Reach curves in a fixed-location channel
In broadcast or online media, reach is built by buying more impressions against a population that is already reachable. In a fixed-location channel, reach is bounded by who walks into the stores you bought, and additional weight in those stores adds frequency rather than reach.
That distinction changes how a plan should be built and how a reach estimate should be read.
What actually drives reach here?
The number of distinct shoppers visiting the stores in the buy over the flight. Adding stores adds reach; adding plays in the same stores adds frequency.
That is the central planning fact and it is the opposite of the intuition brought from impression-based media, where more weight generally means more reach.
How does the curve behave?
Steeply at first and then flattening. In the early days a large share of the stores' regular visitors are reached; after that, the additional reach comes from progressively less frequent visitors.
The flattening point depends on the visit frequency distribution. A store whose traffic is mostly daily regulars saturates quickly; one with more transient traffic keeps accumulating.
Where does the estimate come from?
A model combining store traffic, visit frequency and duplication between stores. Each input has uncertainty and the duplication term is the hardest, because shoppers visit several stores.
Ask how duplication is handled. An estimate that treats every store's visitors as distinct overstates reach substantially in a dense urban area.
What is unduplicated reach worth here?
A good deal, where the channel reaches people other media do not. The argument for the channel is strongest when it is framed as incremental reach against a plan rather than as reach in isolation.
That framing also makes the claim checkable, because incremental reach against a defined plan is a measurable quantity rather than an assertion.
How does store selection change the curve?
More than weight does. A buy concentrated in a small number of high-traffic stores reaches fewer people more often; one spread across more stores reaches more people less often.
Deciding which shape suits the objective is the main planning decision in this channel, and it is frequently made implicitly by a budget rather than deliberately.
How should a reach estimate be presented?
With its inputs and its duplication assumption. A reach percentage against an undefined population is not a claim that can be checked, and a network that states its population and method is inviting the right scrutiny.
What can be measured rather than modelled?
Plays, uptime, and store traffic where transaction data exists. Reach itself is modelled in this channel as in most others, and separating the measured from the modelled in reporting is what makes the estimate credible.
Frequently asked questions
Can reach be measured directly?
Not without identifying shoppers, which most of this channel does not do. What can be measured is store traffic and delivery, from which reach is estimated, and honest reporting distinguishes the two.
Does adding plays increase reach?
Only slightly, by catching shoppers whose visit did not coincide with a play. The main effect of additional plays is frequency, which is a different objective.
How is duplication estimated?
From assumptions about shopping behaviour across stores, sometimes informed by transaction or location data. It is the weakest input in most reach models and it is worth asking about specifically.
What reach is achievable in this channel?
It depends entirely on how many stores are bought and how much of the population visits them. A percentage quoted without a store count and a population definition is not interpretable.
Should I optimise for reach or frequency?
For awareness, reach, which means more stores. For activation among regulars, frequency, which means more weight in fewer stores. The objective should choose.
Does the channel suit a national reach objective?
It suits an incremental reach objective against an existing plan better than a standalone national reach one, because its audience is defined by store footfall rather than by broadcast coverage.
How does this compare to other out-of-home?
Roadside formats reach passing traffic with very short exposure; store formats reach shoppers with longer exposure and a purchase context. Measurement practice across out-of-home is discussed by bodies such as Geopath.
What is the most common estimation error?
Ignoring duplication between nearby stores. In a dense area it inflates reach substantially and it is invisible in the output.
What is the single planning lever?
The store list. Reach in this channel is bought by store, and everything else adjusts frequency. Reach as a concept transfers; the mechanism does not.
### How should a reach estimate be used in a plan?
As one input alongside the store count and the distribution overlap. A reach figure that cannot be reconciled to a number of stores and a population is not doing any work.
Where two networks quote different reach for similar store counts, the difference is in the model rather than in the estate, and that is the thing to interrogate.
Does reach in this channel transfer to a national figure?
Only against a defined population. A store network's reach is best expressed against the population of the areas it covers rather than against a national base, because the latter makes the figure look small and tells nobody anything useful.
### What should be reported about reach after a flight?
The store count actually delivered, the days live, and the reach estimate recomputed on delivered rather than planned inventory. A reach figure from the plan describes an intention.
Where stores dropped out, the delivered reach is lower than planned and the difference should be visible rather than absorbed.
Can two networks' reach figures be added?
Not without deduplication, and in overlapping geographies the overlap can be large. Two store networks in the same neighbourhoods share customers.
Does frequency appear in the same model?
It should, because reach and frequency come from the same exposure distribution. A reach figure quoted without any frequency context describes half the delivery.
Considering NRS Digital Media? Start at nrsdigitalmedia.com, read the media deck, or contact the team.