Marketing budget allocation tool cut cost per acquisition by 10%

A high-street womenswear retailer cut its cost per acquisition by 10 per cent, and kept growing, after QuantSpark built it a bespoke, loyalty-aware tool for allocating digital marketing budget across channels and geographies.

The problem. The retailer was spending digital marketing budget across multiple channels, paid search, paid social and others, and across multiple geographies, without a reliable way to compare where a pound of spend paid back the most. It wanted an analytical toolset that could allocate spend to maximise return on advertising spend, rather than leaving the channel and market split to instinct or to whatever platform was easiest to read.

The harder part was not the analysis. It was turning cost, profitability and payback modelling, built on customer acquisition cost, cost per acquisition and customer lifetime value, into something marketing executives could use daily. A slide deck goes stale quickly. The retailer needed a live tool that recalculated the picture as the market moved and handed back a decision, not just a chart, to the people running campaigns. This is a familiar trap for multi-channel retailers generally: each channel gets optimised in isolation inside its own platform, while nobody optimises the split between them, or weights it by which channels bring back customers who stay rather than one-off bargain hunters.

The method. QuantSpark started with data, not a workshop. Historical cohort analysis established cost per acquisition and customer lifetime value for each channel and market, giving a like-for-like basis for comparing paid search against paid social, and one geography against another. That modelling then fed three connected outputs, handed to the client's own teams for ongoing use rather than left as a one-off recommendation: a self-service analytics data cube pulling together the retailer's web and advertising analytics sources; a live reporting suite tracking key metrics and campaign performance; and a bespoke decision-support tool that allocates marketing budget with customer loyalty factored in, so a cheap acquisition that churns fast is weighted differently to one that sticks around.

The workflow. The build ran from data to decision. Cohort analysis first fixed the unit economics, cost, profitability and payback, per channel and market. That fed a data cube connecting web and advertising analytics, which in turn powered a live reporting suite so the team could watch campaign performance as it happened rather than after the fact. The decision-support tool then converted that live picture into a budget allocation, loyalty-weighted, so spend moved towards channels and segments that brought back customers worth keeping. All three outputs, the data cube, the reporting suite and the decision-support tool, were then handed over for the retailer's own marketing team to run day to day.

The systems. Nothing here required naming a specific vendor stack. The build sits on four categories of system: the retailer's existing web and advertising analytics sources as the raw input; a marketing analytics data cube as the self-service query layer; a live reporting suite for ongoing dashboarding; and the bespoke decision-support tool as the layer that turns modelled economics into a daily allocation call.

The value. The headline result is a 10 per cent reduction in cost per acquisition, achieved while the business kept growing, so the saving came from smarter targeting and allocation rather than from spending less overall. That is the measured outcome of reallocating budget towards the channels and segments the cohort analysis showed converted more cheaply and retained customers for longer: a delivered result, not a modelled or indicative projection. A second, less visible dividend came from the tooling itself: the self-service data cube and reporting suite took manual report-building off the analysts' desks, freeing that time for higher-impact interpretation instead of pulling numbers together by hand.