Stop briefing great ideas in a Google Doc

For a creative industry, marketers spend a horrible amount of time trying to make ideas make sense on a page.

Decks. Concept docs. User journeys. Written briefs. Screenshots of vaguely similar examples. Those can be useful but there’s a limit to how well you can explain something in a Google Doc.

Getting buy-in on a new concept is tricky, especially if you’re asking clients to picture something they haven’t seen or done before. And this is where AI tools and platforms like Claude Design and Loveable are making a real difference.

Marketers used to rely on people's imaginations

Two people can read the same concept and picture completely different things.

That’s not really anyone’s fault.

If I’ve spent hours brainstorming an idea, I already know what I mean when I write “CRM benchmark analysis” or “platform migration hub”. I can picture where someone clicks, what happens next, what the results page looks like, and how we’ll repurpose everything across different channels.

A client might read the same words and imagine something different. Even if they work in marketing. Even if I’ve explained it really well.

There has always been an element of asking people to fill in the blanks.

Decks and write-ups still have a very obvious place, but the more complex or experiential an idea is, the harder it is to communicate in words alone.

Take an interactive diagnostic tool, campaign microsite, or slightly unusual content experience as examples.

You can pour a lot of time into describing all of those things; drawing the user journey, adding screenshots, talking through it on a call until everyone is extremely familiar with slide 14. And the client might understand each individual part.

What’s harder is piecing all of those parts together and seeing the experience through their audience’s eyes. What does it actually feel like to land on it? Where do you click? What happens next? Does the journey make sense when you’re not the person who wrote it?

Tangibility changes the conversation

Tools like Claude Design and Lovable have become incredibly useful because they let us turn a concept into something much more tangible, very quickly.

With Claude Design, that might be a process map, a case study mockup, or a branded content example that makes the direction of a campaign much easier to understand.

With Lovable, you can go even further and build out a landing page, custom tool or interactive experience for people to explore.

It doesn’t need to be perfect. Half of it might even be placeholder copy. The point, at this stage, isn’t to produce the final version. It’s to give the client enough to understand why it’s worthwhile and what the user experience will be.

You’re moving beyond a well-written description and a nicely structured document into something interactive, almost immersive.

And the feedback gets better too

We’ve found that clients react differently when there’s something tangible in front of them.

If we sent over six paragraphs describing an idea, the feedback is probably going to be fairly rational: Does the logic make sense? Does this match our ICP? Is anything missing?

But when you put something visual in front of someone, you get a different kind of response.

“I really like that.” or “This part doesn’t feel like us.”

“This ties in well with a separate campaign we’re running.”

“I can imagine the CFOs we work with using this asset.”

And if they can actually interact with it, you get another layer again.

They click somewhere you didn’t expect or hesitate at something you thought was obvious. All useful signals that help us build better content.

And ultimately, marketing isn't experienced as a strategy document.

Your audience isn't going to read the rationale before deciding whether they like the landing page. They just experience it.

Marketing is felt, not read. Where a written concept gives you considered feedback, something tangible gives you a reaction — and you probably want both.

Human thinking still matters

None of this means AI has suddenly made every marketer a designer or developer. Making something easier to build doesn’t automatically make it worth building (the classic “just because you can, doesn’t mean you should” still stands).

You still need to know what problem you’re trying to solve and have an idea worth exploring.

And you need enough judgement to spot when AI has done exactly what you asked for and somehow still produced something completely wrong (it happens…a lot).

If anything, making execution easier puts more pressure on the thinking that comes before it. Strategy has never been more important than it is today.

But once you’ve got an idea worth exploring, you can bring it to life much faster. And that means better conversations with clients, earlier feedback and more opportunities to get everyone excited about what you’re building together.

Cool things we’ve built recently

A few examples of what this has looked like in practice:

• Commerce Thinking’s Returns Report — see how mid-market ecommerce brands are handling international growth and returns

• Reveni’s Christmas Campaign — take control of your global logistics

• Polaris Growth’s Live Gameshow — test your CRO instincts

• Rotate’s Tech Survey - insights from tech leaders on the state of ecommerce

Different formats, different ideas, same basic principle: make the concept easier to understand, react to and build on. And that's what I’m finding most exciting about AI tools. Not necessarily being able to produce more marketing content, but being able to take the ideas we have so much further.