Google Ads Remarketing and Facebook Remarketing Ads Explained

Remarketing services in the US cost $1,000 to $10,000 a month at an agency, or 10% to 20% of your ad spend, which is the same pricing structure agencies use for paid search generally. Self-serve retargeting platforms typically charge $100 to $500 a month or more in platform fees on top of the media, often tiered by how much you spend. AdBot charges a flat fee by plan, $297 to $1,497 a month and never a percentage of budget, for Meta ads on Facebook and Instagram, with the ad spend going straight to Meta. Google Ads campaigns are not available for new orders right now.

The number to interrogate is not the fee, it is what the fee is charged against. Percentage-of-spend pricing is particularly awkward for remarketing, because remarketing budgets are naturally small: your warm audience is a fraction of your total reach, so a well-run retargeting layer might only be 10% to 20% of your total spend. A percentage model either makes that work unprofitable for the agency to bother with, or pushes them to spend more on it than the audience size justifies.

There is also a media floor worth knowing before you buy anything. If you get fewer than a few hundred relevant visitors a month, your remarketing lists will not fill fast enough to matter and the honest answer is to spend the money on getting traffic first. Our advertising budget calculator works backward from a revenue goal to what you need to spend, and the ROAS calculator tells you what return your margin has to clear before any management fee goes on top.

Google Ads requires a minimum of 100 active visitors or users within the last 30 days before a remarketing list can serve on the Display Network, the Search Network, or YouTube. That threshold is lower than the 1,000-member figure many older guides still quote for Search remarketing, so accounts that gave up on retargeting a few years ago on size grounds are often eligible now. Meta Custom Audiences have a comparable floor of around 100 matched people.

Eligibility and usefulness are different questions though. A list of 120 people will technically serve and will deliver almost nothing, because the platforms need volume to optimize against. In practice a segment starts producing readable results somewhere above roughly 1,000 users, and the useful working range for a warm audience sits between about 5,000 and 50,000. Below that, the right move is fewer, broader segments rather than the ten precise ones a strategy deck will recommend.

List membership duration is the other lever people forget. A 540-day window on a product with a two-week consideration cycle just means you are paying to advertise to people who bought elsewhere last spring. Windows should match how long your buying decision actually takes, which for most ecommerce is 7 to 30 days and for considered B2B purchases can genuinely be six months.

Yes, and the margin is large enough to be worth the setup: reported 2026 benchmark data puts retargeting campaigns at roughly 3.61x ROAS compared with 2.11x for prospecting, retargeting display campaigns convert at two to four times the rate of prospecting display, and Meta retargeting shows the highest average conversion rate of any funnel stage at about 4.1%. The audience already knows the brand, so you are paying to close rather than to introduce.

The caveat that matters for anybody reading those numbers as a promise: retargeting inherits credit for demand your other campaigns created. If you switch off prospecting, your remarketing lists stop filling and the returns collapse within weeks. It is a closing layer, not a standalone channel, and any agency selling it as a self-sufficient acquisition strategy is misreading its own reporting. The honest way to judge it is at account level, the same way you should judge Performance Max.

Attribution also flatters retargeting by default. Last-click models hand the conversion to whichever ad the buyer touched most recently, and the retargeting ad is almost always last by definition. That does not mean the campaigns are worthless, it means the reported ROAS is an overstatement of the incremental gain. The practical test is a holdout: pause retargeting for two weeks and watch total account revenue rather than campaign-level revenue.

Hire a retargeting agency when the work is genuinely creative and cross-channel: you need volumes of new display and video assets produced, the retargeting has to stay in step with email, PR and offline campaigns, or your spend is large enough that a dedicated strategist earns their retainer. Agencies are also the right call when you want somebody accountable for the whole funnel rather than for one layer of it.

Use software when the problem is maintenance. Most remarketing accounts do not fail at the strategy stage, they fail three months in: nobody rebuilt the segments as traffic patterns changed, the same three banners have run since launch, converters were never excluded, and the frequency caps set in January are wrong for December. That work is frequent, unglamorous, and exactly what an automated system does better than a monthly review call. Our comparison of a flat-fee PPC agency model covers the trade-off in full.

If you specifically want cross-site display inventory outside Google and Meta, a dedicated retargeting platform is a legitimate choice and we will say so plainly, since we do not buy that inventory. Our AdRoll alternative page compares that model honestly, including where it wins.

The mechanics are the same on both platforms and take an afternoon if the tracking is already sound. Do it in this order, because getting step one wrong invalidates everything after it:

Ecommerce is the easier case, because the events are unambiguous and the product catalogue does the creative work. Dynamic remarketing shows people the exact items they viewed, pulled from your Merchant Center feed, which means creative production scales without a designer. Cart abandonment is the highest-value segment in almost every store, and it is usually the one running on the same generic banner as everything else. Our ecommerce ads and Facebook ads for ecommerce pages cover the full funnel around it.

Lead generation is slower but often more profitable per conversion, because considered purchases involve several visits by definition. The segments that matter are pricing-page visitors, people who started a form and stopped, and webinar or content downloaders who never booked. The mistake specific to lead gen is retargeting on the wrong outcome: optimizing toward form fills produces cheap unqualified leads, so the campaigns need to be pointed at whatever your sales team calls a real opportunity. For subscription businesses, our SaaS PPC management page covers the trial-to-paid version of the same problem.