A designer cannot read your mind, and the cost of an unclear brief is a revision cycle &mdash; days lost and a worse final image, because each round drifts a little further from what you meant.

Write one brief per image. Three images, three briefs.

[Watch the walkthrough](https://kb.trackiq.com/kb-assets/kbvideo-brief-explainer-5182fbed.mp4)

*Prefer to watch? One hundred and twenty-three seconds on what belongs in a brief, why the aspect ratio follows your traffic split, and the feedback discipline that avoids extra rounds.*

## **What a brief contains**

**Your current main image.** The starting point, so the designer can see what they are replacing and what changed.

**The reference images**, with your notes on what specifically you are borrowing from each. This is the core of the document.

**Links to download the source files.** Every product photograph, logo and asset the designer will need, in full resolution. A brief that makes someone ask for files has already cost you a day.

**A written description of the image you want**, in as much detail as you can manage.

**The dimensions.**

## **Writing the description**

Be specific to the point of feeling excessive. Every ambiguity becomes a decision the designer makes for you.

**Describe what is in the frame and where.** Product position, angle, what surrounds it, what the background is.

**Say what the focus is.** What should a shopper notice first, in that fraction of a second? If everything is emphasised, nothing is.

**Describe the feeling.** Clean and clinical, warm and domestic, energetic. Designers work well from this and badly from adjectives like "premium", which mean nothing concrete.

**Call out anything non-negotiable** &mdash; the product must be unobstructed, a particular feature must be visible, certain text must appear.

**Paste the reference images inline**, next to the paragraphs they relate to, rather than collecting them at the end. Proximity removes interpretation.

**Use annotated screenshots.** Circle and label the exact element you mean on the reference. A marked-up image removes more ambiguity than a paragraph.

![Sessions and conversion, per product](/kb-assets/shot-prod-kpis-b5e45f.png)

## **Choosing dimensions**

Determined by where your traffic comes from, which you established in the previous phase.

| Traffic majority | Aspect | Pixels |
|---|---|---|
| **Desktop** | square | **3000 &times; 3000** |
| **Mobile** | portrait | **2500 &times; 3000** or **2000 &times; 3000** |

**This is not a small detail.** A taller image occupies more vertical space in a mobile results feed, which means more attention before the shopper scrolls. On a mobile-dominant listing, the aspect ratio alone can move click-through rate.

Check the split before deciding rather than assuming &mdash; see [**Understanding Business Reports**](https://kb.trackiq.com/hc/knowledge-base/articles/understanding-business-reports).

## **Working with the designer**

**Match the professional to the job.** Standard graphic design, a 3D render and a photoshoot are three different skills, and a designer strong at one is often weak at another. Decide which you need before commissioning.

**Expect two or three rounds.** The first batch is rarely right, and that is normal rather than a sign of a bad designer.

**Give feedback that is specific and comparative.** Put the delivered image next to the reference and describe the difference concretely: "the product is too small relative to the frame &mdash; match the proportion in reference 2", not "it doesn't pop".

**Batch your feedback.** Collect every note into one message per round rather than sending observations as they occur. Drip-fed feedback produces drift and irritation, and multiplies the number of rounds.

**Keep going until it is right.** You are going to run this image on every impression the product receives for months. A further revision round is cheap by comparison.

## **When you are done**

You should have **two or three finished images**, materially different from each other and from your current one, all built to the correct dimensions.

That is the input for testing &mdash; see [**Testing a New Main Image**](https://kb.trackiq.com/hc/knowledge-base/articles/testing-a-new-main-image).

## **Common mistakes**

**A one-paragraph brief**, which guarantees revision rounds.

**References without annotation.**

**Missing source files**, stalling the job before it starts.

**Designing square by default** when most of your traffic is mobile.

**Vague feedback** that the designer cannot act on.

**Accepting something close enough** because you are tired of the process. This image will be seen more than anything else you own.

## **Related**

- [**Finding Reference Images**](https://kb.trackiq.com/hc/knowledge-base/articles/finding-reference-images) &mdash; the inputs to this brief.
- [**Testing a New Main Image**](https://kb.trackiq.com/hc/knowledge-base/articles/testing-a-new-main-image) &mdash; what happens to the finished images.
- [**Why the Main Image Decides Your CTR**](https://kb.trackiq.com/hc/knowledge-base/articles/why-the-main-image-decides-your-ctr) &mdash; why this is worth the effort.
