Product themes, laid out.
Also called: Product variants · Concept directions · Product archetypes · Candidate types
Laying out the candidate product themes side by side, so the MVP you take forward is chosen on purpose, not defaulted into.
Most products have two or three plausible shapes: stripped-back, feature-rich, premium. Name them all on one page with a one-line verdict each, then pick deliberately. The point is not generating more options. It is choosing the one you would otherwise have backed into.
What product themes and types are
A product theme is one coherent answer to the question “what shape should this product take?”. The same validated idea can almost always be built three or four genuinely different ways, and each one is internally consistent. A heated proofing box could be a stripped-back single heated box, a connected device with an app and zones, or a large-capacity professional unit. None of these is wrong. They are different bets, aimed at different buyers, with different bills of materials and different risks.
The trap is that teams rarely lay the themes out. They drift into one, usually the one the most enthusiastic person in the room described first, and then spend the next six months optimising a shape nobody actually chose. Naming the themes explicitly turns an accident into a decision. You do not need many of them. Two or three real candidates beat a brainstorm of fifteen variants that mostly differ by the colour of the lid.
I should add the obvious caution. The goal here is not to manufacture options for the sake of looking thorough. A wall covered in sticky notes feels productive and decides nothing. Three honest candidates, each one you would genuinely be willing to build, is worth more than a generated long list you will quietly ignore by Friday.
Here is what that page looked like for the proofing box we ran through the framework, so you can see the shape of a good answer rather than a generic template.
Three themes, one page, a one-line verdict each. The chosen theme was not the most exciting one on the table. It was the one that tested the riskiest assumption for the least money, which is usually how these decisions ought to go.
Default versus choose
- Build whatever was described first in the room.
- Never name the alternatives, so never weigh them.
- Discover the rejected shapes later, as regrets.
- Let the price land wherever the parts happen to total.
- Lay out two or three real candidates side by side.
- Give each a one-line verdict and a rough cost.
- Pick on purpose, and record why the others lost.
- Set the £149 price to suit the assumption you are testing.
The difference is not the work; both take an afternoon. The difference is that one of them leaves you able to answer “why this shape and not that one?” without inventing a reason afterwards.
How it fits the bigger picture
Product themes is activity 04.10.06 in the framework, the sixth of eight Evaluate-stage activities. It draws on the earlier Evaluate work, the competitor and market reads, and it feeds straight into Define MVP (04.10.07), where the chosen theme gets cut down to the smallest version worth shipping.
What it can do
It makes the shape of the product an explicit decision. By naming the candidate themes and giving each a verdict, it stops the team optimising a direction nobody actually chose, and it leaves a record of why the rejected themes were rejected, which is worth more than it sounds when someone reopens the argument in month four.
What it can’t do
It can’t tell you which theme is right; it only forces you to choose between the ones you can see. A theme nobody named cannot be chosen. And it does not size the scope. That is what Define MVP does next, taking the chosen theme and stripping it to the smallest viable thing.
See the full 10-stage process →
Try it yourself
Take your validated idea and write down every genuinely different shape it could take. Aim for two or three real ones, not fifteen near-duplicates. Give each a name, a rough bill of materials, and a one-line verdict. Then pick one on purpose and note, in a sentence, why the others lost. If you cannot tell them apart, you have variants, not themes, and the exercise is done.
Want a structured first pass? Start the Free Sprint → and the GPT will help you frame the candidate directions.
Your themes checklist
Project notes: three boxes, one chosen
▸ From the notebook · optional reading
Dan and Anna Hartley in Stockport arrived already in love with the connected version. Half an hour of laying the themes out side by side talked them out of it, gently.
3 min read · click to open
Dan walked in with a clear favourite: the connected box. App, temperature zones, push alerts, the lot. It was, he said, “the obvious version”. I have learned to be wary of the obvious version, because it is usually the one nobody compared against anything.
What ended up on the page
Simple Essential. One heated box, no app, holds 26°C overnight, £149. A bill of materials of roughly £38 to £55, a clean single promise, and the fastest route to finding out whether anyone would actually buy the thing.
Connected. The app version Dan loved. Around £249, a fatter bill of materials, firmware to write and then support forever, and a positioning that flatly contradicted the no-app story we wanted to put in front of the Sourdough School audience.
Pro / large capacity. A multi-loaf unit for keen amateurs and the odd small bakery. A real market, genuinely. Also a second set of ceramic tooling the Hartleys could not pay for in year one.
How the choice landed
Once the three sat next to each other, the connected box rather lost its shine. I asked one question: “which of these tells us fastest whether bakers will pay for a box that just holds the temperature?” Said out loud, it answered itself. The Connected theme was not testing that. It was testing whether bakers wanted an app, which was a different and more expensive question we had no reason to ask first.
We chose Simple Essential. The Pro unit went into a dated V2 backlog, and the Connected idea went into the same backlog with a polite note that it might never come back. Dan was a little deflated for about a day, then relieved, because the chosen box was the one he could actually afford to build before Christmas. (The large-capacity unit did launch in year two, and sold well, precisely because the simple box had proved the demand first.)
— Evaluate stage, project notes, 2026
— Next in Evaluate → Define MVP