CASE STUDY

CASE STUDY

Sector: Ecommerce

Achieving 734k profit growth with propensity-driven attribution

Lovehoney wanted to improve the ROI from paid advertising campaigns

We knew this was possible using our technology, Cubed’s comprehensive revenue modelling data, which attributes value based on each customer's purchase probability profile.

To test this, we ran a geo-split test.

Hypothesis:

By optimising paid search campaigns to cubed full model revenue data we will drive more topline revenue than by optimising to Google-tracked revenue.

Our Methodology:

→Run a robust geo-split test

Using Impact Navigator, a tool from Skai, we conducted a robust geo-split test. This used historic data to determine clear location groups and measure the impact across all GA channels.

→Split the UK into three groups

Using this location data we set up a geo test, splitting the UK into three; two groups of comparable locations (A and B) and a third group of outlier locations (C) not suitable for the test (e.g London).

→Adjusted bidding strategies

We then used cubed full model revenue data to bid to group B, while group A kept bidding to Google-tracked revenue data. We measured the topline revenue and spends of both groups to determine the winning conversion action.

This resulted in +£734K Profit Per Year

(UK only, so far)

The Effectiveness Truth, Proven

Our analysis demonstrated three fundamental truths about modern digital marketing:

 1. Channel Cannibalisation is Real

   - Paid search was competing with, not complementing, organic search

   - Multiple channels were targeting the same customers

   - Budget was being wasted on internal channel competition

 

2. Attribution Models Need Rethinking

   - Traditional models significantly undervalued organic channels

   - Channel interdependencies were misunderstood

   - ROI calculations were fundamentally flawed

 

3. True Incremental Growth Requires New Thinking

   - Focus needed to shift to genuine new customer acquisition

   - Upper-funnel activities were being neglected

   - Budget was being wasted on capturing existing demand

Our Response

Based on these insights, we developed a bespoke approach to:

  1. Target true incremental growth opportunities

  2. Focus on new customer acquisition

  3. Eliminate channel cannibalisation

  4. Drive genuine marketing effectiveness

This renewed strategy:

- Eliminates of wasteful channel overlap

- Reduces paid targeting of existing customers

- Ensure more efficient budget allocation

- Focuses on true incremental growth

Sometimes the most powerful marketing insight comes not from what you do, but what you stop doing. This experiment demonstrates how questioning conventional models and testing fundamental assumptions can reveal millions in potential savings and drive true marketing effectiveness. We believe most brands require a similar audit of channel management.

This experiment gave us the environment to truly test the long held beliefs about the marketing mix. The reality is that marketing has changed exponentially in the last year alone, previously considered best practices are now outdated. This moment requires ambitious brands to face hard questions about whether their marketing is still effective.

This is the sort of challenge we love to solve.

Do you need our help?