Positioning and tone of voice tested with real buyers
Most positioning projects end in a workshop and a slide. We run three positions against each other on your buyers’ channels. Then we build a brand file around the position buyers chose.

When companies need a positioning test
Your company is entering a new verticaland the message for current customers may not work there.
Your company merged two brands after an acquisitionand the combined company needs one story.
Your sales and marketing teams describe the company differentlyand both versions have defenders.
Your category has become crowdedand every competitor claims the same three benefits.
Three positions tested on four channels
Each position leads with a different reason to buy. We choose the channels that can measure each one fairly.
Example test design for three positions
We keep only the cells that suit the question.| Search ads Do buyers already look for it? | LinkedIn ads Does the target title respond? | Outbound Do named accounts reply? | Landing page Do visitors convert? | |
|---|---|---|---|---|
| A. The cost you cut | Click-through and cost per clickNeeds search volume | Click-through by seniority | Reply and positive reply rate | Conversion rate |
| B. The risk you remove | Click-through and cost per click | Click-through by seniority | Reply and positive reply rateOften the strongest arm for risk messages | Conversion rate |
| C. The status you gain | Rarely searchedWe skip search for this arm | Click-through by seniority | Reply rate | Conversion rate |
What your team receives
Three positions
Each has a one-line promise, the proof behind it and the buyer it addresses.
The market test
Capped spend per position on suitable channels with the same dates for all three.
The readout
The winning position per channel with sample sizes and what we learned about your buyers.
Tone of voice
Vocabulary, sentence rhythm and the phrases your company avoids, with examples from your own material.
The brand file
Your position and voice written as instructions for your team and for Claude, ChatGPT and Lovable.
The checker
A list of banned phrases and machine-writing signs your editors check every draft against.
Three example test designs
How a brief for this service usually looks. We write yours with you.
The lead promise for an IT provider entering healthcare
- Arms
- Three positions
- Channel
- Search, LinkedIn and outbound
- Measure
- Click-through and reply rate
Two names for a company formed by an acquisition
- Arms
- Two names
- Channel
- LinkedIn and Google Search
- Measure
- Click-through, plus a trademark screen
Platform or service as the category word
- Arms
- Two arms
- Channel
- Google Search
- Measure
- Click-through and cost per click
These are example designs, and their numbers describe the setup. Real results publish on the experiment log.
Engagement terms
Questions about positioning
What if leadership prefers a position that lost?
We show the numbers and the sample sizes. Your leadership team still makes the call with evidence in hand.
Can we test a position nobody searches for?
Yes. We test it on LinkedIn, outbound or Meta. Search ads only measure existing demand, so we drop search for positions that create a new category.
Does the sprint include naming?
It can. We test names the same way. Each name gets a trademark knockout search and a domain check. That screen does not replace legal clearance.
Test your position before a rebrand or launch
A rebrand or a new vertical is a large bet. A positioning test costs a fraction of it.


