this new pricing tier land? Don't rebuild billing. Say you write a short pitch in Brevo and email the higher tier to five existing customers by hand. If three say yes, you've validated it before touching a line of code.

Why it matters

Minimum viable tests slash the cost of learning. Most teams discover what doesn't work through expensive failures , features customers ignore, channels that never convert, segments that can't sustain the business. A cheap test surfaces those truths early, while changing course still costs almost nothing.

They also kill the endless "will it work?" debate. Instead of trading opinions, you run a test and read the data. That shift , from opinion-led to evidence-led , consistently produces better decisions. And when resources are tight, it's the only sane way to prioritise: test ten ideas cheaply, find the two or three worth real investment, and put your money there.

How to apply it

Start with the assumption. In one sentence, write down what you believe will happen: "If we run a webinar on X, 15% of registrants become marketing-qualified leads." That's your hypothesis.

Then design the smallest test that proves or disproves it , and build no more than that. Testing demand? Ask, don't build. Testing a message? Email a small segment before launching the campaign. Testing a price? Offer it by hand to five customers before touching billing.

Finally, set the decision rule before you run it. Decide up front what counts as success , if you needed 15% and got 8%, is that a go or a rethink? Fixing the threshold beforehand stops you from reading the result however you'd already hoped it would go.

A software company wanted to add a new collaboration feature they believed customers needed. Rather than building it over two months, they surveyed 30 current customers asking about the feature concept and how much they'd pay for it. Only 7 customers showed strong interest. The company revised the concept based on feedback, re-surveyed, and found stronger interest. This iteration through testing took two weeks and cost nearly nothing compared to building a full feature that might have had low adoption.

A SaaS company wanted to test a new enterprise tier at a higher price point. Rather than overhauling their entire pricing, they manually offered the new tier to 5 existing customers, explaining the expanded capabilities. Three customers accepted the new pricing. This manual test validated the concept with minimal risk. Once confidence increased through additional manual tests, they built the tier into their product.

Minimum viable tests reduce the cost of learning in B2B growth. Most teams discover what doesn't work through expensive failures: building features customers don't want, investing in marketing channels that don't convert, pursuing customer segments that can't sustain the business. Minimum viable tests surface these truths early when correcting course is cheap.

Tests also build organisational learning discipline. Rather than debating whether an idea will work, you run a test. Rather than relying on opinions, you rely on data. This shifts decision-making from opinion-based to evidence-based, which consistently leads to better decisions.

For a team with limited resources, minimum viable testing is essential. You can't afford to build and launch every idea to full scale. Testing allows you to prioritise which ideas are actually worth building. You can test 10 ideas cheaply, identify the 2-3 most promising ones, and invest in those.