“Instabrand?” 5 Top Tips for Branding & AI: Unravelling Buzz from Hype

Just pop in your brief… get some amazing creative… some plausible words… and off we go… Right? Sadly, not quite…

At a time when algorithms can write headlines and create logos in seconds, the question of how best to use AI in branding right across industries is a hot topic. 

As an agency that specialises in digital transformation, with in-house data and AI teams, our advice is that the smart approach is ‘blended’. That means: take the best of what AI can do, apply with nuance (the approach is different for strategy vs naming vs creative or copywriting); and with the efficiencies it generates, maximise the moments that truly connect - and empower your brand growth.

Why not just go all in? AI’s limitations

As FLO CEO Andrej Hájek notes: the technology is still evolving fast, whether generative AI for creative work, or analytical AI for efficient data processing, and the next generation will be different again. That is: Artificial General Intelligence, a multifunctional form capable of solving complex problems, whose form is still a work in progress.

Crucially, there are some important things AI in any form CAN'T do, like:

• Secure ownership rights to visual creative, and avoid legal pitfalls

• Go beyond the limits of today’s programming: like a focus group, AI is good with existing knowledge, but not at the surprising, elevating twists that characterise the most impactful work 

• Always get it right  a – there are famous cases of AI pooling existing data and drawing the wrong conclusions. While humans aren’t immune, brand creation always needs seasoned, expert eyes.

Using its potential: 5 Top Tips

→ Naming & Trademarking

We propose a naming tool for the ‘best of both worlds’: created in- house, it blends generative and analytical AI to create on-strategy lists of names that also undergo a number of availability checks – in seconds. 

But these too are imperfect. Take pharma, which needs a human brain, as it’s all about the fallibility of misprescribing drugs that can look or sound similar. So we’ve given prominence to interventions and checkpoints by experts, and never run creative only by machine. 

→ Visual Identity & Creative Direction

As a studio, we don’t advocate for purely AI-designed creative. Instead, we can use it to explore concepts in parallel with our own ideation. To road-test, expand, iterate… But never to be the final design. 

And while it can save time on process, it is no shortcut to truly remarkable, ownable creative. Our award-winning campaigns, e.g. for Porsche, came from people deeply immersed in the topic (in this case, total car nuts) creating something that they knew first- hand would captivate the audience. 

→ Copy & Tone of Voice

AI models are trained on data. We’re trained on nuance and moments of surprise. So while AI is useful at scale to stress-test a tone of voice across platforms (and we can help show you how), we don’t start or end with it. Again, not least for IP reasons, it should also never replace a crafted solution. Machine does general very well, but not unique. And it’s your voice after all. 

→ Brand Strategy & Positioning

AI is famously prone to error in factual data-collection, so is never a replacement for fully fledged insight, out of which we derive a positioning. Also, AI is no help at all in a workshop situation, which is crucial for exploring stakeholder dynamics. These are, as often as not, non-verbal. But any positioning needs validation – and here AI is a fast way to scan, summarise, and pattern-match against competitors, look for untapped opportunity spaces (which we still do the human way, too).

→ Content & Campaigns

Generative tools help us rapidly scale ideas and localise assets, freeing up our creative team to craft moments that actually connect.

Oh, and… yes, I ran this article through AI, but only after it was written. Perhaps unsurprisingly, it seemed to agree. 

How’s that for human and machine in harmony?