# Case Study: A Growing Top Line Was Hiding a Worsening Unit Cost | Crank Source: https://wearecrank.com/case-studies/luxury-department-store-behavioural-2022 Search users converted three times better and content viewers 44% better, both reaching under a quarter of visitors, while cost per acquisition rose 25% inside a year that looked healthy on every headline number. --- ![Luxury department store case study: A Growing Top Line Was Hiding a Worsening Unit Cost](/images/case-studies/luxury-department-store.png?dpl=dpl_2uJjggmfov9ibhArygujfRBb3pax) Luxury Department Store, 2022 Audit # A Growing Top Line Was Hiding a Worsening Unit Cost Search users converted three times better and content viewers 44% better, both reaching under a quarter of visitors, while cost per acquisition rose 25% inside a year that looked healthy on every headline number. Challenge Total advertising cost fell 14 per cent year on year while cost per acquisition rose 25 per cent to £20 and revenue grew 20 per cent, a set of headline numbers that all looked healthy at once, while unit economics quietly worsened underneath them. The two strongest predictors of conversion on the whole site were reaching under a quarter of visitors, and no cross-selling had been built around the third of customers who were already crossing departments to buy. Approach Report by department rather than blended, give site search and editorial content far greater prominence so more visitors meet the two behaviours that already predict conversion, and build cross-department activity around the buying pairs the data already shows. Outcome The cross-department activity was built and deployed, and cross-department trading improved as a result, though the size of that improvement isn't separately documented, so it's stated here as direction rather than a number. Unit cost also became visible underneath the growing top line, so a rising cost per customer could no longer hide behind rising revenue. 01 ## Behaviours that predicted conversion, reaching almost nobody Across a £3.1m month, visitors who used site search converted at 7.9 per cent against 2.5 per cent for those who did not, in the strongest department the gap was 15.0 against 5.4\. Only 21.6 per cent of shoppers ever used search. Visitors who viewed editorial content converted at 5.2 per cent against 3.6 for those who did not, a 44 per cent lift reaching just 4.0 per cent of shoppers. Two of the largest measurable behavioural differences on the site, each available to only a small fraction of visitors. 02 ## One site, six-fold difference inside it Department conversion ranged from 1.5 to 8.8 per cent for new shoppers, with average order value ranging from £96 to £317 and cart abandonment from 62 to 86 per cent, inside one website. New shoppers converted at 4.1 per cent against 6.8 for existing ones, and the gap held across all four departments; new shoppers made up 86 per cent of visitors, so a blended figure flattered acquisition and understated the returning customer base. Between a third and two-fifths of customers bought outside the department that brought them in, 22 to 38 per cent, with clear repeating pairings, and no cross-selling activity had been built around any of it. 03 ## A cost problem hiding inside a growing top line Total advertising cost fell 14 per cent year on year while cost per acquisition rose 25 per cent to £20, driven by shifts within paid search and shopping. Revenue was up 20 per cent, so every headline figure looked healthy throughout, a deteriorating unit cost concealed inside a growing business. The analysis itself was careful to flag its own limits before drawing conclusions: basket behaviour was recorded inaccurately, traffic sources were being misrepresented, and year-on-year comparison was constrained by data retention, stated at the front of the report rather than discovered later, so a reader could tell which findings were solid and which provisional. 04 ## What building on the evidence changed Reporting by department rather than as one blended figure made department-level unit economics visible for the first time, so a rising cost per customer could no longer hide behind a growing top line. Site search and editorial content were given far greater prominence, putting more visitors in front of the two behaviours that already predicted conversion, and cross-department activity was built and deployed around the buying pairs the data had already shown, though the resulting uplift isn't separately quantified here. 05 ## What this means for you A growing top line hides a worsening unit cost for exactly as long as you look only at totals. ## What behavioural data showed, and what was built from it 7.9%vs2.5% Conversion rate, search users vs. non-searchers 1.5%–8.8% Conversion rate range, across departments of one site +25% to £20 Cost per acquisition, even as total cost fell 14% Deployed Cross-department activity built around the buying pairs the data already showed Drawn from Crank's account audit conducted in 2022 for this luxury department store. Client not named. ## Want to know what this looks like in your own account? These findings came from asking straightforward questions before any budget moved. We're happy to do the same for yours. [ Talk to Wearecrank](https://wa.me/447457416922?text=Help%20my%20business)[Get in touch](/contact) [← Back to all case studies](/case-studies)