How Connota works

Every review meeting runs on instinct. Usually good instinct: one person feels the concept is off-brand, another senses the message isn't landing. What's missing is the why. Connota finds it, through a methodical, evidence-based examination of the work itself. Your hunch becomes something you can point at.

Connota is built on decades of brand science, the research your best strategists quote. That research is clear about what effective brand work requires: consistency, distinctiveness, cultural awareness, message fidelity. Connota examines your work against exactly those conditions and shows you the evidence. It won't pretend to read minds — it will tell you, before anyone sees the work, whether it meets the conditions that research links to brands that win.

1. Teach it your brand once

The Brand Brain is the reference layer every analysis is measured against. It holds your identity: voice, positioning, visual system, colours with their roles, typography, market definitions. And it holds a registry of what makes you recognisable: logo marks, your registered tagline, exact colour codes and your actual value propositions.

You can feed it knowledge too: free-form notes, PDFs, URLs and documents imported straight from Google Drive. Connota reads them and remembers what matters. A calibration meter shows how complete your Brand Brain is, so you know exactly what to add for sharper results.

Why this matters: the quality of feedback depends on the quality of the reference. Because Connota knows your registered tagline and your exact brand colours, it can tell the difference between on-brand and almost-on-brand. That's the drift human reviewers miss, because it looks about right.

2. Upload the work, attach the briefing

Upload a single asset, a whole campaign set or a multi-frame sequence (Stories, carousels) and attach the briefing it has to deliver on. Working in Figma? Connect it and import design files directly.

Then choose the question you're asking. Campaign mode asks: will this deliver for this audience? Brand-concept mode (for merchandise, internal concepts and identity explorations) asks: is this us? It deliberately won't nag you about a missing call-to-action on a hoodie.

3. A multi-pass examination

Your work is examined the way a seasoned brand strategist would examine it, in a fixed series of checks, the same rigorous routine every time:

• Briefing fit

  Does the work deliver what the briefing asked for?

• Brand fit

  Is this recognisably, correctly you? Voice, visual system, message fidelity.

• Channel fit

  Does the execution respect the norms of the channel it will run on? Judged against a channel knowledge layer you can extend with your own rules.

• Cultural fit

  How the work reads across the markets you operate in, grounded in documented cultural research, not stereotypes.

• Set coherence

  Do the assets work as one campaign, reinforcing each other instead of competing?

• Distinctive-asset coherence

  Are the assets that make you recognisable present, prominent and consistently executed across every piece?

4. Evidence, down to the pixel

Every finding follows the same anatomy: the evidence (a direct quote or observation from the asset), the finding itself, why it matters, and a concrete action. The AI writes the findings; it doesn't get to pick the score. Scores are calculated from the findings by fixed rules, so a number always traces back to the evidence that produced it.

Colour claims are backed by an actual measurement instrument: Connota measures the pixels. It can tell you your brand teal is measured absent on a cap's front panel, with the numbers to prove it, and recommend a Pantone check at production stage.

And findings aren't buried in a text report. The analysis view pins them on the asset itself: markers on the exact locations, a composition lens, a ruler for alignment checks, an attention lens. Sequences get position-aware, frame-by-frame treatment.

5. Rehearse with a Virtual Focus Group

From the target-group description in your briefing, Connota casts a small panel of detailed, realistic personas: inspectable, editable and reusable. Run your assets past the group and read individual reactions plus a synthesis.

To be clear about what this is: a rehearsal room, not market research. It doesn't count votes and never claims statistical validity. It shows you how your message can be read by the audience you described: the misreadings, the resonances, the questions. You walk into the real audience prepared.

6. Take it to the people who decide

One analysis serves everyone involved: a management summary for leadership, the full evidence trail for the people doing the fixing, polished PDF exports, and revocable public share links that need no account to view.

Send the management summary straight to a Slack channel, or push finding actions into whatever tool your team runs on through the outbound webhook.

Who feels the difference

That's who Connota makes a real difference for.