The Very Group’s Nicole Dutton is a marketer who has been through her fair share of digital transformation. She says unlocking customer data is key to helping multi-category retailer Very embrace the next phase of e-commerce growth.
We had completed a digital journey a century in the making. With origins stretching back to mail order catalogue businesses founded in 1923, The Very Group has become a digital-only, e-commerce powerhouse in the U.K. In our full year 2024 financial year alone, Very Group made 44.8 million deliveries and £2.1 billion in revenue.
But new technological advances, changing privacy regulations, and evolving shopper behaviour called on us to turn to a new page — again.
To support the next phase of our growth, we wanted to harness the ability of AI-powered ad platforms to effectively reach the right customers, at the right time.
The only problem? The success of this technology lies in the usage of first-party customer data being utilised for marketing activity — and how we unlock the power of ours needed modernising.
Customer data bottleneck
The boom in home shopping brought a surge of new customers to Very — but not all customers shopped the same way. Many made small, sporadic purchases, while others were loyal, high-value shoppers. We needed to ensure every pound of acquisition marketing spend focused on finding and engaging these most valuable customers.
With 3.7 million active Very customers, a logged-in environment, and our own flexible ways to pay proposition, we were sitting on one of the richest first-party datasets in the retail world. We know our customers extremely well — but our marketing was struggling to find our bullseye: more of our highest-value customers.
The reason was twofold:
- We were working with a monolithic, aging tech stack. While our customer data was very well understood internally, we weren’t unlocking the full potential of it across our marketing platforms. For instance, we were limited to manual, infrequent uploads of data.
- Privacy trends and the addition of consent banners meant we could no longer measure in the way we used to, losing sight of performance indicators and, therefore, impacting performance.
We were at a digital crossroads. For a business whose catalogues were once a mainstay in U.K. living rooms, it was time to turn over a new leaf and adapt.
Transforming our marketing with customer data
We started by addressing the measurement privacy challenge. We matched our e-commerce conversion data against Google accounts in an anonymised way using Google’s enhanced conversions feature. We also collected anonymised user data when customers opted out of cookies by using the consent mode feature. Additionally, we moved to Google Analytics 4, which allows us to see a broader range of user events. This work led to a combined 19% uplift in measured conversions, leading to a 7% increase in return on ad spend (ROAS).
With this new measurement foundation in place, we worked with our agency Jellyfish to turn our customer data into marketing gold.
Automating first-party data transfer
We set out to capture and activate more data into our marketing platforms, faster.
Richer data: Previously, we were leveraging data only from opted-in email customers – a small subset of our total customer base. By changing our privacy policy to cover more activities, we significantly increased the pool of usable first-party audience data.
Faster data: Whereas we previously updated our data manually, we are now able to automate the upload on a daily cadence.
The actions meant we improved freshness and accuracy of our data, built on a privacy-centric foundation. Our tests also revealed a 13% uplift in conversions, because we are reaching an audience that is engaged from clicking to actually buying.
Optimising for long-term customer value
Aiming to reach higher-frequency, longer-lasting, and bigger-spending shoppers, we also pivoted our ad-buying goals from mere conversion revenue and return on investment to highest-margin and longer-term results.
Bidding for profit: Traditionally, we had bought Search ads to meet sales demand goals. But we realised that not all demand is equally profitable — some products have higher margins, some customers spend more than others. Now that we can add product transaction data to our ad platform, our algorithm understands how to bid against the most-profitable conversion outcomes.
Focus on the future: Because we know how important our credit card holders, repayment customers, and frequent shoppers are, we can now create lookalike audiences. This means we can optimise our ads to reach the right future audiences, at the right time.
Predicting shoppers’ payment preference
Our repayment and credit options aim to help families get more out of life, even in difficult times. And we know these shoppers tend to be stickier, repeat customers. Freeing our customer data helped us build a model that identifies the most likely credit card customers from our browsing traffic.
We now use historical customer event and behaviour data to help predict which customers are most likely to purchase using one of our Very finance options. This is done through Google Tag Manager in real-time as customers browse the site. This allows us to create predictive audience segments we can use to re-engage users with custom messaging.
It is like a crystal ball for picking out and reconnecting with our most valuable future customers.
Closer to customers
Our new practices mean we have a more solid basis for our investment decisions, one which helps us achieve our goal of getting “close to our customers”.
Making technology shifts like these starts with people. As a marketing department, we needed to step out of our own silo to build relationships with the C-suite, our commercial finance partners, and customer experience team. This included interview workshops with more than 20 group stakeholders and forming a cross-functional ad-tech squad.
That is how you gain the trust that’s needed to provide investment for a strategy that turns your customer data into a customer value flywheel.
