Remarketing Lists (RLSA): Bid More on Warm Visitors

2. The 1,000-User Rule Is on a Page Google Forgot

Quick Answer: Almost every RLSA guide says a search remarketing list needs 1,000 users and can hold them for 180 days. Google’s current documentation says 100 active users in the last 30 days, with a 540-day cap. Two Google help pages disagree, and the older number is the one the internet copied.

This changes who can use RLSA at all. Google’s page on how data segments work says the Search Network needs 100 active users in the last 30 days, capped at 540 days. Google’s page on data segments for Search ads still says 1,000.

The last row is probably where it started. Google kept a 1,000-user floor, but only for customer lists uploaded before 1 February 2024. A rule that survives on legacy uploads got copied everywhere as the rule for everything.

So don’t take either number on faith. Open Audience manager and read the Size column for Search: your own account beats both help pages.

3. How Long Until Your List Is Big Enough to Serve?

Quick Answer: A Malaysian SME site running around 3,000 sessions a month typically reaches the 100 active-user floor within a week of tagging. Under 1,000 sessions a month it takes roughly six weeks, and only about one in eight of those accounts ever reaches the old 1,000-user mark.

Not every session becomes a list member. Consent choices, tracking prevention in Safari and Firefox, and Google’s definition of an active user each take a cut. The gap between Analytics and Audience manager is normal.

Above 8,000 sessions a month, eligibility is a non-event: tag on Monday, list live by Thursday. Under 1,000, the list is a six-week project that then sits near the floor, one quiet month from dropping under.

That is the fork. Above the middle tier, RLSA is normal account structure. Below it, traffic is the constraint, and the budget belongs in organic search.

4. Targeting vs Observation: The Setting That Decides Everything

Quick Answer: Targeting restricts your ads to people on the list and nobody else sees them. Observation shows ads to everyone and simply reports on list members. Pick Targeting on a small list and the campaign quietly stops serving. This dropdown causes more silent RLSA failures than anything else.

• Observation is the safe default. The campaign runs as before; list members are reported separately, so you can see whether they behave differently.
• Targeting is a hard filter. The ad group goes invisible to everyone not on the list. Useful when that is the plan, such as bidding on broad terms you’d only buy for past purchasers.

The failure looks like this. Someone attaches a 400-user list on Targeting, and the eligible audience collapses from everyone in Klang Valley to 400 people who mostly aren’t searching today. Impressions fall off a cliff, no error appears, the account looks broken.

The second reason to prefer Observation is now the important one. Google states that for Smart Bidding campaigns, first-party segments added under Observation are used as signals. It is no longer the timid option; it is how you feed the algorithm.

It is a standard handover check, alongside our 12-check Google Ads audit and proper PPC management.

5. Membership Duration Is the Lever, Not the Bid

Quick Answer: Membership duration decides who is on your list, which decides whether it serves and how well it converts. Malaysian e-commerce accounts peak at a 7-day window. Property, education and aesthetics peak around 180 days. The default 30 days is wrong for both.

Duration is the setting everybody leaves alone and everybody should touch. Chosen once, defaulting to 30 days, it quietly decides the result: a 7-day list and a 540-day list hold different people.

The diagonal is the story. E-commerce decays hard: someone who browsed a bag last week is a live buyer, someone who browsed in February is a stranger. Property and education run the opposite way, peaking at 180 days, because a first enquiry about a condo starts a six-month conversation.

The 30-day default suits neither. Too generous for e-commerce, diluting a sharp list with cold browsers. Far too short for high-consideration buys, expiring people two months before they purchase.

One caveat if you run Smart Bidding, straight from Google: the strategy already accounts for how long someone has been on your segment, so you don’t need to split segments by duration. You still pick one, since it governs who is on the list. But 7/30/90/180-day ladders are work the algorithm now does for you.

6. How Much More Is a Warm Visitor Worth?

Quick Answer: On manual or enhanced CPC campaigns, cost per lead bottoms out around a +30% bid adjustment on Malaysian SME accounts. Past that you pay steadily more for a conversion rate that has stopped climbing. On Smart Bidding campaigns the adjustment does something else entirely.

Here is the ladder, measured only where an adjustment still adjusts a bid.

Conversion rate does climb with the bid, and 3.9% to 5.9% is real. It stops climbing long before the cost does. Between +30% and +100%, conversion rate gains half a point while cost per lead goes backwards by RM 41.

Now the part that reframes the ladder. Google’s documentation states that under a conversion-based automated bid strategy, bid adjustments no longer adjust your bids and instead prioritise which segment a click gets attributed to when overlapping segments could claim it.

Read that twice if you run Target CPA or Maximise Conversions. That number settles an attribution argument between two lists; it does not raise your bid. Most accounts we inherit have a +40% sitting there in that belief.

So, honestly: on manual bidding, bid about 30% more. On Smart Bidding, most Malaysian accounts now, forget the modifier and get the list right. The mechanics are in our Smart Bidding guide, and the same logic drives budget optimisation on Meta.

7. The Shrinking List: Why Coverage Keeps Falling

Quick Answer: The share of site visitors who make it onto a remarketing list falls year on year as browser tracking prevention and consent banners spread. Roughly two in three were capturable in 2022. It is closer to two in five now, and the direction has not changed.

Your list is not shrinking because you did anything wrong. The plumbing it depends on is being dismantled one browser release at a time.

The bottom row explains the internet’s advice problem. When barely a quarter of accounts clear 1,000 users, a guide built on that floor describes a product most readers cannot buy. Dropping it to 100 is what keeps RLSA alive for everyone else.

The middle row is the one to watch on your account. Clearing the floor today is not clearing it in March, and lists near the line fail silently.

You can push back on the top row. Server-side tagging recovers a meaningful share of the visitors browsers drop, which is why it is now standard on accounts that depend on audiences: see fixing the data browsers now block. Consent mode does the rest, starting with clean conversion tracking, which in Malaysia is a Personal Data Protection Act question too.

8. How to Build a Remarketing List That Actually Serves

Quick Answer: Build one list, not twelve. Set its duration to your sales cycle, add it under Observation, and wait for Audience manager to call it eligible before touching anything else. The order below is what stops a list being a decoration.

How to set up a remarketing list for Search ads

1. Confirm the tag is firing everywhere. Google tag on every page, consent mode configured. An untagged thank-you page is why your converters list is empty.
2. Create one list: all visitors. Skip cart abandoners and pricing-page viewers on day one. Five segments of 40 people serve nothing; one list of 600 serves.
3. Set membership duration to your sales cycle. Seven days if people buy in a session, 180 if they take months. Change it now: it applies to everyone already on the list.
4. Add it under Observation. Never Targeting on a first list.
5. Wait for the Size column to say eligible. Check Search specifically: a list can be eligible for Display and too small for Search at once.
6. Only then act on it. Manual bidding: start at +15%, work up to +30%, stop. Smart Bidding: leave the adjustment alone.

Step two is the one people argue with, and the one that saves most accounts. Segmentation feels like sophistication. On a Malaysian SME site it is division: one eligible list, cut into five that aren’t.

9. Where Remarketing Lists Go Wrong

Quick Answer: Remarketing lists rarely fail loudly. They fail by never becoming eligible, by sitting on Targeting on a tiny list, or by carrying a bid adjustment Smart Bidding ignores. All three look identical in a report: a line item with no data.

• Targeting on a small list. The audience collapses to a few hundred people and impressions vanish. No error, no warning.
• Twelve lists, none eligible. The account looks strategic and serves nothing.
• Purchasers left on the buyers list. Exclude converters, or pay a premium to sell people what they just bought.
• No consent mode. The tag fires, the list stays empty, and nobody checks the Size column for a quarter.

The common thread is silence. A broken keyword shouts at you: cost with no conversions. A broken list says nothing, which is why the fastest thing an executive can do is open Audience manager monthly and read one column, in the same rhythm as your automated rule alerts.

And plainly: for a Malaysian SME with modest traffic, remarketing lists won’t transform the account. They make a good campaign a bit better. When the underlying cost of acquiring a customer is broken, no audience setting fixes it.

10. Conclusion

Quick Answer: Remarketing lists still work, but not through the lever the guides point at. Eligibility, membership duration and tagging quality decide the outcome. The bid adjustment is the smallest lever on the page, and on Smart Bidding campaigns it isn’t a bid lever at all.

The head term promises you can bid more on warm visitors. You can, on manual bidding, up to about 30%, after which you pay more for the same people.

Everywhere else the honest version is less exciting and more useful. Get the tag right so the list fills. Set the duration to how your customers actually buy. Add it under Observation. Check the Size column says eligible, and check again next quarter.

Judge it on whether cost per lead moved, not on whether the audience row looks busy. Our guide to a good return on ad spend sets that benchmark, and the ZenWeb Google Ads team sets audiences up this way on every account.

11. Frequently Asked Questions

1. What are remarketing lists in Google Ads?

Remarketing lists, now called data segments, are groups of people who visited your site and triggered your Google tag. A rule decides who joins; a membership duration decides how long they stay. Applied to a Search campaign, the list becomes RLSA.

2. How many users does a remarketing list need for Search ads?

Google’s current documentation says 100 active users in the last 30 days. An older Google page still says 1,000, and most guides copied that figure. Customer lists uploaded before 1 February 2024 do still need 1,000. Check the Size column rather than trusting either.

3. How long can someone stay on a remarketing list?

Up to 540 days for Search and Display, with a 30-day default. The 180-day cap you’ll read elsewhere is out of date. Match the duration to your sales cycle: a week for impulse e-commerce, up to 180 days for property or education.

4. Do RLSA bid adjustments work with Smart Bidding?

Not the way you’d expect. Google states that under a conversion-based automated bid strategy, adjustments stop adjusting bids and instead prioritise which segment a click is attributed to. Add the list under Observation and leave it alone.