Pathmatics is a digital advertising intelligence platform owned by Sensor Tower, the app and market intelligence company. It estimates how much brands spend on digital advertising, how many impressions they buy, which creatives they run and where those ads appear, across social networks, display, online video, connected TV (CTV), retail media and, since 2026, linear television. Platforms do not publish what each advertiser spends with them. Pathmatics fills that gap with outside estimates built from a panel of consenting users, automated ad collection and statistical models. Agencies, brands, publishers, ad sales teams, analysts and journalists use it as one of the few outside views of where advertising money goes.

How the estimates are built

Pathmatics does not, in general, receive spend data from advertisers or platforms. Its numbers are modelled in three stages.

The first stage is observation. Sensor Tower's product page lists three inputs: first-party panel data from opted-in users, automated web crawlers that detect creatives, placements and ad paths, and partnerships and platform integrations. The panel matters most for social and app advertising, where ads are targeted to logged-in users and a crawler visiting a page sees nothing. Sensor Tower's data sourcing page says its proprietary panel covers more than 10 million consumers who use the company's own mobile apps, a list that includes the screen-time tools ActionDash and StayFree, the Luna VPN and Adblock Mobile. The company says data is de-identified on the device.

The second stage is pricing. Each impression served to the panel or data aggregators is assigned a cost per thousand impressions (CPM), according to an EMARKETER methodology profile, and impressions multiplied by CPM give a spend estimate per creative.

The third stage is classification. Each creative is tagged to an advertiser, brand, category and placement, and Sensor Tower says it uses AI to tag creatives at scale. Because Pathmatics also records the path an ad took, from publisher through intermediaries to the viewer, it can split activity into programmatic and direct buys.

A worked example shows the arithmetic. If a model credits a sportswear brand with 400 million US Instagram impressions in a month at an assigned CPM of $8, the estimated spend is $3.2 million. Both inputs are estimates, so an error of 25% in either compounds in the dollar figure.

The output is a set of reports: spend and impression estimates, creative libraries with flight dates and calls to action, site targeting, share of voice and category rankings. A Site Strategies view shows which advertisers run on a given publisher and which are absent.

Coverage as of October 2026

According to the product page, Pathmatics covers Facebook, Instagram, YouTube, TikTok, X, Snapchat, LinkedIn, Pinterest and Reddit, plus display and video on mobile and desktop, and over-the-top (OTT) streaming on Netflix, Amazon Prime Video, Disney+, Hulu and Peacock. Core markets are the US, Canada, the UK, Germany, France, Italy, Spain and Australia, with selected markets in the Middle East and North Africa, Latin America and Asia-Pacific. A March 2026 release added Thailand, Norway, Sweden, Denmark and Austria.

Retail Media Insights, launched inside the Pathmatics interface in July 2025, tracks onsite, offsite, paid social and OTT activity for more than 40 retailers in the US, the UK, Canada and Europe. Linear TV arrived in March 2026, covering 120 US networks and 6,000 programmes. In August 2026 the company began tracking US ChatGPT ads, with mobile data from April 1, 2026 and desktop data from June 1.

Origin and evolution

Pathmatics was founded in Santa Monica by Gabe Gottlieb, its long-time chief executive, and a co-founder who, like Gottlieb, had worked at Microsoft, according to a socaltech interview. Sources disagree on the year. Socaltech's report on the company's Series A in March 2016 said it was founded in 2010; Sensor Tower's 2021 acquisition announcement says 2011.

The name described the original idea. "Pathmatics is in the name, and that refers to how we understand how ads are delivered across channels," Gottlieb told socaltech. Early customers were publishers, followed by ad tech companies, agencies and brands. Bertelsmann Digital Media Investments led the $3 million Series A. In November 2016 Pathmatics bought WhatRunsWhere, a six-person Toronto company selling ad tracking to small businesses.

The method moved with the market. Programmatic targeting and logged-in social feeds made the most valuable ads invisible to crawlers, and the industry's answer was panels. Market Track's Competitrack added an opt-in panel in 2018 for that reason. By 2021 Sensor Tower was describing Pathmatics as a panel-based product covering Facebook, Instagram and Twitter as well as display, video, mobile and native.

Sensor Tower announced the acquisition on May 24, 2021, with a growth investment from Riverwood Capital. Terms were not disclosed. Pathmatics then had more than 250 customers and 7,000 platform users, had raised $7.7 million and had doubled brand software revenue in 2020. The deal promised TikTok, YouTube mobile, Snap and OTT coverage. Sensor Tower then bought data.ai, its main app intelligence rival, in March 2024; TechCrunch reported Oliver Yeh as Sensor Tower's chief executive at the time. Sensor Tower says it added 66 channel and country bundles in 2025. On July 28, 2026, Sensor Tower announced that a Pathmatics integration in Google Ads, in beta since late 2025, lets advertisers pull their Facebook, Instagram and TikTok creatives into Performance Max and Demand Gen campaigns.

The naming question

The brand is in transition. Sensor Tower's product page is still titled Pathmatics, and the July 2026 Google Ads release refers to "Sensor Tower's Pathmatics platform". Yet the same page labels its main image "Digital Advertising - Insights (formerly Pathmatics)", and Sensor Tower's August and September 2026 reports use "Digital Advertising Insights" alone. No formal rename announcement was found as of October 2026.

Why it matters for marketers

The large platforms report revenue, not who spends it. Third-party ad intelligence is therefore how the market learns about shifts in advertiser behaviour before earnings calls, or in places that earnings calls never cover. ChatGPT is the clearest recent case. OpenAI has stated a $1 billion annualised ad run rate without monthly figures, and Sensor Tower's panel supplied much of the detail around it: about 1,200 US advertisers in August 2026 and financial services rising from 2% to 13% of tracked US spend between April and August, while shopping and retail fell from 37% to 21%. The same firm's estimate that ads per user per hour on the mobile app rose 163% over the period was one of the few outside measures of how OpenAI was loading its inventory.

The data also gives scale to platform-level trends. Sensor Tower estimated in August 2026 that video had passed half of Instagram's ad placements in the second quarter. Its September 2026 report put US digital ad spend at about $201 billion for the year to July 2026, up 15%, with Facebook the largest channel at $42.9 billion. Earlier, during the July 2020 Facebook boycott, Pathmatics showed Facebook's top 100 advertisers spending about $30 million less than a year before, according to MediaPost.

Limitations and disputes

Every Pathmatics figure is an estimate. Sensor Tower's standard caveat states that estimates "may not accurately represent metrics for any app or advertiser" and should be used for directional trend analysis. The company publishes no error margins. Spend is impressions multiplied by an assigned CPM, so a brand that negotiates deep discounts, buys through private deals or runs heavily discounted remnant inventory will be overstated, and a premium buyer understated.

The panel is a second source of uncertainty. It is drawn from users of Sensor Tower's own apps, including a VPN and an ad blocker, and no independent audit of its representativeness has been published. Coverage thins outside the core markets, and observed creatives reveal nothing about conversion rates or return on ad spend.

The sharpest dispute came on X. Sensor Tower reported in November 2023 that 50 of X's top 100 US advertisers from October 2022 had stopped spending, and Reuters reported 51 the following month. In January 2026 X's head of advertising, Monique Pintarelli, said 97 of the top 100 had returned. Sensor Tower's data, reported by Digiday, showed 64 of the 2022 top 100 spending in 2025, at about half their earlier level. One measure counts presence, the other spend; neither side published underlying data.

Methods also vary between vendors. Nielsen's Ad Intel tracks 5.5 million brands across 23 media types, while BIScience's AdClarity relies on a 30-million-user panel in 51 markets.

Pathmatics versus adjacent terms

  • Platform ad libraries. Meta's Ads Library and Google's Ads Transparency Center show creatives an advertiser is running, with spend only for political or regulated ads in some regions. They are first-party and free but give no commercial spend. Pathmatics models spend across platforms.
  • Nielsen Ad Intel. A cross-media monitoring service rooted in TV, print and radio. Pathmatics began in digital and added linear TV only in 2026.
  • Sensor Tower App Advertising Insights. A sibling product covering in-app ad networks and the demand-side and supply-side platforms (DSPs and SSPs) trading app inventory.
  • Share of voice. A metric, not a product. Pathmatics is one source of share of voice estimates in paid media.

Recent developments

Linear TV and five new markets arrived in March 2026. ChatGPT tracking followed in August, and by October 2026 PPC Land was treating Sensor Tower's monthly US totals as one of the few third-party measures of an ad business that had by then expanded to 89 countries. The Google Ads integration moved out of beta in July 2026. US podcast advertising, listed on Sensor Tower's 2026 roadmap, had not been confirmed as launched as of October 2026.

Timeline

  • 2010 or 2011 - Pathmatics founded in Santa Monica, led by Gabe Gottlieb (sources differ on the year).
  • March 3, 2016 - $3 million Series A led by Bertelsmann Digital Media Investments.
  • November 24, 2016 - Acquisition of Toronto-based WhatRunsWhere reported.
  • 2018 - Omnicom Media Group extends exclusive Pathmatics partnership for the Middle East and Africa.
  • August 2020 - Pathmatics data on Facebook's top 100 advertisers during the July boycott reported.
  • May 24, 2021 - Sensor Tower acquires Pathmatics, with Riverwood Capital investment; terms undisclosed.
  • October 2022 - Sensor Tower adopts differential privacy for panel data.
  • November 2023 - Pathmatics by Sensor Tower reports 50 of X's top 100 US advertisers had stopped spending.
  • March 18, 2024 - Sensor Tower announces acquisition of data.ai.
  • July 2025 - Retail Media Insights added to the Pathmatics interface.
  • Late 2025 - Google Ads creative integration enters beta.
  • January 2026 - Sensor Tower reports 66 new channel and market bundles added during 2025.
  • March 2026 - Linear TV (120 US networks) and five new markets added.
  • July 28, 2026 - Google Ads integration announced for Performance Max and Demand Gen.
  • August 2026 - US ChatGPT ads tracking launched under the Digital Advertising Insights label.
  • September 2026 - State of Digital Advertising 2026 puts US digital ad spend at about $201 billion.

Summary

Who. Pathmatics was founded by Gabe Gottlieb and a fellow former Microsoft employee, and has been owned by Sensor Tower since May 2021. Its users are brands, agencies, publishers, ad sales teams, investors and journalists.

What. A digital advertising intelligence platform that estimates competitors' ad spend, impressions, creatives and placements, using an opt-in panel, crawlers and models that assign CPMs to observed impressions. It is increasingly marketed as Sensor Tower Digital Advertising Insights.

When. Founded in 2010 or 2011, acquired in 2021, and expanded in 2025 and 2026 into retail media, linear TV and ChatGPT ads.

Where. Strongest in the US, Canada, the UK, major EU markets and Australia, with selected coverage in other regions, across social, display, video, CTV, retail media and US linear TV.

Why. Platforms do not disclose who spends what, so marketers rely on outside estimates for competitive and market views. Those estimates are directional, unaudited and sometimes disputed by the platforms they measure.