Artificial intelligence is moving incredibly quickly, with new tools launching weekly with promises to revolutionise how the cycling industry does business. But for many brand marketing and sales teams, separating the hype from the genuinely helpful tools is a daunting task.

Enter Manual Focus, a Sheffield-based marketing consultancy dedicated to helping endurance sports and cycling brands navigate this new frontier. Founded by James Vickers, an industry veteran with deep roots in emerging tech, Manual Focus aims to shift how the bike trade uses AI.

Rather than just churning out social media captions, the consultancy advocates for using AI at the strategic level, utilising innovative, proprietary tools like the ‘Rider Panel’ to simulate focus groups and test launch plans before budgets are committed.

BikeBiz caught up with the founder of Manual Focus to discuss the origins of the company, the pitfalls of lazy AI-generated copy, and why true marketing success still comes down to deeply understanding the rider.

Tell us about Manual Focus. What inspired you to start the company?

Manual Focus is a marketing consultancy in Sheffield for endurance sports and cycling brands. We work wherever new technology is changing how people decide what to buy, and right now that means AI. As well as marketing work for clients, we help smaller companies put AI to good use inside the business, and we build our own research tools under the Light Meter name.

Every week, one of the big AI studios, like OpenAI or Anthropic, who make ChatGPT and Claude respectively, launches a new AI product, each one promising to do more than the last. The trick for cycling brands’ marketing and sales teams is to work out which of those tools will actually help, which jobs to hand them, and how to fit them into the website, social posts and business development work they’re already running.

Cold outreach and sales is a good example, because you can now research each prospect in detail and build a landing page for them, instead of sending everyone the same link. I’ve spent more than 15 years marketing in emerging tech, usually a little before the mainstream got there. AI felt like the next one, so I built a business around it.

How have your past roles informed the way you approach Manual Focus?

I was on the founding team at RGT Cycling, which grew exponentially before Wahoo Fitness acquired it. That’s where I learned how much it’s worth getting into a market early. I’ve also worked at several cycling brands, and these days I consult for Ribble Cycles.

Those roles taught me that good marketing starts long before the campaign. The brands that do well know exactly who they’re for, what those people want and what they’re trying to achieve, and they plan from there. The ones that struggle tend to jump straight to tactics, like a new channel or a new product, without a plan behind it. Manual Focus is built around getting the planning right first, and then using whatever technology helps deliver it.

What’s your one piece of advice for cycling brands using AI in their marketing?

Use it to plan, not just to produce. Most brands start using AI for output, like writing posts or making images, which feels like the easy end. The bigger win is earlier, in the planning: working out where you are now, where you want to get to, how you’ll get there and how you’ll know it’s working. AI can now help with all of that, including the parts that used to need expensive research, like understanding what your customers want and testing a plan before you spend on it. That’s what our Light Meter and Rider Panel tools are for.

That matters because good marketing is about helping people become the version of themselves they want to be, whether that’s a first sportive, a faster season or just getting out more. The better you understand that, the better everything after it works. AI gives brands of any size a way to understand their customers properly, and that’s where I’d start.

One word of caution on using AI to make the marketing itself, especially promotional images and copy. Set clear rules for it, then read and review everything carefully. Strip out the tell-tale signs of AI-written content, like the infamous em dash, because if you can’t be bothered to pay attention, why would your customers?

What motivated you to build the Rider Panel, and how can brands make the most of it?

The Rider Panel grew out of Light Meter, our AI scoring framework. The challenge most brands have is that market research is expensive, so they only do it for their biggest launches and go on instinct for everything else. I wanted brands to be able to test a launch plan before they commit the budget.

The Rider Panel is a bit like a focus group made of simulated riders. Each one is an AI character with a life story of their own, from who they live with to the kit they already own. They go through a season month by month and tell us why they bought or why they walked away. Then we change one thing, such as the price or where the product is advertised, and the same riders go through the same season again, so we can compare the plans.

Brands get the most out of it when they bring two or three plans they’re torn between. The panel narrows those down, and then the brand takes the best one to real customers. In our first study, a made-up folding smart trainer sold to 26 of 500 simulated riders when it had independent reviews, and to 3 without them. A £399 Black Friday price took it to 41 buyers.

I should be clear that these results are simulated, but they attempt to link tactics to sales and surface opportunities, weaknesses, and customer feedback that might have been ignored.

What’s the ultimate goal for Manual Focus over the next 12 to 18 months?

To show how the Rider Panel can work for brands as part of how we help them plan. It comes in at the planning stage, giving brands a cheaper way to understand their customers and test a plan before they spend money. Then we help them carry that plan through to the campaign itself.

More broadly, I’d like testing a launch before you spend on it to become the norm for endurance brands, and we can already build a panel for any country using that country’s own data, so we can expand into more markets and product categories.

A note – The baseline awareness figure is a setting in the simulation, not a survey result, and the report shows how it was set.