What to ask before signing a retail media contract
A retail media contract commits a budget to inventory a buyer generally cannot inspect. The questions that make it verifiable are straightforward and they are considerably easier to ask before signature than after.
None of them are adversarial. A network confident in its operation answers them readily.
What should be asked about inventory?
Which stores, how many screens, where the screens are positioned, loop length and commercial slot count. Five answers, and together they define what is being bought.
A network that cannot provide a store list is asking for a commitment to an unspecified estate.
What should be asked about delivery?
How delivery is measured, whether reporting is store-level, what uptime is and how it is measured, and what happens when delivery falls short. The uptime method matters: device heartbeat and verified playout are different measurements.
What should be asked about measurement?
What outcome data the network holds, whether store-level analysis is possible, whether holdouts can be arranged, and who performs the analysis. Those determine whether the campaign can be evaluated at all.
A network with transaction data at store level can support measurement that an impressions-only network cannot, and that difference is worth establishing before the buy rather than after.
What should be asked about the estate?
How stores are recruited, what the churn rate is, how many stores left in the last year, and what happens to a campaign when a store leaves. Churn is a normal feature of an independent network and its rate is informative.
What should be asked about content?
The content policy, category restrictions, the approval process, competitor separation, and how quickly content can be pulled. Five answers, and they constitute the brand safety position in this medium.
What should be asked commercially?
Pricing basis, minimum commitment, cancellation, make-good mechanism and what happens if the store list changes materially. The last is specific to this channel and it is frequently absent from standard terms.
What does a poor set of answers look like?
Vagueness about the store list, uptime measured only by device status, no store-level reporting, and no defined make-good. Any one of those is workable; all four together describe a channel that cannot be verified.
Frequently asked questions
Is it reasonable to ask for a store list?
Entirely, and a buyer is committing budget to specific premises. Confidentiality arrangements can address commercial sensitivity without withholding the list altogether.
What if the network will not disclose uptime?
That is a significant gap, because every impression figure rests on it. A network that measures uptime and declines to share it is making a choice a buyer should weigh.
Should measurement be contractual?
The data access and the design should be, because a measurement agreed informally tends not to survive a disappointing result. Putting it in the contract makes it a commitment.
How long should a first contract be?
Short enough to evaluate and long enough to run a proper test, which usually means a single flight with a measurement design attached. Multi-year commitments before any evidence are a poor first step.
What about exclusivity?
Ask whether category exclusivity is available, at what level and for how long. It can be worth more than a rate reduction.
Who owns the campaign data?
Establish it in writing, including delivery logs and any analysis outputs. Ambiguity here surfaces when the relationship ends.
Can I speak to existing advertisers?
Asking is reasonable and networks vary in what they can offer. Case studies published by a network are a starting point rather than independent evidence; the case studies a network publishes describe its own work.
What about the store operator relationship?
Ask how operators are compensated and supported, because a network with unhappy operators will have screens switched off. It is a delivery question in a commercial disguise.
What is the single most revealing question?
What does your network not cover. A provider that names its own gaps is demonstrating the understanding that makes everything else credible.
### How should the answers be recorded?
In writing, attached to the contract, so that the commitments survive a change of personnel on either side. Verbal assurances about reporting and measurement have a short life.
It also makes the second negotiation easier, because the baseline is documented rather than reconstructed from memory.
Where an answer is unsatisfactory, record that too. A known gap that has been priced into the decision is a different position from one discovered after the buy.
Is it reasonable to ask all of these?
Yes, and a network that finds them unreasonable is telling you something. They are the ordinary questions a buyer asks about inventory they cannot inspect.
What if time is short?
Ask about the store list, delivery measurement and the make-good mechanism. Those three cover the largest risks and they can be answered in a short conversation. Industry practice on measurement is discussed by bodies such as Geopath.
Should the answers be revisited at renewal?
Yes, because estates, methods and policies change between contracts. A renewal negotiated on the position established two years earlier is negotiating about a network that no longer exists in that form.
What if the network is new?
A short track record is not disqualifying and it changes what evidence is available. Weight the operational answers more heavily, since a new network has methods and practices even where it has no case studies.
Should the questions be asked in writing?
Yes, because the answers become part of the record. A verbal assurance about reporting granularity has a short life once personnel change.
What if answers arrive late?
Delay the commitment rather than proceeding on assurances. The questions are quick to answer for a network that knows its own operation.
Considering NRS Digital Media? Start at nrsdigitalmedia.com, read the media deck, or contact the team.