What to check if your output doesn’t look right
If your output does not look quite the way you expected, it usually does not mean anything has gone wrong. In most cases, it simply means one part of the workflow has not been set up, interpreted, or applied in the way you expected.
Brand Ready works across AI tools and Word, so the final result depends on which route you are using:
a brand skill inside ChatGPT or Claude
a branded Word template inside Word
or both together
First, check where you are working
The first question to ask is:
Am I expecting the branding to happen inside the AI tool, or inside Word?
This matters because the brand skill and the branded Word template do different jobs.
if you are working in ChatGPT or Claude, the brand skill helps the AI generate more brand-aware output
if you are working in Word, the branded Word template is what applies the document styling
if you are copying text from AI into Word, the template should usually take over and apply the document formatting automatically
A common source of confusion is expecting the same kind of branding behaviour from the AI tool and the Word template. They work differently.
If you are using a branded Word template
If your document in Word does not look right, the most likely reason is that the content has been pasted in using a method that keeps the source formatting instead of letting the template take over.
Check the following:
are you working inside the branded Word template, not a blank document?
did you open the with logo or without logo version from your toolkit?
did you paste the content in normally, so Word could apply the template formatting?
did you use a paste option such as HTML Format or another source-formatting option that may preserve external formatting instead?
if needed, have you applied Word styles manually to headings or sections that did not map cleanly?
In most cases, when content is pasted into the template in the normal way, the template should apply the formatting automatically. Manual styling is usually only needed when the pasted content comes in with formatting that overrides the template, or when the structure needs a little tidying afterwards.
If you are using a brand skill in ChatGPT or Claude
If you have installed your brand skill but the output still does not feel quite right, check the following:
is the skill definitely installed?
are you working in ChatGPT or Claude, where Brand Ready skills are supported?
did you mention the skill clearly in your prompt, for example Create a proposal using my brand skill?
are you expecting the output instantly, when the skill may take a little longer to work?
are you asking the AI to create something your current brand guidance defines clearly?
Skills often produce better results when the task is clearly framed. If the output feels too generic, try being more specific about what you want the AI to create.
For example:
Create a proposal using my brand skill
Draft a one-page report using my brand skill
Create a presentation outline using my brand skill
If your brand guidance is still evolving
Sometimes the issue is not that the skill is failing. It may simply be that the brand guidance available to it is stronger for some output types than others.
For example, your brand guidelines may define document styling clearly, but say much less about presentations, landing pages, or other digital formats. In that case, the AI will still do its best to work within your brand system, but it will be drawing from whatever guidance is currently available.
That means the quality and precision of the output depends partly on the quality and specificity of the brand material behind it.
Over time, you may find that your brand guidance needs to evolve as new use cases emerge. If you begin using Brand Ready for presentations, web pages, campaign concepts, or other formats that were not fully defined in the original material, it is natural to refine and expand those guidelines over time.
If you are using Copilot
Copilot does not currently support Brand Ready skills in the same way as ChatGPT and Claude.
That means if you are working in Copilot, your main branded workflow is likely to be your Word template rather than your skill.
In this case, the branding happens in Word through the template and styles, not through a skill running inside the AI tool.
If the platform itself seems to struggle
Sometimes the issue is not your Brand Ready setup at all. It may simply be a temporary platform issue.
For example, ChatGPT or Claude may occasionally:
fail to create a document on the first attempt
return a partial result
ignore part of the instruction on one pass
behave inconsistently from one prompt to the next
This can happen even when your skill is installed correctly and your prompt is clear.
If that happens, the simplest thing to do is try again. In many cases, a second attempt works perfectly well.
You can also try:
rephrasing the prompt slightly
making the task a little narrower
asking for one section at a time instead of the full output
returning to the branded Word template workflow if you need a faster or more reliable route
If the colours or fonts do not look right
Sometimes the issue is not that the branding is missing, but that it looks slightly different from what you expected.
Check the following:
are you using the branded Word template rather than a standard Word file?
have the correct styles been applied where needed?
are you comparing the document against the style guide, rather than against raw website colours or other brand materials?
are you expecting the full brand palette, when the Word template is intentionally limited to a tighter document palette?
It is also worth remembering that the brand skill and the branded Word template are not identical tools.
the brand skill can use a broader set of brand assets and references
the branded Word template uses a more controlled document system inside Word
So some differences are intentional.
If your skill output is taking longer than expected
This is normal.
A Brand Ready skill often takes longer than a standard AI prompt because the AI is working with more instructions, more references, and sometimes more files.
That can mean waiting a few minutes rather than a few seconds.
If you want a faster workflow, the branded Word template is often the better place to start.
A quick reset if you are not sure
If you are unsure what is happening, the easiest reset is:
Open your branded Word template
Add a heading and a paragraph manually
Apply Heading 1 and Normal
Check that the styles look branded
Then paste in one small section of AI-generated content and see whether the template takes over as expected
This quickly tells you whether the template is working as expected.
Most common reasons output does not look right
In practice, it is usually one of these:
you are not working inside the branded Word template
the content was pasted using a format that preserved source formatting instead of letting the template take over
the brand skill is not installed
you are using Copilot and expecting skill behaviour
the AI output needs a clearer prompt
your current brand guidance does not yet define that type of output clearly enough
the platform itself has had a temporary wobble
you are expecting instant output from a skill-based workflow
you are comparing the output to the wrong reference point
In short
If your output does not look right yet, the issue is usually not the content itself. It is usually a question of setup, workflow, paste behaviour, or expectation.
Start by checking whether you are using the brand skill, the branded Word template, or both, and then make sure the correct tool is doing the correct job.