Campaign performance comparison

Campaign performance comparison is the practice of evaluating multiple marketing campaigns against the same normalised metrics so they can be ranked fairly. A 2,000 pound email campaign and a 50,000 pound paid media campaign cannot be judged on raw revenue, because the larger campaign will always produce more in absolute terms. Comparison works by reducing every campaign to efficiency ratios, such as profit per pound spent or cost per qualified lead, that hold regardless of scale.

Comparison matters because budget is finite and campaigns compete for it. Without a common basis, the loudest campaign or the one with the biggest headline number wins the next round of funding, even if a quieter campaign returned more per pound. Normalised comparison strips out size and exposes true efficiency, so the team funds what works rather than what looks impressive. It is the difference between allocating budget on evidence and allocating it on instinct.

The discipline lives in making campaigns genuinely comparable. The same attribution model, the same cost definition, and the same time window have to apply to every campaign in the set. Compare a campaign measured on last-touch and media-only cost against one measured on multi-touch with full costing and the ranking is meaningless. Comparison is only as fair as the consistency underneath it.

A campaign comparison is built by computing the same efficiency metric for every campaign in the set, then ranking them. The core ratio is gross profit divided by total cost, but a sound comparison layers in volume and quality so a tiny but efficient campaign is not ranked above a large one that drives real scale. The inputs below are what make the comparison trustworthy.

1. 1

   A consistent efficiency ratio

   Pick one primary measure, usually gross profit per pound of cost, and compute it identically for every campaign. This is the spine of the ranking and must use the same attribution and costing rules throughout.

2. 2

   A volume measure

   Record the absolute profit or customers each campaign produced alongside its ratio. A campaign with a superb ratio on tiny spend may matter less than a slightly less efficient one driving the bulk of pipeline.

3. 3

   A quality measure

   Track lead quality or retention of the customers each campaign won. Two campaigns with the same cost per lead are not equal if one delivers leads that close and stay while the other delivers churn.

4. 4

   A common time window

   Hold the measurement period constant across campaigns. Comparing a campaign with a 90-day window against one cut off at 14 days unfairly penalises the slower-converting campaign.

A worked comparison shows why ratio alone is not enough. Campaign A spends 2,000 pounds and returns 8,000 pounds of gross profit, a ratio of 4.0. Campaign B spends 40,000 pounds and returns 120,000 pounds, a ratio of 3.0. On efficiency, A wins. But B produced 120,000 pounds of absolute profit against 8,000 from A, and B may be the only campaign capable of scaling. The right read is to scale A toward its efficiency ceiling while keeping B as the volume engine, a decision the comparison only surfaces because it holds both ratio and volume in view.

A metric tree turns a flat campaign ranking into an explanation of why each campaign sits where it does. Instead of a leaderboard that says campaign A beat campaign B, the tree decomposes the efficiency ratio into the drivers behind it, so the team can see whether one campaign won on cheaper acquisition, stronger conversion, or higher deal value.

The comparison decomposes into the same branches for every campaign: how efficiently it acquired attention, how well it converted that attention, the value of what it closed, and the cost it carried. Laying the trees for two campaigns side by side shows exactly where the winner pulled ahead. A campaign may rank top not because it is broadly better but because its cost per lead was half the rest, a fact a bare ranking would hide.

This is the gap between a dashboard and a decision. A dashboard ranks campaigns by ROI. The tree explains that the top campaign won purely on conversion rate, which tells you to copy its landing page across the others rather than simply shifting all the budget to it.

There is no universal benchmark for a comparison itself, because the right reference depends on the channels in the set. What benchmarks well is the spread between your best and worst campaigns and how decisively you act on it. The bands below describe what a healthy comparison spread looks like and what it signals.

Treat a very wide efficiency gap as a prompt to check the measurement first, not just the campaigns. Extreme spreads are often an artefact of one campaign being credited generously by last-touch attribution while another is costed in full. A comparison that flatters one channel and punishes another tends to reveal a methodology flaw before it reveals a real performance gap.

Improving a campaign comparison means making it fairer and more actionable, not just running more campaigns. A comparison earns its keep when it reliably points budget toward the levers that produce return. The metric tree makes that possible by showing where each campaign won or lost rather than only that it did.

KPI Tree models each campaign as a branch in a shared tree with RACI ownership on every node, so the media team owns acquisition cost while the web team owns conversion strength and the comparison is no longer a static spreadsheet. When one campaign pulls ahead or falls behind, the change is pushed to the accountable owner rather than waiting for the next review. The verified impact loop then checks whether copying a winning lever into another campaign actually improved its ranking, closing the gap between a dashboard that compares and a team that acts.

1. 1

   Comparing raw revenue across sizes

   A large campaign always wins on absolute revenue. Without normalising to efficiency ratios, the comparison just rewards whoever had the biggest budget.

2. 2

   Mixing attribution models in one ranking

   If some campaigns are measured on last-touch and others on multi-touch, the ranking compares methodologies rather than campaigns. Use one model throughout.

3. 3

   Ranking on efficiency alone

   A tiny campaign with a brilliant ratio can top the table while contributing almost nothing to pipeline. Always hold absolute volume alongside the ratio.

4. 4

   Ignoring lead quality differences

   Two campaigns with identical cost per lead are not equal if one delivers customers who churn. Bake quality and retention into the comparison, not just front-end cost.

5. 5

   Comparing across mismatched time windows

   A slow-converting campaign cut off early will look worse than a fast one measured fully. Hold the window constant so the comparison is fair to longer sales cycles.