Can you see a competitor's ad spend? What each ad library shows in 2026

Can you see a competitor's ad spend?

A competitor's ad spend is not published by Meta's, Google's or LinkedIn's ad library for a commercial advertiser. On October 11, 2026, we opened the same advertiser, Nike, in Meta's Ad Library, Google's Ads Transparency Center and LinkedIn's Ad Library, and none of the three showed a spend figure for any of its ads. Spend appears in exactly one case: ads about social issues, elections or politics, where Meta and Google show a spend range per ad; LinkedIn shows none at all. Everything else a library shows about a commercial ad is a date, a format, a targeting summary or an audience count.

There are three kinds of answer, and it helps to know which one a competitor falls into. Political and issue ads carry spend and impressions ranges in Meta's and Google's libraries. Ads shown in the European Union carry targeting and an estimated reach or impressions figure, because EU law requires it. Every other commercial ad carries only its content, its dates and its format, and on Meta it may not be in the library at all. The sections below set out what each library showed on the day, why the EU rule stops at reach, how spend estimates are made, and what overads shows.

What the Meta Ad Library shows for commercial ads

Meta's Ad Library shows a commercial ad's creative, its start date, whether it is active, the Meta platforms it appeared on, and, for ads shown in the EU, the targeting the advertiser chose and an estimated reach. Meta's Ad Library API page, read October 11, 2026, lists what every archived ad carries: a Library ID, the ad creative, the Page name and Page ID, delivery dates and the platforms. The same page lists spend, impressions and demographic distribution as fields for ads about social issues, elections or politics only, in ranges such as 1K to 5K and 10K to 50K. For ads delivered to the UK and the EU it adds estimated impressions, EU targeting and reach demographics, and advertiser and payer information for the EU.

One Nike ad shows what that means in practice. Library ID 1870984290951787 was active on October 11, 2026, had started running on September 2, 2026 and had six versions. Its Transparency by location panel listed the EU audience as France, Germany, Italy, the Netherlands, Poland and Spain, ages 18 to 65 and over, all genders, and an estimated EU reach of 8,110,419 Meta accounts, broken down by country, age range and gender. The panel carried no spend line and no impressions line. Nike's advertiser view with Germany selected had an About tab that showed Page transparency details, such as the Page's creation date of May 21, 2008 and the countries of the people who manage it, and no spend figure. Third-party guides that describe a 12-month spend tile on EU advertiser profiles did not match what we saw that day.

The second thing to know is which ads are in the library at all. Meta's API page says the archive holds ads about social issues, elections or politics delivered anywhere in the world during the past seven years, and ads of any type delivered to the UK or the EU during the past year. Its note on the required country parameter adds that ads which did not reach any EU location are returned only if they are political or issue ads. A competitor that advertises only in the United States, India or Australia therefore has no commercial ads in the archive, and a search that returns nothing is not evidence that they run none.

Spend does appear in Meta's library for one category. Meta's Ad Library report covers ads about social issues, elections or politics since October 2019 and shows totals and spend by advertiser and by location. With India selected on October 11, 2026, it listed 5,161,696 ads and a total spent of 7,941,084,877 rupees, with a per-Page table of amount spent and number of ads for the chosen date range. Nothing on that report covers a commercial advertiser.

What Google's and LinkedIn's libraries show

Google's Ads Transparency Center and LinkedIn's Ad Library show less per ad than Meta does, and neither shows spend for any commercial ad. Google's ads transparency help page describes the center as a searchable repository of advertisers and the ads they have served on Google platforms, with the advertiser's name, location and verification status and the ads served over a period. For ads served in certain countries or regions, the page adds targeting information, the total number of recipients for each ad and a subject-matter label, and it says Google provides API access to data about ads served in the EEA.

In the Ads Transparency Center on October 11, 2026, a search for nike.com returned about 100,000 ads across several advertiser accounts, including Nike, Inc., Nike Retail BV and Nike Japan, each marked verified. The Nike, Inc. advertiser page showed the legal name, that the advertiser is based in the United States, that it had verified its identity, and about 9,000 ads. One ad's details showed a last-shown date of October 10, 2026, the format Text, that it was one of five variations, and a region selector set to anywhere. No run length, impressions or spend appeared on the commercial ad.

Google's political report is the contrast. Its United States political advertising view covers election ads published through Google Ads and Display & Video 360 since May 31, 2018, and each card carries a times-shown range, a date range and a spend range. On October 11, 2026, one advertiser's card read 1 million to 1.25 million times shown, October 6 to 10, 2026, and 35,000 to 40,000 dollars spent. Those three numbers exist for political ads and for nothing else in the center.

LinkedIn's Ad Library help page says the library includes ads that ran on LinkedIn after June 1, 2023, that an ad stays for one year after its last impression, and that it usually appears shortly after its first impression. Every ad shows a preview, the format, the advertiser name, the payer name and whether the ad is restricted. For ads targeted to the European Union the library adds impressions, targeting and the dates the ad ran, and the search form gains an impression-range filter.

One Nike job ad in LinkedIn's library, ad 1427944723, showed on October 11, 2026 that it ran from June 2 to October 10, 2026, with estimated total impressions of 200,000 to 300,000, a country split of 34 percent United States, 20 percent India, 4 percent Netherlands and 4 percent United Kingdom, and a location targeting summary of North America, Asia and five other regions. No spend appeared.

TikTok's Commercial Content Library was not checked. Its pages and TikTok's newsroom refused connections from India, where this check ran and where the Ministry of Electronics and IT blocked TikTok with 58 other apps on June 29, 2020. This guide makes no claim about what TikTok's library shows.

Why the EU rule stops at reach and never reaches spend

The EU's Digital Services Act is the reason every large platform runs an ad library with targeting and reach, and the reason none of them publishes spend. Regulation (EU) 2022/2065, dated October 19, 2022, sets out in Article 39 what a very large online platform must put in its ad repository, through a searchable tool and an API, for as long as an ad runs and for one year after it was last shown. Paragraph 2 lists seven items: the content of the ad including the product, service or brand; the person on whose behalf it is presented; the payer, if different; the period it was presented; whether it was targeted at particular groups and the main parameters used, including exclusions; commercial communications identified under Article 26; and the total number of recipients reached, with aggregate numbers by Member State for targeted groups.

Spend is not on that list. That is why Meta's EU panel shows targeting and an estimated reach, why LinkedIn adds impressions and targeting only for EU-targeted ads, and why Google's help page limits recipients and targeting to ads served in certain countries. Each library publishes what the law requires for the ads the law covers, and the one-year retention in Meta's and LinkedIn's libraries matches the one-year period in Article 39. Outside the EU, nothing in the regulation applies, which is why Meta keeps only political and issue ads for the rest of the world.

Political spend has a different origin. Meta's report and Google's political view predate the DSA, and both are built around a paid-for-by disclaimer and a spend range. Those are the platforms' own political advertising policies at work, and they are the only place a spend figure appears in any of the three libraries.

Why some tools show competitor spend estimates

Tools that show a competitor's spend are modelling it, because no platform publishes the number for a commercial ad. Sensor Tower's Pathmatics page, read October 11, 2026, describes its method as a combination of first-party panel data from millions of users who opt in to sharing data, automated web crawlers that detect creatives, placements and ad paths, partnerships and platform integrations, and machine learning models to estimate spend and impressions. The page calls its outputs ad spend and impression estimates, and it does not state the panel size, the markets the panel covers or how the models are validated.

A modelled figure can still be useful, as long as it is read as a model. Three questions separate a usable estimate from a decorative one. Which markets and devices does the panel cover, and is the competitor's audience inside them? Is the number an observed count, such as an EU reach from a library, or a modelled one? And does the vendor publish an error range against a known budget? A vendor that cannot answer the second question is selling a guess with a currency sign on it.

The same caution applies to anything derived from a library. An EU reach of 8 million accounts says how many people saw an ad at least once in six countries. It does not say how much the ad cost, how long it will run, or whether it sold anything, and a tool that turns it into a spend figure has added an assumption the data does not contain.

What you can learn about a competitor without spend data

Without spend data, a competitor's ads still show what they are selling, to whom, where, for how long, and in how many versions, and those five things are enough to read a strategy. The useful fields are the ones every library publishes: the creative, the start date, the platforms, the landing page in the creative's link, the EU reach or impressions range where it exists, and the number of variations under one ad. A recurring check of those fields, run by hand, looks like this:

1. Set the country to an EU member state in Meta's library before searching, or a commercial competitor's ads will not appear.
2. Record each ad's Library ID, start date, platforms and number of versions, and the landing-page domain in its link.
3. Open the EU transparency panel and note the targeted countries, the age range and the estimated reach.
4. In Google's center, search the competitor's domain, note how many advertiser accounts point at it, and note the last-shown date and format of the newest ads.
5. In LinkedIn's library, filter by advertiser and an EU country, and note the run dates and the impressions range.
6. Repeat on the same weekday and compare. Pick a threshold, 60 days for example: an ad still active past it has kept its budget in the advertiser's eyes, and a new landing page in the link means a new offer.

Two readings to avoid. A long run is not a performance score, because a retargeting ad on a small daily budget can outlast a scaled launch by months. And an ad that disappears from Meta's library may have aged out after a year rather than stopped, because the archive holds commercial ads for one year and offers no history before that. Meta's API page also notes that the delivery stop time it publishes is when the advertiser wants delivery to end, and that a blank one means the ad runs until the advertiser stops it or the budget is spent, so a blank is not a sign that an ad is still running.

The one thing a library cannot show is what a competitor changed on their own site. Pricing pages and landing pages move more often than ad creative does, and watching them is a separate job that needs a crawler rather than a library. Our guide to why many ad tools can read your campaigns but not change them covers the related question of what a tool can see in your own ad accounts.

Where overads fits

overads is AI marketing automation built only for marketing, and its Competitors view is built only on observed data: it shows what a library or a crawl returned and leaves blank what it could not observe. The view has three tabs. The Landing pages tab uses an overads crawler that re-reads pinned competitor pages once a day and records each detected change as the text before and the text after, with provenance on every response, and the same data is on the API and the MCP server as crawl changes and the list_page_changes tool. The Ad archive tab searches Meta's Ad Library and indexes an advertiser by its Meta Page ID: overads reads that Page's archived ads once a day over the EU's 27 countries and the UK, asks Meta only for the fields it publishes for commercial ads, keeps each sighting for one year to match Meta's own retention, and shows how long each ad stayed in the archive against the ads the same advertiser launched the same month. The Instagram tab reads a competitor's public posts through the customer's own connected Instagram professional account, and shows a notice instead of numbers when Meta has not granted the permissions it needs.

What overads does not do:

• It does not show competitor spend, impressions, click-through rate or return on ad spend, and it does not estimate them. Meta publishes those for political and issue ads only, so there is no observed figure to show for a commercial competitor.
• It does not read Google's or LinkedIn's libraries, and it does not cover TikTok.
• It cannot read Meta's archive on its own. Reading it depends on one credential that has passed Meta's identity confirmation, and when none is usable the Ad archive tab says the advertiser was not read rather than showing zero ads.
• It shows a dash, not a zero, wherever a figure is missing. A dash under reach means Meta did not publish one, and a dash under the run-length comparison means the sample was too small to compare.
• It holds no write access to any ad account. Its own ad reporting reads Meta, Google and Snapchat read-only, and every workflow starts out asking for approval before anything goes out.

A team that wants a modelled spend number is better served by a vendor that states its panel and its method, such as the one above. A team that wants a one-off look at a competitor is better served by the three libraries themselves, which are free and need no account. overads fits a team, or an agency with several clients, that wants the observed fields recorded every day without a person doing the weekly round by hand, next to the pricing and landing-page changes that no library shows. Plans are on the pricing page.