Read time: 04'24''
22 September 2022
Why online retailers are missing hidden sales opportunities
Unsplash © Bruno Kelzer

Why online retailers are missing hidden sales opportunities

It’s been sixteen years since ‘Google’ became a recognised verb in the Oxford English Dictionary, and it is still our first port of call when looking for something. We’ve all been there. Who hasn’t Googled for a new pair of Veja shoes or a ME+EM maxi dress? Google Shopping allows shoppers to search for any product online to easily compare items and prices across sellers – but a simple listing is only the starting line for retailers to gain a competitive advantage.

In the eCommerce industry, I personally hear from founders, CMOs and CFOs monthly, and the general gist is that performance marketing is the only area that they don’t have a forensic understanding of how the budget is contributing to their bottom line.

Retailers are often frustrated and feel trapped in a bidding war on the items they broadly predict consumers want to buy. Business leaders are exasperated by slow data analysis and ineffective campaigns leading to zero revenue. As purse strings get tighter, the future looks bleak for online sellers using the same limited levers to sell products. So what can be done?

Acting on the intangible

Optimising tangible factors is a game that everyone in consumer retail is playing. Typically this will centre around seasonal trends – retailers will ensure they are competing with the best ad strategy accompanied by compelling, search-optimised copy on items like swimwear and paddling pools during the summer months, for example. Arguably, this has become a zero sum game for retailers as CPC (Cost Per Click) becomes higher with more Google ad buyers in the market. This strategy can only take you so far.

The intangible factors can’t be predicted and are impossible for humans to plan for. Unexpected weather changes, a viral TikTok video or a celebrity spotted wearing a particular item of clothing can instantly change what consumers want to buy. Unexpected trends don’t fit neatly into agency working hours – they need to be acted on in real-time before the buying moment is lost. This is where AI steps in.

Machine learning technology has the ability to closely monitor changes in Google search trends and match that intel with retailers’ inventory data to automatically action the best commercial advertising decisions for retailers in real-time. Processing thousands of data points in seconds is something that humans simply can’t do – unlocking new growth opportunities for retailers beyond paying more for ineffective advertising.

Humans simply can’t do it all (and that’s fine)

Agencies can certainly support, but when it comes to utilising a platform like Google to sell products online, there’s a lot to manage – multiple levers to pull, alongside reports and metrics to watch and adapt closely. That’s why freeing up time and reducing the effort for an in-house team or agency team by removing manual processes and automating data mining, analysis and optimisations is a no-brainer.

Do you know the spread of marketing budget across your inventory? Which SKUs are driving revenue? What % of your inventory is Google ignoring? Could the budget be better allocated? If you work in retail and can’t confidently answer these key performance questions and know how you’d make improvements, then something’s not right.

In my opinion, if retailers are planning to ride through this period of inflation with the same tools and techniques they’ve been using for years, then they are planning to fail. AI provides a new competitive edge in a tough market and cannot be ignored.

What the data tells us

According to McKinsey’s 2021 Global State of AI Survey, based on responses from more than 1,800 executives, AI adoption and its impact is continuing to grow. 67% of all respondents saw revenue uplift and 79% saw cost reduction (up significantly from 44% last year).

Specifically within the eCommerce industry, machine learning technology allows retailers to match their inventory data (supply) with consumer data (demand) in real-time to automate advertising decision-making and drive better business results. This may sound obvious, but this level of data connection is not yet the norm. At Upp, we exposed that retail businesses are spending money wastefully, and up to 40% of advertising spend generates zero revenue.

CPCs increase with demand during predicted sales periods – often resulting in a loss on items e.g. fans and air con units during the recent heatwave. However, one of our clients, an online marketplace, saw an 89% increase in revenue, and a P&L improvement of 113% thanks to an unforeseen spike in wardrobe sales that we capitalised on. Whilst fans and aircon drove GMV (Gross Merchandise Value) achievement, it was the sales of wardrobes that drove profitability achievement due to the low CPC – higher margin, lower ad cost – a hidden opportunity that was discovered by machine learning.

Without AI this was simply an impossibility…

Similarly, Charles Tyrwhitt recently achieved a 42% increase in UK monthly orders and a 14X return on investment by automating ad optimisation. With machine learning technology working at SKU-level, retailers like Charles Tyrwhitt can target buyers with hyper accuracy and show them the exact product they’re looking for, precisely when they’re looking for it.

Our industry-first AI performs millions of micro-optimisations dynamically, balancing unexpected trends with changes in advertising costs – all within a framework of business objectives set in the Upp platform. A real-life example of this is Poundshop, which drove its impressions on Google Shopping up by 375% whilst maintaining its ROAS target – through acting on the hidden opportunities and unexpected trends that traditional retail performance marketing cannot capture.

The best way to predict your future is to create it

Reinventing how retailers trade, make decisions and act is critical right now, especially in a world where disposable income is shrinking and uncertainty is high. In the highly competitive eCommerce landscape, retailers have to think smarter and integrate AI to scale and automate their retail performance.

In the words of Abraham Lincoln, ‘the best way to predict your future is to create it’. With this in mind, it’s easy to see that AI technology has the power to process data much faster than humans and even learns and optimises on the job. Most businesses have realised that Google Shopping (soon to be Performance Max) is an essential channel for retailers to sell products and offers a huge opportunity for eCommerce growth, but many are still learning how to apply a data-led approach. The best advice I can share is to take advantage of the technology at your fingertips and you’ll reap the benefits in what’s a new era for retail.

Drew Smith is CEO and cofounder of Upp.