What he is actually building
Thomas McKenna runs a company that writes product descriptions. This sounds unglamorous, and it is, which is precisely the point. Somewhere in the catalogue of every large retailer there are tens of thousands of product pages, most of them written once by someone who has long since left, never touched again, and quietly failing to be found. McKenna's company, Ocula Technologies, exists to rewrite all of them - at once, on brand, and in a way that both Google and the newer answer engines can understand.
He is CEO and co-founder. The company was born in 2021 out of a specific irritation: McKenna and his co-founder, the data scientist Dr. Gregory Fletcher, kept watching retailers try to make profitable use of their own product data and fail. Vendors failed them. In-house proofs of concept failed them. The observation that became the company's founding line is blunt - "the majority of retailers and brands globally fail to make profitable use of one of their greatest assets: data."
The bet underneath Ocula is a bet about how search is changing. For twenty years, being found online meant ranking in a list of blue links. Increasingly it means being the answer a conversational AI gives when someone asks it what to buy. The industry has a clumsy new acronym for this - AEO, answer engine optimization, the sequel to SEO - and Ocula was built for that world slightly before most brands were willing to admit it was arriving. The technology moves product data from keyword-stuffing toward intent, generating titles, descriptions and metadata designed to be legible to an algorithm that increasingly talks back.
The client list is the proof. Marks & Spencer, Frasers, Boots, AO - more than 25 major retail customers, with the company splitting its people between London and Belfast. In December 2025, M&S went public about the partnership: Ocula's AI generating search-optimised product copy across the British institution's online catalogue, without losing the M&S voice. That last part is the hard part, and it is the part McKenna keeps insisting is a human-plus-machine job rather than a machine-only one.
The one-line thesis
A mid-size brand can't out-hire a giant's content team. It might be able to out-compute one. That gap - between the brands with armies of copywriters and everyone else - is the market Ocula is trying to erase.
Funding, stacked
Backers include Praetura Ventures, Lloyds Banking Group, Castelnau Group, NYO Capital and Jose Luis Gomes.
What the platform claims to move
Figures as stated by the company and in trade press. Bars are illustrative.
A resume that reads like a retail-data greatest hits
Before founding a startup, McKenna assembled the sort of background that makes the pitch credible in a retail boardroom. Procter & Gamble, the company that arguably invented modern brand management. dunnhumby, the data science firm best known for building Tesco's Clubcard into a loyalty-data machine. Monitor Deloitte and Bain & Company, two of the strategy consultancies retailers pay to be told what is wrong with them. He holds an MSc in Business from Alliance Manchester Business School.
There is a familiar arc here, and McKenna followed it: the consultant who spends long enough diagnosing the same problem for other people that he eventually decides to just go and solve it himself. The problem, in his telling, was always sitting in plain sight in the product data. The people best placed to see it were the ones being paid to advise, not to build. So he switched sides.
The company he built is deliberately technical at its core. The co-founder is a data scientist. The team, as McKenna describes it, is "some of the brightest science, engineering and product talent in the UK and beyond." When it came time to place the research-and-development center - the actual engine room - he did something a little contrarian.
Consulting and data roles across Procter & Gamble, dunnhumby, Monitor Deloitte and Bain & Company.
Co-founds Ocula Technologies with Dr. Gregory Fletcher. Becomes CEO.
£5m from Lloyds Banking Group and Castelnau Group. Belfast software HQ opens with £11m R&D commitment.
Series A closes. Client base passes 25 major retailers.
Marks & Spencer partnership goes public.
How he talks about the work
The majority of retailers and brands globally fail to make profitable use of one of their greatest assets: data.
We've achieved this by bringing together some of the brightest science, engineering and product talent in the UK and beyond.
By combining their team's expertise with our AI, M&S are creating search optimised, high quality product copy at scale, helping more customers find the right products faster.
Users consistently tell us that our generative AI is levelling the playing field with their most advanced competitors.
The specifics
Ocula's AI is credited with a 15% revenue lift for the Kansas City Chiefs' online merchandise store - an American football team getting help from a Belfast lab.
The company runs on two co-founders: McKenna as CEO and Dr. Gregory Fletcher as CTO. Two people, 25-plus enterprise retail clients.
Its flagship tool, Ocula Boost, auto-generates SEO-optimised product descriptions and product-page improvements in bulk.
The people are split between London and Belfast; the AI platform itself is built by the Northern Ireland team.
McKenna's stated ambition has included tripling the client base within two years and building serious North American sales.
The pre-founder resume - P&G, dunnhumby, Deloitte, Bain - is why the retail-data pitch lands in a boardroom.
Level the field, then keep building
The aspiration McKenna returns to is not really about copywriting. It is about access. If generative AI genuinely lets a mid-size brand produce the same volume and quality of on-brand product content as a giant, then one of the durable advantages of scale - the ability to throw a hundred writers at a catalogue - quietly evaporates. That is the "levelling the playing field" line, and it is either a marketing slogan or a real structural shift, depending on how the next few years of AI search shake out.
For now the plan is more concrete: triple the client base, deepen the Belfast R&D center toward 50-plus staff, push into North America, and keep making product pages that both people and machines can actually read. The unglamorous text on the internet, it turns out, is where a surprising amount of retail money is quietly won or lost.
The stakes, plainly
Search is shifting from lists you scroll to answers you're given. In that world, whoever writes the product data that machines trust decides which products get surfaced. McKenna is betting that job is worth a company. Twenty-five retailers, and counting, agree enough to pay for it.