Digital drive-to-store spend jumped 33% in 2025 to 2.1 billion euros (Alliance Digitale). But the money goes to search and brand websites. The increment is elsewhere.
On 20 October, Alliance Digitale presented the second edition of its drive-to-store barometer, produced with France Pub and reported by Minted. The figure doing the rounds: 2.1 billion euros invested in digital drive-to-store in 2025, up 33.4% year on year, while digital as a whole grew only 8.2%. Everyone reads it the same way: store traffic is now won online. I read something else. When you look at where the money goes, it goes to search and to brand websites. That is, to people who had already decided to come. My view: a good part of those 2.1 billion does not create traffic, it collects it. And the increment, the kind that fills a store that has just opened, sits in the part of the plan this barometer shows shrinking.
Digital drive-to-store: what the barometer says
Orders of magnitude first. Drive-to-store across all media weighs 7.6 billion euros in 2025, up 21.1%, in a French advertising market of 35.2 billion that fell 1.3%. Digital accounts for 28.2% of those 7.6 billion and becomes the top lever, gaining 2.7 points, just ahead of promotion at 28.1%. Traditional media take 19.8% of the total: television 7.8%, down 1.7 points, out-of-home 4.5%, press 3.8%, down 1.1 points, radio 3.2%, cinema 0.4%.
Virginie Sablé, managing director of France Pub, gives the overall reading: in 2024, image accounted for 66% of advertisers' investment and traffic for 34%; the rebalancing heads towards 60% for branding and 40% for drive-to-store. In other words, advertisers are moving money from awareness to immediate sales. I understand the reflex in a year when the market is down. I am not sure it is the right one, and I will come back to it.
One honest point on the numbers: the first edition, published in September 2025 on 2024 data, counted 6.3 billion of drive-to-store, including 1.6 billion in digital. Going to 7.6 and 2.1 billion is an increase, but it is also the second edition of a young methodology, and the press coverage details neither the sample nor the scope. I take the trend, not the decimal.
Where the money goes: search and brand websites
This is the part of the barometer that interests me. Within digital drive-to-store, paid media, sponsored links, display, banners, make up 43%. The rest, 57%, is owned media: the brand's websites and apps 17%, its social networks 14%, CRM 10%, content marketing 10%, influence 6%. Search alone represents 23% of all drive-to-store, 33% among retailers, and it is the first lever in the mix of national campaigns, at 30%, up 3 points. In local campaigns, which account for 41% of the amounts, owned media weighs 80%.
Read that split with a media buyer's eye. Search captures an intent that already exists: someone types "organic store open Sunday" and you pay to sit above the map. The website, the app, the CRM talk to people who already know you. The brand's social networks reach your followers. None of that is useless; I buy it for my clients. But none of these levers brings into a store someone who had no idea of going there. They convert demand. They do not manufacture it.
For an established network, with known stores and loyal customers, collecting demand is good business. For a network that is opening, a D2C brand going physical, a retailer that wants to gain share in a city where it is number three, it is not enough. The demand you need, nobody types into Google. You have to go and create it in the catchment area, and that is the job of out-of-home, geotargeted mobile, digital audio and, at larger scale, television. Exactly the lines the barometer shows falling within drive-to-store. The market is strengthening the net and weakening the fishing.
The real issue: nobody can measure the increment
In the same article, Minted relays a Bonial survey of 113 media buyers. 81% of advertisers consider that reach alone does not allow them to evaluate a campaign's effectiveness. In-store traffic is the most used KPI, by 69% of them, ahead of incremental traffic, 48%, in-store revenue, 37%, and in-store sales, 36%. Incremental sales are tracked by only 28%, incremental revenue by 23%. And two figures that say it all: 52% say they cannot allocate their investment from visit and sales KPIs, and 82% see large variation depending on the attribution method.
When 82% of buyers see the result change with the method, the method is the result. That is why search wins: it is the easiest lever to attribute, not the most effective. A click on "directions" after a brand search counts itself. A person who saw three screens in the street, heard a spot in the car and pushed the door on Saturday does not get counted, unless you planned to count them. So the market invests where measurement is comfortable, and calls it performance. I described the same mechanism for retail media: the platform that sells the space is also the one that writes the report.
My rule for drive-to-store
One indicator that does not come from a sales house: incremental traffic, measured with exposed areas and comparable control areas, store by store. Search and owned media last in the plan, to collect what the rest has created, never first. Isochrone targeting around each point of sale, not a region. And a reach budget, out-of-home, mobile, audio, that does not shrink because search is easier to justify in a board meeting.
How I build a drive-to-store plan for a network
I start with the map, not the channel. Every store has a real catchment area, in minutes, not kilometres. For Comptoirs de la Bio, we replaced the paper leaflet with an isochrone set-up around each store: digital screens within ten minutes, mobile follow-up on people who had actually been in the area. Budget halved compared with the leaflet, and 100% of the set-up tracked. I detailed the mechanics in hyper-local: the secret to filling a point of sale. The leaflet, by the way, is precisely that "promotion" at 28.1% that digital has just overtaken.
I create demand with local reach. Geotargeting now lets you buy digital out-of-home screen by screen, at 4 to 9 euros CPM, and mobile within 100 metres of a point of sale: for Deblock, encircling metro stations and geotargeting mobile at 100 metres produced more than 30 million impressions. Digital audio can be targeted by city for one or two euros more on the CPM. Three different contacts, in the same area, for three weeks, and a brand becomes familiar to people who were not looking for it. That is what we are working on right now with Banette, a network of artisan bakeries: social relay for the network, but above all local pressure where the bakeries are, not a national plan divided by the number of outlets.
I put search at the end of the chain. Brand, "near me", opening hours, directions: local search captures what reach has triggered. I keep it, I watch it, and I look at one thing first: brand search volume in exposed areas versus control areas. If reach works, search rises on its own. If search rises without reach, the demand already existed, and that is proof you were collecting.
I measure before launching. Exposed areas and control areas, comparable stores, traffic measured through passive location data and receipts, baseline recorded before the first screen. I described the protocol in how to measure point-of-sale traffic from a billboard. It is not sophisticated. It is just decided beforehand, not rebuilt afterwards with whichever method flatters.
Two signals this week confirm that local out-of-home remains an asset that counts. Stratégies reported on 19 October that JCDecaux had opened discussions on a possible acquisition of Giraudy, 50 million euros of revenue in 2025, 664 of whose 8 m² panels in the Paris region have just moved to Cityz Media under a sub-sales agreement; the paper's analysis is that the market leader may have an interest in absorbing it to raise its prices. And Values.media has connected its tools to Displayce's AI agents through the MCP protocol, to cross INSEE statistics, weather and DOOH inventory from the plan's design stage. Local out-of-home is becoming more expensive to own and easier to plan. For a buyer, this is the time to be good at it.
What I tell founders
The barometer says the market is moving money from image to traffic, and within traffic, from media to search and owned. That is rational for a CFO who wants a clean attribution line. It is dangerous for a brand that needs new customers, because you end up paying more and more to collect a demand you no longer feed. If you are opening points of sale, or if your network has plateaued, the question is not "how much do I put in search", it is "how many people within a ten-minute radius heard of me this month, and what did it change in traffic". We can run that diagnosis store by store during a free audit. And for a first order of magnitude of local reach, the reach and frequency calculator gives a baseline in two minutes.
