Benchmarking a new channel: fair comparisons for DOOH

It's planning season, and a media team is deciding whether screens in independent stores earn a line on next year's plan. Someone asks the natural question: how does this channel benchmark against the others? The tempting shortcut is a comparison table borrowed from a deck. The honest answer is that fair benchmarking means comparing channels on the same outcome, and for a new channel, building your own benchmarks beats borrowing anyone else's.

That answer takes more work. It's also the only version that survives scrutiny.

Why do cross-channel benchmark tables mislead?

Because the metrics aren't measuring the same thing. A social channel reports clicks and video views. Digital out-of-home reports modeled impressions built on the opportunity to see, the standard maintained by measurement bodies like Geopath. Lining those up in one column implies an equivalence that doesn't exist; a click and a modeled exposure at a register are different events with different meanings.

Context differs too. An impression served to someone scrolling in bed and an impression at a checkout counter, seconds from the shelf, occupy different moments in the path to purchase. A table that prices them identically isn't neutral; it's quietly biased toward whichever channel's metric is cheapest to manufacture.

What does a fair comparison look like?

Compare channels on the outcome you actually want, which for most brands in this channel is sales. Cost against incremental sales, measured the same way for each channel under comparison, is the fairest common denominator available. Reach-based comparisons can also be fair, provided each channel's reach is unduplicated and honestly derived; resources from the Out of Home Advertising Association of America are a reasonable starting point for understanding how the OOH side of that ledger is put together.

Fairness also means comparing jobs, not just numbers. A channel reaching shoppers at the point of decision should be judged on conversion-adjacent outcomes; a broad-awareness channel on audience building. Asking every channel to win every metric guarantees a muddled plan.

How do you build internal benchmarks for a new channel?

Run a structured first flight and let it set the bar. Define test and control stores, flight length, and target SKUs before launch, then record the results as your baseline: what delivery cost, what the sales comparison showed, how the read varied by region and store type. Your second flight benchmarks against your first. By the third, you have a channel history that no external table could have given you.

Keep the write-up of each flight short and standardized: same tables, same definitions, same baselines. Benchmarks only stay comparable if the measurements behind them hold still from flight to flight.

This is more practical on a network built for it. Campaigns on NRS Digital Media can be scoped by zip code, retail channel, and SKU with no minimums, so a benchmark-building test doesn't require a leap of faith, and the transaction data behind NRS Insights gives the outcome side of the benchmark a solid footing.

When should you distrust a borrowed benchmark?

Whenever it arrives without a method. A benchmark is a measurement plus a context, and stripped of the context, the number is decoration. Whose campaigns produced it? Which categories, which stores, which baseline, which year? If those answers aren't attached, the number can't transfer to your situation, and adopting it just outsources your judgment to an unknown methodology.

A channel confident in its measurement will happily show you how a number was made. Asking costs nothing and reveals plenty, and that willingness is itself a useful benchmark.