Comparing costs across channels that are not comparable
Media plans compare channels on cost, and the comparison usually uses a common unit whose definition differs between them. The result looks like arithmetic and is closer to an assertion.
Knowing which comparisons are defensible is more useful than refusing to compare at all.
Why do impression definitions differ?
Because each medium measures what it can. An online impression is a served ad; an out-of-home impression is a modelled opportunity to see; a broadcast impression comes from a panel-based audience estimate.
Each is a reasonable answer within its own framework and none of them mean the same thing.
What happens when they are compared directly?
The medium with the loosest definition looks cheapest. That is a property of the definitions rather than of the media, and a plan built on it systematically over-allocates to whichever channel counts most generously.
That failure is common and it is usually invisible, because the comparison is presented as a table of numbers.
What can be compared?
Cost per outcome, where an outcome is measured the same way across channels. Sales lift per pound spent, measured by comparable designs, is a genuine comparison.
It is harder to produce and it answers the question a budget decision actually asks.
What about cost per store per day?
For comparing two in-store networks, it is unambiguous and useful. It does not transfer across channels and it does not adjust for traffic, so it is a within-channel unit.
Within a channel, unambiguous units beat sophisticated ones that rest on differing assumptions.
How should a cross-channel table be presented?
With the definition under each column. A table of CPMs with no definitions is a table that cannot be checked, and adding one line per channel makes the comparison honest without making it longer.
Where the definitions are too different to compare, saying so is more useful than producing a number.
What is the role of a common currency?
It exists precisely to make comparison possible within a medium, and cross-medium currencies are much harder. Out-of-home measurement practice is developed by bodies such as Geopath, and digital standards by bodies such as the IAB.
Aligning to a currency is a signal about a network's seriousness, and it does not make its numbers comparable to a different medium's currency.
What should a buyer do practically?
Compare within channel on cost, and across channel on outcome where possible. Where outcome comparison is not possible, treat cross-channel cost tables as directional and say so.
Frequently asked questions
Is any cross-channel CPM comparison valid?
Between channels using the same measurement framework, reasonably. Across frameworks it compares definitions, and the differences are usually larger than the cost differences being examined.
What is the best cross-channel unit?
Cost per measured outcome, where the outcome is defined identically. It requires measurement in each channel and it is the only comparison that settles a budget question.
Should I discount a channel with a loose definition?
Rather than discounting arbitrarily, ask for the definition and make the comparison on outcome instead. Arbitrary adjustments introduce a different error.
Does viewability help?
It tightens a digital definition and does not make it equivalent to a physical one. The categories remain different.
How do agencies handle this?
Practice varies and the better approach states definitions explicitly. An agency that presents a cross-channel cost table without them is presenting a table rather than an analysis.
Can a network claim a lower CPM than digital?
It can compute one, and whether the claim means anything depends on the definitions. A network making the comparison without stating them is making a claim that will not survive scrutiny.
What about reach comparisons?
They face the same problem plus the duplication issue. Reach across channels should be deduplicated with a stated method or presented separately.
Is there a simple honest framing?
Compare cost within a channel, and compare channels on what they achieved. That is defensible and it is usually available.
What should a buyer insist on?
A definition under every number. It costs a line and it prevents most of the errors in this area.
### What should a planner do when asked for a single comparison?
Give the comparison with its caveat in the same sentence, or give an outcome-based comparison instead. A table of CPMs presented without definitions will be used as though the definitions matched.
Where an outcome comparison is genuinely unavailable, saying that the channels cannot be compared on cost is a legitimate answer and it protects the plan.
Is cost per store per day useful across channels?
No, it is a within-channel unit. Its value is that it is unambiguous for comparing two store networks, which is exactly the comparison a CPM makes hardest.
### What should a media team standardise internally?
A definitions sheet: what an impression means in each channel the business buys, and how each channel's cost is computed. One page, maintained, referenced in every plan.
That document ends most cross-channel cost arguments, because the disagreement is usually definitional rather than factual.
Should the sheet be shared with suppliers?
It helps, because a supplier who knows how their figures will be read presents them more carefully. It also signals that claims will be checked.
Does this slow planning down?
It speeds it up after the first time, because the recurring argument about whose number is bigger has already been resolved into a comparison of definitions.
Considering NRS Digital Media? Start at nrsdigitalmedia.com, read the media deck, or contact the team.