Drako, a Montreal-based programmatic outcomes company, said on September 10, 2026 that campaigns using its custom or syndicated audiences in any demand-side platform will now carry Foot Traffic Attribution at no additional charge, removing a measurement line that the company says is frequently billed at roughly 50 cents per thousand impressions.

In Short

A Canadian advertising data company now includes store visit reports, which show whether people who saw an ad later walked into a shop, at no extra cost when advertisers use its audience lists. Those reports are often sold as a separate add-on, so the change touches agencies and brands running campaigns for shops, restaurants, car dealers and tourism businesses. The reports come only with the company's own audience data or managed campaigns, and its tests of whether ads actually caused extra visits remain separate products.

What changed on September 10

The offer, as Drako describes it, runs through two doors. Agencies that manage their own media gain access when they activate Drako custom or syndicated audiences inside whatever buying platform they already use. Managed-service clients receive the measurement as part of campaign execution handled by Drako's AdOps team. According to Drako, the offering is available immediately for eligible campaigns, although the company did not define eligibility beyond those two conditions, name any buying platforms, list markets or set a minimum spend.

Implementation is described in one sentence. The no-charge offering can be deployed through an agency's existing DSP using a conversion pixel, with direct platform integrations available in what the company calls "certain environments." Which environments those are was not specified.

Foot Traffic Attribution, in Drako's definition, connects devices exposed to advertising with subsequent visits that a measurement provider can observe at physical locations. European practitioners tend to call the same practice footfall attribution. According to Drako, it has become a common component of retail, restaurant, automotive, tourism and other drive-to-store campaigns.

Drako was founded in 2017. The company describes itself as combining proprietary audience data, hands-on campaign management and real-world measurement tools, the latter including foot traffic attribution, geolift and market share lift analyses.

The price of a visit signal

The commercial argument rests on a comparison. Online conversion tracking, according to Drako, is generally treated as a basic campaign utility. Visit attribution is not. It is frequently sold as a paid add-on, and the release puts the typical price at about 50 cents per thousand impressions.

That figure is Drako's characterization of the market rather than a published rate from a named vendor, and the release does not identify whose pricing it reflects. Taken at face value, the arithmetic is simple. A charge of that size adds $500 for every million impressions served. A drive-to-store campaign delivering 20 million impressions would carry $10,000 in attribution fees; one delivering 100 million would carry $50,000.

Its weight relative to media cost depends heavily on the inventory. DataBeat's figures for April 2026 put average United States mobile CPMs at $1.44 and connected television at $4.68. Against the mobile number, a 50-cent measurement line would equal roughly 35 percent of the clearing price. Against connected television, it would be closer to 11 percent. Those DataBeat numbers are publisher-side clearing prices, before buy-side platform and data fees, so the ratio a buyer actually experiences would be lower. The direction of the comparison holds nonetheless: on inexpensive mobile inventory, a flat per-thousand measurement charge is not a rounding error.

Stewart Sullivan, CEO and co-founder of Drako, framed the decision in terms of paying twice. "Advertisers should not have to pay a second toll for a partial signal about the media they already bought," Sullivan said, according to the release. "FTA has a real job: helping a trading team see what is working so they can move budget while a campaign is live. Questions about changed behavior and incremental visits require a different methodology. We are making FTA free because basic optimization belongs in the campaign workflow. Premium measurement fees should be reserved for insight that creates incremental value."

Fees already under scrutiny

The complaint about stacked charges is not new. Programmatic buyers have spent the past two years dissecting the take rate at each link of the supply chain. IAB Spain's supply-side platform guide, citing the ANA Programmatic Transparency Benchmark 2025, found that transaction costs covering DSP fees, data costs and SSP costs consume 26.1 percent of every dollar invested, and that only 41 percent of programmatic investment reaches genuine, measurable, viewable impressions.

Data costs have been a particular target. When The Trade Desk overhauled its data marketplace on September 29, 2025, the company said advertisers using third-party data typically invested nearly 20 percent of their media costs on it. Its Audience Unlimited product replaced per-segment pricing with tiered rates of 3.3 percent and 4.4 percent of impression costs in Control Mode, and included data at no additional cost in Performance Mode. By that yardstick, a 50-cent measurement add-on equals the higher 4.4 percent tier on an $11.36 CPM, and exceeds it on anything cheaper.

There is a structural parallel between the two moves. In both cases a cost that commonly appears as its own line is folded into a product carrying a price of its own. Who pays for the visit data, then? Drako's release does not disclose what its custom or syndicated audiences cost, so it is not possible to determine from the announcement whether the attribution expense is absorbed by the company or reflected elsewhere in the data fee. What changes for a buyer is the invoice structure: visit reporting no longer appears as a separate charge on eligible campaigns.

A partial count by design

Drako is explicit in the release about what the product cannot do. According to the company, attribution reports reflect only a portion of real-world visits, and the share depends on three variables: the measurement provider's device coverage, the boundaries drawn around each location, and the length of the attribution window.

Each of those is a technical choice with measurable consequences. Device coverage determines how many exposed phones can be observed at all; a provider whose panel draws on a limited set of apps sees a limited slice of shoppers. Boundaries decide what counts as a visit. Location vendors have long offered a choice between simple radii and polygons traced around a building, and Adsquare, when it extended store visit tracking to Austria and Switzerland on January 30, 2025, mapped audience signals onto two-dimensional polygons of store locations. A generous radius around a mall entrance captures passers-by; a tight polygon around a single storefront can miss people who parked and walked in from another side. The attribution window sets how long after exposure a visit still gets credited, and a longer window will report at least as many visits as a shorter one against the same campaign.

None of those parameters are disclosed in Drako's announcement. Nor does the release name the measurement provider whose observations underpin the reports. It refers only to "a measurement provider" in general terms.

Given those constraints, Drako positions the product narrowly. According to the company, visit attribution is most useful as an in-flight optimization tool: it can help media buyers compare campaign tactics, identify stronger-performing approaches and redirect budget while a campaign is still active. Questions about whether a campaign generated incremental visits, or changed consumer behavior, require different measurement methodologies.

That framing aligns with the long-standing methodological objection to attributed visit counts. A matched visit establishes sequence, not cause. Someone who saw an advertisement and then walked into a shop may well have walked in anyway, and the problem of endogeneity bites hard in location campaigns, because targeting by proximity concentrates exposure on the people most likely to visit regardless. Comparisons between tactics measured under identical parameters avoid some of that problem, because the coverage, boundary and window limits apply equally to each, although differences in whom each tactic reaches can still skew the result. That is the use case Drako has chosen not to bill separately.

Where causal measurement sits

For post-campaign evaluation, Drako offers two other products. Its geolift studies compare exposed and control groups to estimate incremental impact. Its market share analysis examines changes in visitation relative to competitors over time. According to Drako, treating the tools as distinct allows advertisers to match each measurement method to the question it is equipped to answer, instead of relying on one attribution number as a complete account of campaign performance.

The release does not state pricing for either study, the minimum campaign size or geography needed to run a geolift test, or how control groups are constructed.

The split mirrors how the wider market has been drawing a line between attribution and incrementality. Geographic experiments have become more accessible during 2026. Jamloop paired household-level holdouts with Recast's GeoLift methodology for connected television campaigns on July 21, 2026, comparing performance across ZIP codes and designated market areas. Google moved its open-source Meridian GeoX geo-experiment tool out of beta on September 9, 2026, claiming experiments 31 percent cheaper. TikTok has offered advertisers Geo Lift Tests to estimate its contribution to sales.

Location vendors have also tried to close the gap from inside attribution products. Adsquare's Attribution Dashboard, opened on June 30, 2025, used a proprietary Control Condition methodology that simulates behavior for unexposed audiences, accounting for seasonality, public holidays and weather, so that mid-flight reporting carries a modelled baseline without manual control groups. Drako's approach runs the other way: it separates the optimization signal from the causal study and prices them differently.

Measurement inside the buying platform

Drako's announcement lands in a market where visit measurement has been migrating steadily into the tools media buyers already use. StackAdapt and Adsquare placed an integrated footfall product inside StackAdapt's buying interface on September 10, 2024, exactly two years before Drako's release, using opted-in SDK-derived location data across EMEA, APAC, North America and Latin America. The Trade Desk brought Adsquare's location-derived data into Audience Unlimited on June 17, 2026, tying it to store visits and in-store conversions. Amazon DSP followed in July 2026 by listing Adsquare location segments in the United States, the United Kingdom and four more markets.

The latest integrations have stretched the concept further. Azira and Adobe embedded cross-border footfall attribution inside Adobe Advertising DSP on August 25, 2026, starting in Asia-Pacific, with Azira building 22 measurement macros into each ad pixel. That announcement disclosed neither attribution windows, minimum dwell time, control methodology nor match rates. Two days later, Azerion's acquisition of mJourney added a product that sells footfall and store visit rate as outcomes without naming the measurement supplier. StackAdapt, for its part, has pushed past presence toward purchases, connecting its platform on July 30, 2026 to Affinity Solutions transaction data covering more than 150 million credit and debit cards.

Against that backdrop, Drako's model differs in one respect. The platform integrations above attach measurement to a particular DSP. Drako attaches it to its audience product and lets the buyer choose the DSP, with a pixel as the common denominator. Whether a pixel-based deployment delivers the same match quality as a native integration is not addressed in the release, which says only that direct integrations exist in some environments.

Location data and the rules around it

Any visit measurement product depends on precise location data, and the legal ground under that data has shifted repeatedly in 2026. The Federal Trade Commission closed its four-year case against Kochava with a stipulated order signed on May 4, 2026, approved by a 2-0 Commission vote, barring the company and its subsidiary Collective Data Solutions from selling sensitive location data without affirmative express consent.

States have gone further. Virginia Governor Abigail Spanberger signed SB 338 on April 13, 2026, banning the sale of precise geolocation data within a 1,750-foot radius from July 1, 2026, making Virginia the third state with an outright ban after Oregon and Maryland. Vermont's Act 138, signed on June 16, 2026, defines precise geolocation as a radius of 1,850 feet and requires data brokers to disclose whether they collect it, with core provisions taking effect on January 1, 2027.

Accreditation offers another reference point. Foursquare Visits became the first location data product accredited by the Media Rating Council under guidelines issued in 2017, a milestone reported in August 2020. Drako's release does not mention accreditation, the sources of its location observations, the consent mechanisms behind them, or the jurisdictions in which the measurement operates. For a Montreal-based company offering the product "in any DSP," the geographic scope is a material open question.

What the release leaves unanswered

The announcement is specific about positioning and thin on parameters. Several items that determine how the reports behave in practice are absent: the identity of the measurement provider, device coverage or match rates, the default attribution window, whether buyers can adjust location boundaries, the markets covered, the DSPs with direct integrations, the reporting cadence, and the price of the audiences that unlock the offer. The same gaps have appeared in other recent footfall announcements covered by PPC Land, including the Azira and Adobe integration.

What the release does establish is a pricing position. A company selling audiences and causal studies has decided that the attribution layer between them will not be billed separately on eligible campaigns, and has said explicitly, in its chief executive's words, that the number it is giving away is "a partial signal."

Timeline

Summary

Who: Drako, a Montreal-based programmatic outcomes partner founded in 2017, led by CEO and co-founder Stewart Sullivan. The change affects agencies activating Drako custom or syndicated audiences in their own DSPs, and advertisers using Drako's managed service.

What: Foot Traffic Attribution, which links ad-exposed devices to subsequent visits observed at physical locations, is now included at no additional charge on eligible campaigns. Drako says the product is often sold elsewhere for about 50 cents per thousand impressions and captures only part of real-world visits. Deployment runs through a conversion pixel in the agency's DSP, with direct integrations in unnamed environments. Geolift and market share analyses remain separate products for post-campaign evaluation.

When: Announced on September 10, 2026, and available immediately for eligible campaigns.

Where: Any DSP, according to Drako, via pixel deployment. The release does not specify markets or name the measurement provider.

Why: According to Drako, visit attribution is a partial signal best suited to in-flight optimization, and basic optimization belongs in the campaign workflow rather than in a separate fee. Sullivan argued that "premium measurement fees should be reserved for insight that creates incremental value," pointing to geolift and market share studies as the tools for those questions.