Behind the networkBehind the network

Your network's report shows a sale on Wednesday 14 October, credited to a publisher you approved last year. The same day, an ad on your brand search carried a different name in Google's “About this advertiser”: a company that isn't on your partner list.

You ask the network who that is, and the answer comes back as a sub ID: a short string from the publisher's link. It's half an answer.

Both records can be accurate. The publisher in your report may be a sub-network with its own publishers behind it, and the lookup that joins the two names sits with that intermediary.

How the extra layer works

A sub-network is a business that joins your programme as a single publisher and offers it on to its own members. Those members, its sub-publishers, join the sub-network rather than applying to your network one by one. They usually give up part of their commission in return.

So approving a sub-network is one decision about one business. It isn't a review of every site behind it, and you aren't told when a new sub-publisher joins. A sub-publisher you removed, or one barred from your programme, can come back in through that layer without applying to you again.

Networks' rules for this differ. One large UK network's code holds sub-networks solely responsible for their sub-publishers, as if the actions were their own, bans sub-networks of sub-networks, and asks for each sub-publisher's ID in the first click reference. Another UK network's sub-network policy is still voluntary.

What a sub ID is, and isn't

A sub ID (some networks say click reference, or click ref) is free text the publisher adds to its own tracking link. Neither you nor the network sets it. It can hold a website, a placement such as “Header banner, page 2”, or an internal code. It belongs to the click and, through it, to the sale.

It doesn't name the Google ad, the keyword or the Google Ads account. On one large UK network, the sub-publisher's ID there is the sub-network's internal code, not a name, and its dashboards of which sub-publishers promote which advertisers aren't shared outside its compliance team.

What reaches you varies too. On one network a link can carry six click refs, and advertisers see only the first. On another, the brand sees none of the partner's three sub IDs, only a separate shared ID. One sub-network gives brands the publisher's domain, never the page. So ask which field you receive. The useful identifier is the one your team can see and the intermediary can map to a publisher and a domain.

Two names for one ad

Google's “About this advertiser”, behind the three-dot icon beside a search ad on desktop (the info icon on mobile), shows a verified advertiser's legal or trade mark name and where it's based. For an unverified advertiser it shows the payments profile name. Where a verified name differs from the payments profile's, Google's help says the disclosure adds “Ads funded by” with the profile's name.

That name is whoever holds the Google Ads account. It can be a company that appears nowhere in your network's report, which lists the sub-network instead. A sub ID can't close that gap, because it was never about the ad. The lookup has to come from the intermediary: this code, on these sales, is this publisher at this domain. The guide to finding the partner behind an ad lines the name up with the ad's other clues and your own records.

Why it's easy to miss

Every record you hold is in order on its own. The network's row names a publisher you approved, and the sale came through a valid tracking link. Auction Insights labels other advertisers by display domain, not by Google's name for them, so it won't join the two either.

Removing a publisher from your direct partner list doesn't touch a different approved route either. If the same site reaches you through a sub-network, your removal never reaches it. The restriction has to reach the underlying source.

What it costs

Imagine forty pound orders at five per cent commission, so two pounds each. Fifty of them in October came through one sub-publisher's links, after clicks on an ad on your brand search.

Your report shows £100 to the sub-network, mixed with its other sources. How it's split with the publisher behind it doesn't appear. If those shoppers were already searching for your name, that £100, and any network fee on it, went on customers who were already looking for you.

Where it shows in your records

The ordinary explanations

Most of this has a plain answer, and you can check several before anything goes to the network.

• The advertiser Google names is the company behind a publisher you approved: a parent company, or an agency that runs its account, under a name you've never seen in your report.
• Your own agency, a reseller or another team at your company runs ads that show on your brand search.
• A sub-publisher joined after your paid search rules were set, and the sub-network never passed them on.
• The code that matches nothing on your list is the sub-network's internal code, which was never meant to.
• A search engine referrer can come from an ordinary organic result, a page that ranks for your name and a code.
• Your own sale or newsletter lifted brand searches and the sub-network's sales on the same days.

What to do

1. Record each sighting: the search, the date and time with its time zone, the location and device, the ad as it read, and the advertiser Google names. Take a full screenshot. Don't click the ad: a click costs its advertiser money and can set tracking cookies.
2. Look the advertiser up in the Ads Transparency Center: search its name or your web address with the region set to the United Kingdom, and note its advertiser id (it starts with AR) and whether Google has verified it.
3. Search your network's publisher list for that name, and list your programme's sub-networks.
4. Pull the sub-network's transactions for the days around the sighting, with click time, the click ref you receive, the referrer and the order references.
5. Write to your network with the advertiser Google names, the time you saw the ad and the click refs on the matching sales. Ask for the publisher and domain behind each, from the sub-network, in writing. A code without its lookup is unfinished.
6. If the source is outside your terms, ask for it to be excluded by name, with the date that takes effect. One sub-network says it usually acts by the next working day, though sales can still track until earlier cookies expire, so ask how earlier clicks will be treated.
7. Review the matching commissions before your validation window closes. What you can do covers that deadline, and reporting an ad under Google's Unfair advantage policy.
8. Write sub-networks into your programme terms: your paid search rules bind their publishers, and the intermediary enforces them, names its publishers on request and acts on named exclusions. If one won't, decline it or end the relationship under those terms. On one network, that can be on seven days' notice or immediately.

Networks differ in what they'll do. On one, a sub-network must acknowledge a notice within 24 hours and resolve it within 48. Another can withhold commission for up to 90 days where it confirms activity outside its policy.

Seeing it in Godotoro

Godotoro searches google.co.uk from UK addresses in desktop browsers, recording every ad as it showed. It never clicks an ad and never requests an ad's click link.

Each ad carries the advertiser Google names for it, from “About this advertiser”: the verified name, where they're based and their Ads Transparency id, with a link to their ads on the Ads Transparency Center. When Google lists an affiliate network's address for the click, Godotoro names the network. Who's bidding shows, per search, how many checks each advertiser was seen in, its position, when it was first and last seen, and its distinct ads.

That's the identity to take to your network, with the times beside it, so it can ask the intermediary to account for the traffic. Godotoro describes what was seen; what it means under your terms is yours to decide. To see who appears on your searches, check your brand.