It can be a fine line between success and failure in eCommerce.
Attracting people to your website is only the first step. And if you think that showcasing your best-selling products to visitors is enough, you may be wrong.
Andr茅 Brown co-founded merchandising platform Attraqt in 2002 and led it as CEO for 16 years – overseeing the group鈥檚 admission to London鈥檚 junior AIM public market in 2014 – before leaving and turning his attention to his next venture.
Advanced Commerce, founded in 2020, has built GrapheneHC, a platform which combines three technologies 鈥� site search, visual merchandising and product recommendations 鈥� to help clients deliver a better, more personalised online shopping experience on their websites, increasing sales and reducing drop-off rates.
鈥淲e’re the only platform which is actually designed for merchandisers: the giveaway is that everybody else talks about product discovery, which is very much a frontend way of looking at things – a lot of that’s about surfacing your best-sellers and so on,鈥� he tells 老九品茶Cloud.聽
鈥淭he reality is merchandisers don’t really care what the best sellers are. What they care about is stock that’s proving hard to shift. So what are you going to do about that?聽
鈥淚t鈥檚 all about sequencing.鈥�
Brown said his original vision for the company – which placed fifth on our RetailTech 50 ranking this year – was somewhat different to the platform of today.
鈥淢y original plan was to build a framework for headless commerce. That’s why the company is called Advanced Commerce,鈥� he acknowledges. 鈥淏ut we quite quickly realised that the whole headless commerce space is really an enterprise space. We realised that by the time we built it out – long sales cycles, long deployment cycles – we’d run out of money.聽
鈥淎s part of the journey, we’d started developing a 鈥榮earch and merch鈥� microservice to illustrate the framework鈥� we realised that we understood this really well, and so we pivoted.鈥�
Integrating with existing eCommerce platforms, GrapheneHC allows users to blend product attributes like price, size and 鈥榠n stock鈥� with site metrics such as popular, trending and average basket adds, plus further metrics from external data feeds if required. GrapheneHC then uses this blend to determine a personalised product display sequence for the potential customer.
It also has a merchandising templates function which gives retailers complete control over where merchandising blends are applied on a page.
It says that the average shopper only looks at the first three pages of an online store, which can be as little as 5% of the product catalogue: delivering unique digital experiences increases online store performance and also the connection a customer feels with the retailer鈥檚 brand.
RetailTech 50 – UK’s most innovative retail tech creators for 2023
At a recent roundtable of RetailTech 50 and MarTech 50 firms – held in association with Refresh PR – Brown produced a deck of cards to make his point. 鈥淚f you all visited the same website, and these cards represented the products on that site, should you all be shown them in the same sequence?鈥� he asked, peeling them off one by one.
鈥淭he same customer profile is being applied to everyone, [which makes no sense].鈥�
After the joker made its planned appearance in the sequence, he further illustrated his message. 鈥淎s you segment your customer base, you鈥檝e got people with different preferences for, say, brand, colour and price. There鈥檚 a lot of focus right now on trying to make sure that you鈥檙e showing an appropriate product sequence to each customer.
鈥淏ut there鈥檚 also an inherent conflict between AI and human merchandisers: AI is data-driven personalisation, whereas if you鈥檙e a merchandiser, your role is to promote the brand鈥檚 DNA on the website 鈥� and the way you do that is through the way you curate pages.聽
鈥淲e鈥檙e doing a lot of work at the moment about how you can blend the two together.鈥�
Brown says a big learning from his time at Attraqt was to not build connectors to every platform but instead 鈥渢o build everything API-to-API鈥�.
鈥淚n my previous business, we built a Magento connector and it was just awful because you get conflicts and contention with other plugins. The API approach is much better,鈥� he explains.
鈥淲e also learned a lot of lessons about how to do the onboarding process: you explain to clients ‘these are your options’ rather than saying ‘what would you like?’鈥�
Advanced Commerce launched without funding but has since been through three oversubscribed fundraises and grown its workforce to 20 people, including tech specialists who worked with Brown at Attraqt. It now serves 40 clients.
The CEO says the primary focus is on the UK market. 鈥淲e do have clients in Europe, Australia and America, but right now we are very focused on just growing our UK customer base,鈥� he says.
鈥淭he next round of funding we’ll do is a Series A – it might be next year.鈥�