Building a test roadmap for a new channel
A brand new to in-store media usually runs one large campaign, measures it ambiguously, and concludes something general. A staged roadmap answers more, costs less in total, and produces evidence that survives a change of personnel.
The sequence matters more than the size of any individual step.
What is the first step?
A feasibility check rather than a campaign. Does the category sell in these stores, is the product distributed, can the network provide store-level data, and can a holdout be arranged. Four questions answered before any spend.
Several brands stop here, correctly, because one of the four answers rules the channel out.
What is the second step?
A small powered test with a clear activation objective, a holdout, and a pre-agreed measure. Small in stores, not small in design: an underpowered first test produces an ambiguous result that ends the programme for the wrong reason.
The objective should be the one most likely to show an effect, because the purpose of the first test is to establish whether the channel does anything.
What is the third step?
Creative or weight variation, once the channel has shown an effect. Testing creative before establishing the channel works confuses two questions, and a flat creative test in a channel that does not work teaches nothing about creative.
What is the fourth?
Scaling with continuous measurement rather than a one-off. A programme that measures once and then runs on faith is a programme that will not detect a decline until it is large.
How long should the roadmap take?
Longer than a single campaign and the steps are individually short. A year covering feasibility, a first test, a creative test and an early scale-up is a reasonable shape for most brands.
Compressing it into a quarter produces the single-large-campaign outcome the roadmap exists to avoid.
What should be documented?
Each step's design, result and decision, in one place. That record is what allows the programme to survive a change of brand manager, which is the most common way institutional learning about a channel is lost.
What should the exit criteria be?
Agreed at the start: what result at each stage would stop the programme. Without them, a disappointing result becomes a negotiation rather than a decision.
Frequently asked questions
Is a staged approach slower?
It reaches a confident answer faster than one large ambiguous campaign followed by an argument. The elapsed time is longer and the time to a usable conclusion is shorter.
What if leadership wants a big test?
Size the big test properly and treat it as step two with step one done first. The feasibility check takes days and it prevents the most expensive failure mode.
How much budget should the first test take?
Enough for power in the stores required, which is a calculation rather than a proportion. An underfunded first test is worse than no test.
Should the network be involved in designing it?
In the delivery, yes, and the measurement design benefits from independence. A test designed entirely by the seller is a weaker basis for a decision.
What if the first test is flat?
Diagnose before concluding: compliance, power, measure, creative. A flat first test with an identified cause is a step forward; one taken as a verdict ends a programme prematurely.
Can steps be combined?
Feasibility and the first test can overlap. Combining the effect test and the creative test in one underpowered design is the combination to avoid.
Who should own the roadmap?
One named person, with the document. Roadmaps owned by a rotating group become a set of unconnected campaigns.
How does this apply to an agency relationship?
The roadmap should be shared, because an agency planning against a staged programme plans differently from one planning a single campaign.
What is the highest-value single step?
The feasibility check. It is nearly free and it prevents the most common expensive mistake. General out-of-home planning practice is discussed by bodies such as OAAA.
### How should the roadmap be documented?
As one page per step: the question, the design, the result and the decision. Four lines per step, updated as each completes, held by a named owner.
That document is what survives a change of brand manager, which is the most common way a programme loses its history and repeats a test it already ran.
It also makes the internal case for continuation considerably easier, because the argument is a sequence of results rather than a recollection.
How much of the roadmap should be agreed up front?
The sequence and the exit criteria; the design of later steps depends on earlier results. Committing to designs for every step in advance is planning for a world where nothing is learned.
What if a step produces an ambiguous result?
Diagnose before proceeding, because carrying an unresolved ambiguity into the next step compounds it. An ambiguous result with a known cause is a step forward; one taken at face value is not.
How should the roadmap survive a personnel change?
Through the document rather than the person. One page per step, held somewhere findable, is what allows a successor to continue rather than restart.
Should the network see it?
Sharing the sequence helps, because a network that understands the programme's stages plans its support differently from one selling a single flight.
Considering NRS Digital Media? Start at nrsdigitalmedia.com, read the media deck, or contact the team.