> ## Content Index
> Fetch the complete content index at: https://www.livain.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# How to read a competitor's channel mix from the outside: the low-cost stack I use before trusting an internal narrative
- URL: https://www.livain.com/blog/read-competitor-channel-mix-from-outside-ad-library-transparency-center/
- Published: 2026-09-15T07:00:00.000Z
- Updated: 2026-09-15T06:59:59.000Z
- Description: Someone will tell you a competitor is outspending you. Ask how they know. Most of what a competitor does in advertising is now public by law. Here is the low cost stack, the spend triangulation and the error bars that make an outside-in read defensible.
- Author: Remco Livain
- Tags: Competitive Intelligence, Marketing Tech, Paid Media, Strategy, SEO

Someone in your next management meeting is going to say that a competitor is outspending you, or that a market is under-invested and needs more budget, or that a rival's relaunch is going brilliantly. Ask how they know. Nine times out of ten the answer is a feeling, something a sales rep heard, or a conversation at a trade fair last spring.

That is a bad basis for moving budget between markets, and you don't have to accept it any more. Most of what a competitor does in advertising is now public by law, and you can assemble a defensible read of it in an afternoon, for somewhere between nothing and a hundred and fifty euros a month.

I did exactly that this week for a clinic group across five markets. The useful output wasn't a spend figure. It was being able to sit in a discussion about which market deserves the next tranche of budget and say: here is what is observably happening in each one, here's how confident I am, and here's what that contradicts.

Two caveats before the method, because they matter. This is company-level intelligence from public archives and public pages — no individuals, no private data, nothing a platform hasn't already published because a regulator required it. And every number comes out with error bars attached, because directional is the honest word for all of this.

### Why this exists at all

Article 39 of the EU's Digital Services Act obliges very large platforms to keep a public repository of every ad shown in the EU: the creative, the advertiser, who paid, the run dates, the targeting parameters and aggregate reach — and to keep it for a year after the ad last ran. Meta's implementation is the [expanded Ad Library](https://about.fb.com/news/2023/08/new-features-and-additional-transparency-measures-as-the-digital-services-act-comes-into-effect/?ref=livain.com). The obligations are [getting broader, not narrower](https://www.lewissilkin.com/en/insights/2026/01/23/the-eu-raises-the-bar-on-ad-transparency-102mdvw?ref=livain.com). Google runs its own archive in the [Ads Transparency Center](https://blog.google/innovation-and-ai/technology/ads/announcing-the-launch-of-the-new-ads-transparency-center/?ref=livain.com).

So a year of a competitor's creative, flight dates, funnel type and rough audience is sitting in public, and most teams have never opened it for anyone other than themselves.

### The stack, in order of value per euro

| Source                                          | What it answers                                                  | Cost         |
| ----------------------------------------------- | ---------------------------------------------------------------- | ------------ |
| Meta Ad Library                                 | Live ad count, flight dates, funnel type, reach per market       | Free         |
| Google Ads Transparency Center                  | Search / display / video split, account structure                | Free         |
| SimilarWeb (free tier)                          | Direct vs organic vs paid vs social shape, country split         | Free         |
| SEO visibility index (Sistrix, Semrush, Ahrefs) | Is organic invested in or abandoned?                             | Starter seat |
| Subscribing to everything                       | Cadence, offers, segmentation; mail headers reveal the CRM stack | Free         |
| Monthly view-source                             | Pixels switched on or off; UTMs = can they attribute at all      | Free         |
| Review velocity                                 | The only per-location footfall proxy                             | Free         |

**Meta's ad archive is the richest thing on the list and it's free.** Per market you get the live ad count, the campaign themes, when each flight started and stopped, and the funnel type — whether the call to action goes to a booking page or to a native lead form. That last one tells you more about a market's maturity than any spend estimate. The transparency panel then gives reach plus the country and demographic breakdown. Summed reach is a usable proxy for relative pressure between markets; reach times a category CPM band gives you a euro range. Snapshot it monthly and you have a trend, which is what you actually wanted.

**Google's Transparency Center gives counts and formats, not money.** Still useful: the split between search, display and video shows where effort goes, and the advertiser entities listed reveal account structure — two different legal entities running ads for the same brand usually means two teams, two agencies, or a market that was bolted on.

**A traffic estimator for the channel split.** The free tier of [SimilarWeb](https://www.similarweb.com/?ref=livain.com) is enough to see the direct-versus-organic-versus-paid-versus-social shape and the country split. It's the closest single proxy to a channel-mix pie, and it is an estimate — treat the shape as real and the decimals as decoration.

**An SEO visibility index answers one binary question.** [Sistrix](https://www.sistrix.com/?ref=livain.com) is the standard in DACH; a starter [Semrush](https://www.semrush.com/?ref=livain.com) or [Ahrefs](https://ahrefs.com/?ref=livain.com) seat does the same job elsewhere. With a quarter of history the question you're answering is binary: is organic being invested in or abandoned? You don't need precision for that. A site whose blog has been dark for over a year and whose sitemap hasn't moved since 2024 has already answered.

**Subscribe to everything.** Newsletter, member signup, SMS opt-in, one tagged address per market. You learn send cadence, offer strategy and segmentation sophistication — and the mail headers fingerprint the ESP, which tells you their CRM stack. Nobody volunteers that in an interview.

**View-source once a month.** One page per market, and list the pixels present. A pixel appearing is a channel being switched on; a pixel disappearing is one being switched off, or [something breaking that nobody has noticed yet](https://www.livain.com/blog/no-login-to-the-tag-manager/). The real tell is whether the booking links carry UTMs at all — if they don't, that company cannot attribute its own funnel, whatever the dashboard shows.

**Review velocity is the only window onto individual locations.** New reviews per site per month is a crude footfall proxy, and it is the one metric in this whole exercise that is about branches rather than brands.

Then social cadence and follower growth, to separate maintained channels from zombie handles — [Social Blade](https://socialblade.com/?ref=livain.com) gives you the trend curve without manual counting. Then the soft signals: the careers page, because open roles are investment intentions stated out loud; agency credits in the footer; the [Wayback Machine](https://web.archive.org/) for how the promise has changed.

### One spreadsheet, one row per market per month

That's the whole deliverable. Ad count and summed reach, Google ad count by format, channel percentages, visibility score, emails received, reviews added, followers, pixels present. Three to four hours a month to maintain. The first snapshot is a baseline and tells you very little; the second one is where it starts paying, because trend beats level every time.

For the spend triangulation, write the assumptions into the sheet rather than hiding them in a formula: reach times assumed CPM for social, paid-search share times visits times a market CPC benchmark for search. Then state the error band out loud — call it ±50% and mean it. A number with an honest band attached survives a meeting. A confident number that turns out to be invented costs you the room.

> The point of an outside-in read is not to know their budget. It's to be unable to be told a story that the public evidence contradicts.

### What it's for, and what it isn't

It is not a substitute for account access, and it should never be presented as one. What it is good for is the moment when an internal narrative and the observable world disagree — when a market is described as under-invested and its ad archive shows three times the volume of the market that is supposedly the priority, or when a relaunch is described as ongoing and every creative stopped in the same week four months ago.

This is also the clearest example I have of where the leverage in AI tooling actually sits. Pulling, parsing and reconciling nine sources across five markets is exactly the kind of work that used to need a junior analyst for a fortnight, and now takes an afternoon. But the judgement calls — which CPM band is defensible, which estimate is too weak to put in front of a board, which of the nine signals answers the question being asked — are still [the part that needs the reps](https://www.livain.com/blog/the-death-of-generic-ai-why-deep-domain-expertise-is-the-only-real-leverage-left/). Without them you get a beautifully formatted sheet of numbers with no error bars, which is worse than no sheet at all.

Start with the ad archive. You can do it this afternoon, for nothing, on your own company first. What your own year of ads says about you is usually the more uncomfortable read.

### Sources & further reading

**External**  
[Meta — new features and additional transparency measures as the DSA comes into effect](https://about.fb.com/news/2023/08/new-features-and-additional-transparency-measures-as-the-digital-services-act-comes-into-effect/?ref=livain.com)  
[Lewis Silkin — the EU raises the bar on ad transparency](https://www.lewissilkin.com/en/insights/2026/01/23/the-eu-raises-the-bar-on-ad-transparency-102mdvw?ref=livain.com)  
[Google — announcing the new Ads Transparency Center](https://blog.google/innovation-and-ai/technology/ads/announcing-the-launch-of-the-new-ads-transparency-center/?ref=livain.com)

**Tools in the stack**  
[Meta Ad Library](https://www.facebook.com/ads/library/) (free) · [Google Ads Transparency Center](https://adstransparency.google.com/?ref=livain.com) (free) · [SimilarWeb](https://www.similarweb.com/?ref=livain.com) (free tier is enough) · [Sistrix](https://www.sistrix.com/?ref=livain.com) · [Semrush](https://www.semrush.com/?ref=livain.com) · [Ahrefs](https://ahrefs.com/?ref=livain.com) · [Social Blade](https://socialblade.com/?ref=livain.com) · [Wayback Machine](https://web.archive.org/)  
*No affiliate links, no arrangement with any of these. The first two do most of the work and cost nothing.*

**Related posts**  
[Your analytics dashboard is lying to you by leaving things out](https://www.livain.com/blog/analytics-dashboard-lying-by-omission/)  
[No login to the tag manager. The container told me everything anyway.](https://www.livain.com/blog/no-login-to-the-tag-manager/)  
[The death of generic AI: why deep domain expertise is the only real leverage left](https://www.livain.com/blog/the-death-of-generic-ai-why-deep-domain-expertise-is-the-only-real-leverage-left/)