Has your ad agency suggested retargeting everyone who visits the website? For a manufacturer whose buyers take months to decide, it can work, but only as a way to stay familiar to people who already found you. It won't find new buyers. And it only runs if the ad platform can match enough of your visitors: Google needs at least 100 active users in the last 30 days, LinkedIn at least 300 matched members, and Meta suggests waiting for several hundred. Three questions decide whether it's worth paying for: is the audience big enough, what will it do across a long decision, and where does the ad send people back to?
Is your audience big enough for B2B retargeting at all?
Google and LinkedIn set a floor below which your audience won't serve, and Meta suggests one. Figures quoted online disagree, so here's what each platform's own help page says as of October 2026.
- Google Ads: "a minimum of 100 active visitors or users within the last 30 days" for Display, Search and YouTube, per Google's page on data segments. The 1,000 still quoted is, on that page, the older rule for customer lists uploaded before February 2024.
- LinkedIn: the audience "must match at least 300 member accounts before it becomes active," per LinkedIn's website retargeting guide. LinkedIn's own audience size page calls 300 the minimum for any ad set and suggests 50,000 "to drive results." That's general ad-set advice, not a retargeting benchmark, but treat 300 as where ads can run, not where they work well.
- Meta: no fixed number on the pages I read. Its troubleshooting page suggests waiting until you have "several hundred people" in the audience.
Your monthly visitor count isn't your audience. Meta only counts visitors who use its apps and leaves out people who've opted out of behavioral advertising. LinkedIn only counts visitors it can match to a member account. Then there's who visits a manufacturer's site: your own staff, existing customers looking up a part number, job applicants, vendors. They all count toward the floor and none of them is who the ads are for.
So the number in your analytics is a ceiling. The real size is whatever each platform reports once its tag has been collecting for a few weeks.
Why install the tags months before you decide
LinkedIn's and Meta's audiences don't fill backward. LinkedIn's audience "starts building on the day you create it," and Meta's help page says its audience "won't backfill." A visitor who came last spring isn't in an audience you create today.
That makes the first step cheap. Put the tags on the site now, build the audiences, run nothing, and look again in 60 to 90 days. Then you're deciding from the reported size, not an estimate, yours or the agency's.
What can retargeting do over a six-month decision with several people involved?
It keeps your name in front of someone who already looked. That person might be the engineer who read your capabilities page after meeting you at a show, or someone a customer referred, during the quiet check a referred buyer runs before picking up the phone. Over a long decision, being familiar when the shortlist gets written is the whole job.
What it can't do is reach the people who never came to your site. If the purchasing manager and the plant manager who sign off never visited, they aren't in the audience.
That's a limit of website retargeting. LinkedIn also lets you upload a list of companies, such as the accounts already in your pipeline, and refine it by job function or seniority, which can reach people there who never visited. It's a separate tactic with its own floor: the list has to match at least 300 member accounts, and LinkedIn recommends 1,000 or more companies.
LinkedIn keeps website visitors for 30 to 180 days, depending on the setting, so a buyer who visited seven months ago has already dropped out. Google lets an audience keep people for up to 540 days, but the 100-user minimum still counts only the last 30, so a long window doesn't rescue a site with little fresh traffic.
Expect few clicks. Someone not ready to buy yet sees the ad and scrolls on, and that can still be the ad doing its job. Judge it over a period as long as your sales cycle, and start with conversion tracking: every quote request gets recorded, ideally in your CRM next to the ad that brought the person back. Click attribution will undercount people who only saw the ad, so also compare quote requests and searches for your company name in Search Console for the months before and during the campaign, and ask new enquiries how they found you. Trade shows and busy seasons move those too, so treat them as signals, not proof.
Where does the ad send them back to?
This part is a website question more than an ad question. The person you're retargeting already saw your site and left without getting in touch. Send them back to the same homepage and you're asking the page that didn't convince them to try again.
With one sitewide audience, which on a small site is often the only one that clears the floor, you can't tell what any one person read. So point the ads at the proof a quick visit usually skips: a project like the work they'd send you, the certifications, the capability detail with tolerances and materials. "Request a quote" asks for the decision they haven't made yet. Proof they probably didn't reach gives them a reason to make it.
Page-by-page sequencing comes later, once narrower audiences clear the floor on their own. Then someone who looked at a process can see its tolerances and certifications, and someone who read only the overview can see a project like theirs.
That only works if those pages exist. If your site has one services page and a contact form, the ads have nowhere useful to point, and building the pages comes first. The landing-page check in Google Ads for manufacturers applies here too. When the site itself is the gap, that's manufacturing website design, not an ad budget.
A decision rule before you sign off
- Tag first. Install the platform tags and build the audiences with nothing running. Wait 60 to 90 days.
- Read the sizes the platforms report. Below the floor, don't run it yet. The money does more getting you found in the first place.
- Narrow only where the numbers allow. Near the floor, the platforms push the other way: Meta's troubleshooting page suggests fewer conditions, "Anyone who visits your website", or a longer window. Well above it, test a second audience that leaves out the careers page and the customer login, so staff, applicants and customers stop soaking up the budget. If only "everyone" clears the floor, that's likely why the agency proposed it, and it can still be worth running small. Confirm the campaign is held to that audience: Meta applies Advantage+ audience automatically to new campaigns where available, and it treats a custom audience as a suggestion, showing ads to "other audiences when it's likely to improve performance." Ask too how often each person in a small audience will see the ad.
- Point the ads at proof, not the homepage. Per-page ads wait until narrower audiences clear the floor on their own.
- Judge it on your sales cycle, with quote requests tracked, not on the first month.
If most of your work comes through trade shows and referrals, this can be the whole paid plan, which is the "Retargeting only" option on choosing B2B paid channels.
