Calculate EPC with a clear denominator
Earnings per click is attributable earnings divided by clicks. If a campaign produces 180 currency units from 1,500 clicks, the result is 0.12 per click. Some reports express earnings per 100 clicks, which would show 12 for the same example. Check the reporting convention before comparing two numbers that appear to describe the same metric.
You also need to know what earnings means in that report. Pending commission, approved commission and money already paid are different amounts. A useful comparison names the status, currency and date range beside the figure. The arithmetic is simple; choosing consistent inputs takes more care.
Give conversions time to arrive
Today's clicks may generate conversions tomorrow or next week. If you compare a mature campaign with a new one after a few hours, the newer campaign can look weaker simply because its conversion window has barely started. Review groups of clicks from comparable periods and allow for the ordinary delay in that buying journey.
Small samples need particular restraint. One sale can move a low-volume EPC sharply, and the next group of visitors may behave differently. Keep the click count and conversion count beside the average. When you decide whether to continue a test, include how much uncertainty remains rather than treating the first promising number as a forecast.
Separate the placements you can actually change
A blended EPC may hide useful differences between an email placement, a comparison page and a paid advertisement. Use permitted placement identifiers so you can connect a result to an editorial or buying decision. Avoid sending names, email addresses or other personal details in those identifiers.
Everflow explains how click-level transaction identifiers support attribution in its S2S tracking documentation. That technical association helps identify the referral; it does not by itself explain why the visitor converted. For that, compare the offer, creative, audience and landing page together.
Compare revenue with the cost of reaching the audience
Suppose an illustrative paid test has approved EPC of 0.16 and media cost per click of 0.11. The difference is 0.05 per click before other expenses. Creative production, tools, staff time and reversals can reduce the eventual margin. If you use unpaid search traffic, acquisition still consumes editorial and maintenance resources even though there is no media invoice for each click.
Do not raise a budget solely because revenue per click exceeds ad cost for a short period. Check whether the approval rate is stable and whether the next volume increment is likely to come from the same type of placement. The campaign scaling guide covers that review in more detail.
Use EPC to ask a better question
When EPC falls, investigate what changed: source mix, device mix, creative, availability, tracking or validation. Work from a dated change log and compare affected segments. Lower earnings can reflect a product issue or a tracking issue as well as audience fit.
EPC is most useful when it leads to an action you can test. You might remove a misleading promise, repair a broken destination or compare two clearly described offers. Read how to improve campaign quality for a practical way to connect the metric with the visitor's experience. None of the numerical examples here are a promised return or historical network result.
Put the principle into practice.
Discuss the campaign-specific requirements with MMN before committing traffic or budget.
Talk to Partnerships