Two shoppers wearing winter clothing inside a store.

Path to Purchase: planning retail media across store, site and offsite

The path to purchase is often drawn as a clean funnel. Shoppers rarely behave that neatly.

A woman with store displays in the background.

A shopper may notice a product in social media, compare it on a retailer site, visit a store for another reason, see an in-store message, check a review on their phone and buy a substitute because the first product is unavailable. Another shopper may repeat the same grocery order with almost no visible exploration.

Path-to-purchase planning should not force both people into one sequence. Its job is to decide which touchpoint can change which decision, how one channel hands off to the next and what evidence will show progress.

A three-layer planning framework

The three layers below are this article's planning framework, not an industry standard. They are useful only if the campaign makes each state and transition observable.

The shopper decision

Name the decision before the channel. Is the campaign trying to create category consideration, win a brand choice, bring the shopper to a retailer, remove product confusion, complete a basket or encourage repeat? A shopping mission describes the trip purpose; a shopping moment describes the observable state in which media may act.

The touchpoint job

Assign one main job to each media environment. Offsite may create or refresh memory. Onsite may capture active category intent. In-store may interrupt habit, resolve choice near the product or remind the shopper while the basket remains open.

The handoff

Define what should happen next. An offsite impression might lead to a retailer product page, a saved offer, a store visit or simply better brand recognition at shelf. An onsite product view might hand off to store availability or pickup. An in-store exposure might lead directly to purchase or later repeat.

A path is useful when these handoffs are explicit.

Why the path is not one universal funnel

Research on omnichannel buying journeys organizes touchpoints across pre-purchase, purchase and post-purchase phases while emphasizing that consumers move among channels.[1] Other omnichannel work treats the journey and channel choice as an integrated continuum rather than separate online and offline systems.[2]

For retail media, that has four implications.

First, the same channel can perform different jobs. Offsite media can create awareness or reactivate a known category buyer. Onsite media can support discovery or close a high-intent purchase.

Second, one shopper can move backward. A product detail page can create new questions and send the shopper back to comparison.

Third, stores are not merely the last step. They can create discovery, trial and memory for a later ecommerce purchase.

Fourth, not every journey needs many contacts. Routine purchases may require fewer interventions. More touchpoints can add cost and irritation without adding value.

Give each environment a clear role

The channel uses below are planning options to test. They are not settled rules for every category.

Offsite

Use offsite media when the campaign needs reach before active retailer browsing, when the brand must build memory, or when retailer audiences can identify a relevant group outside owned properties.

Define the intended handoff. If the next useful action is a retailer visit, the creative and landing experience should make that easy. If the job is memory, do not judge the channel only by immediate clicks.

Onsite

Onsite media sits close to active shopping behavior. It can help a product enter consideration, explain a difference, win a category search or complete a basket.

Proximity to purchase does not guarantee incrementality. Sponsored placements can harvest demand from shoppers who would have bought anyway. Measure both conversion and what changed versus a credible comparison.

In-store

In-store media can reach shoppers near products and while a physical basket is open. It can support navigation, reminder, substitution, cross-sell and trade-up.

The channel depends on store traffic, zone traffic, availability and the chance that the shopper actually encountered the unit. A store-level sales rise is not proof that everyone in the store saw the message.

A worked plan: a new household cleaner

Consider a hypothetical launch with two valid routes: an offsite-to-onsite-to-store path and a direct onsite-to-purchase path. The campaign wants to establish a reason to consider the product and create first purchase.

Offsite prospecting

Eligible audience: Category buyers with no recent purchase of the new product.
Message task: Explain one defensible use case.
Next state: Retailer product or category page.
Evidence: Qualified landing visit or exposed holdout assignment.
Suppression rule: Cap frequency and suppress recent trialists.
Fallback: If visits do not rise, test reach, creative and landing friction before adding channels.

Onsite category browse

Eligible audience: Shoppers viewing or searching the relevant category.
Message task: Make the difference easy to evaluate.
Next state: Product detail, basket addition or store check.
Evidence: Detail view, add-to-basket and conversion.
Suppression rule: Suppress after purchase and avoid repeating answered claims.
Fallback: If views rise but baskets do not, test price, proof and content.

Store-ready state

Eligible audience: Shoppers near a stocked store or checking local availability.
Message task: Confirm the benefit and where to find the item.
Next state: Target-zone visit and purchase.
Evidence: Observed target-zone visit evidence plus store-SKU outcome.
Suppression rule: Do not serve when product availability is stale or false.
Fallback: Switch to an approved category message or stop delivery.

Post-purchase

Eligible audience: Verified first-time buyers where permitted.
Message task: Support correct use and observe repeat.
Next state: Repeat or category switching.
Evidence: Repeat within the category's expected cycle.
Suppression rule: Suppress redundant acquisition media.
Fallback: If identity coverage is weak, report aggregate repeat rather than inventing a person-level chain.

Path A moves from offsite exposure to retailer research, then to a stocked store. Path B starts with active onsite browsing and can end in ecommerce purchase without an offsite or physical-store contact. A shopper who buys directly onsite should not be forced into the longer path for reporting convenience.

The measurement plan keeps reach, qualified visits, onsite behavior, in-store opportunity, first purchase and repeat separate. The combined campaign still needs a closed-loop measurement or experimental design that states what can and cannot be linked.

Define transitions before buying media

A transition needs four elements.

Entry condition: what shopper state makes the touchpoint eligible?
Message job: what question or barrier should the touchpoint address?
Next state: what should the shopper know or do afterward?
Evidence: what observable event indicates that the handoff occurred?

The evidence may be imperfect. Cross-device, cash and store purchases can break identity chains. The answer is not to invent certainty. Use aggregated or experimental designs where person-level linkage is unavailable.

IAB's commerce-media definition covers retailer assets, first-party data and activity beyond retailer-owned properties.[3] IAB Europe's updated standards provide a common measurement funnel and definitions across commerce media.[4] Those frameworks help create shared language, but the campaign still needs its own channel jobs and data limits.

Budget by marginal job, not channel prestige

Treat this section as a planning heuristic, not a budget formula. Before moving spend, ask what an additional unit is expected to do and what evidence supports that expectation.

If offsite reach has saturated the eligible audience, more spend may only add frequency. If onsite search already captures existing demand, the next budget may belong in discovery or an experiment. If an in-store unit sits in low-traffic zones, creative changes will not solve the placement problem.

Budget should move only after the team checks saturation or frequency, tests whether onsite media is harvesting existing demand, examines zone traffic and availability, and reviews an incrementality result. The next investment should address a demonstrated constraint rather than channel prestige.

Common mistakes

Drawing a generic funnel and adding channels afterward. The diagram describes stages but does not assign work.

Giving every channel the same conversion KPI. Channels contribute at different distances from purchase.

Counting contacts without sequence. Five impressions do not reveal whether the shopper moved forward.

Treating last click as the whole path. The final touchpoint may collect demand created elsewhere.

Ignoring product truth and availability. A clean handoff fails when price, content or local inventory is wrong.

Assuming more integration means more personalization. Integration can also mean consistent product facts, suppression and measurement definitions.

Planning checklist

Before launch, confirm:

the shopper and commercial decision are stated in one sentence;
every environment has one primary job;
each handoff has an entry condition and next state;
creative and product facts remain consistent across channels;
store and product availability are included;
identity and coverage gaps are documented;
frequency and suppression operate across reachable channels;
channel metrics match channel jobs;
the total plan has an incrementality design;
the next budget decision is defined for strong, weak or mixed results.

FAQ

Is the path to purchase the same as a funnel?

No. A funnel summarizes decision stages. A path maps the touchpoints and handoffs that may move a shopper among those stages.

Should every campaign use offsite, onsite and in-store media?

No. Use the environments needed for the decision. A routine replenishment campaign may need only one or two.

Which channel is closest to purchase?

Onsite and in-store often sit near a transaction, but closeness varies by category, fulfillment and shopper behavior. Proximity does not prove causality.

How should cross-channel performance be reported?

Report delivery and outcome metrics by channel job, then evaluate the combined plan against a counterfactual where possible.

Who should coordinate the path?

One accountable planning lead should coordinate brand, retailer, ecommerce, shopper and measurement teams. Channel owners can execute parts, but the handoffs need one design.

Bottom line

Path-to-purchase planning is the design of channel jobs and handoffs around a real shopper decision.

Start with the barrier. Give offsite, onsite and in-store media distinct work. Measure each environment against that work, and judge the combined plan by the behavior it changed.

Sources

[1] Furquim et al., The Different Phases of the Omnichannel Consumer Buying Journey: A Systematic Literature Review and Future Research Directions.

[2] Hübner et al., The omnichannel continuum: Integrating online and offline channels along the customer journey.

[3] IAB, Defining Commerce Media and Its Ecosystem.

[4] IAB Europe, Commerce Media Measurement Standards V2.1.

Ready to see how this works in practice?

Footprints AI helps brands and retailers measure what matters. See our customer success stories or get in touch to discuss your retail media strategy.