Retargeting Measured Against A Holdout, Not An Attribution Window

Campaigns across Google Ads and Meta, run by one senior lead, and read against a group of people we deliberately do not advertise to.

Get My Free Retargeting Audit

Komplex runs retargeting and remarketing from Austin, Texas, with a presence in New York City and Reykjavik. We are a Google Partner and have run digital work for companies from local service businesses to global retail brands. Campaigns run across Google Ads and Meta, using your own first party data rather than bought audiences. The work covers segmentation, suppression, frequency, dynamic product ads, and a measurement approach that compares against people we hold back. It sits inside our display and video advertising service.

Google Partner

Founded In 2014

You Own Every Account

No Long Contracts

Retargeting Reports The Best Numbers In Your Account, And Some Of Them Are Not Real

Look at any account and retargeting will have the strongest return on ad spend in it. That is not because retargeting is the strongest channel. It is because you are advertising to people who already visited your site, already know your name, and in many cases were already coming back. When one of them buys, the platform counts the sale. It has no way to know whether the advert changed anything. So the reported figure mixes real influence with credit taken for work somebody else did. Nobody can separate the two from inside the report.

Isometric illustration of a marketer reviewing keyword research as Komplex ranks number one in search results.

The Only Honest Way To Find Out Is To Stop

Hold back a portion of the audience, show them nothing, and compare what they do against the people who saw the ads. Whatever gap appears is what retargeting actually added. It is simple, it costs a little reach, and it is uncomfortable. The number that comes back is almost always smaller than the one in the platform. It is also the only figure worth building a budget on. We set the holdout up at the start, not after somebody senior starts asking hard questions.

Ask About Your Account

What Retargeting And Remarketing Management Covers

Eight lines of work. Two of them are about showing fewer ads to fewer people, which is where most of the improvement comes from.

Audience Segmentation

Cart abandoners, product viewers, blog readers, and past customers are four different problems. Treating them as one audience is the most common fault we find.

List Hygiene

Windows that match how long your purchase actually takes. A thirty day window on a product people buy in two days is paying to reach people who already decided.

Dynamic Product Retargeting

Showing the specific items someone looked at, driven off your product feed. Powerful for ecommerce, and entirely dependent on the feed being clean.

Display Retargeting

Image placements across the Google Display Network, with exclusion lists maintained so the ads are not funding sites built to carry advertising.

Frequency Capping

The difference between a reminder and a nuisance. Uncapped retargeting damages the brand it is supposed to be supporting.

Suppression

Removing people who already bought, already converted, or already said no. Every impression served to them is money spent making someone slightly annoyed.

Creative Rotation

Refreshing what people see before it becomes wallpaper. Small audiences see the same ads far faster than anyone expects.

The Incrementality Read

A holdout group set up at the start, and a monthly comparison. This is the number we report and the one we would defend in front of your board.

Start With What Is Running Now

Read only access is enough. Most retargeting problems are visible in the audience settings within twenty minutes.

Get The Free Audit

RANKED: MOST INCREMENTAL TO LEAST

Eight Retargeting Audiences, Ranked By How Much They Actually Add

Every one of these will report a good return. Only the ones near the top are likely to be adding much. The order is what we would fund first and cut last.

1
Cart Or Form Abandoners, Under Forty Eight Hours
Genuine intent, interrupted, still fresh. This is the one audience where retargeting reliably changes an outcome, and it should be capped tightly because it works quickly or not at all.
2
High Intent Page Viewers Who Did Not Convert
Pricing pages, demo requests, contact pages. People who got close and stopped. Worth a specific message rather than the same ad everyone else sees.
3
Product Viewers With A Long Consideration Cycle
Furniture, kitchens, software, anything bought over weeks. Retargeting keeps you present during a decision that was always going to take time.
4
Lapsed Customers Past Their Repurchase Window
People who bought once and are overdue. This is close to email territory and works best when the two are coordinated rather than both firing at once.
5
Video Viewers Who Watched Most Of It
A real signal of interest, and cheap to act on after a video campaign. Smaller audience than people expect, so keep expectations proportionate.
6
General Site Visitors, Thirty Day Window
The default setting in most accounts and the middle of this list for a reason. It includes people who bounced in four seconds and people who read everything, treated identically.
7
Everyone Who Touched The Brand, Ninety Days
Very large, very diluted, and reports beautifully because it contains all the people who were coming back regardless. This is where the flattering numbers come from.
8
Existing Customers With No Suppression
Not a strategy, a mistake. You are paying to advertise to people who already bought, then counting their repeat purchases as advertising success.

Which Of These Are Running In Your Account?

Usually more than anyone realises, several of them overlapping, all bidding against each other.

Check My Audiences

WHERE IT RUNS

Google Ads, Meta, And The List You Already Own

Retargeting is bought on the platforms you already have accounts with. Here is what each one is good for and what changed on it recently.

Google Ads

Display, YouTube, and search remarketing lists. The widest reach of the three, and where dynamic product ads run off your feed. Managed inside the same account as your paid search work.

Your Own List

Customer match and audience upload. First party data you already hold, used for suppression as much as for targeting. It is the cheapest and most durable audience source you have.

What Changed Recently

Chrome kept third party cookies after announcing it would remove them. Safari and Firefox still block them by default, so a meaningful share of your audience was never reachable this way anyway. First party lists matter more than the news cycle suggested.

Running It In More Than One Place?

Then the frequency question is bigger than either platform can see on its own. That is worth looking at.

Ask About Overlap

What Is Retargeting, And Is It Different From Remarketing?

Retargeting means showing paid adverts to people who already interacted with your business, usually by visiting your website or app. Remarketing means the same thing. The two words are used interchangeably. There is a mild historical convention that remarketing meant email follow up and Google's product name, while retargeting meant display advertising. In practice, in 2026, if someone uses one word rather than the other it tells you nothing about what they mean.

One thing did change. Google announced in April 2025 that it would not remove third party cookies from Chrome after all, and shut down its replacement advertising interfaces that October. So cookie based retargeting still works in Chrome. Safari and Firefox have blocked third party cookies by default for years, which means a meaningful share of any audience was never reachable this way. The practical conclusion is unchanged: first party lists and platform owned data do the heavy lifting. Komplex runs this from Austin, Texas.

WHO RETARGETS WITH US

Retargeting For Ecommerce, Software, Home And Kitchen, And Local Services

The value of retargeting rises with how long the decision takes. Our accounts have covered footwear, skincare, telecom, software, and kitchens and home improvement.

See If We Fit

Ecommerce And DTC

Cart abandonment and dynamic product ads do most of the work. The single biggest risk is spending heavily to reach repeat buyers who needed no persuading.

B2B And Software

Small audiences, long cycles, and a real danger of showing the same fifteen people an advert two hundred times. Frequency capping matters more here than in any other sector.

Home And Kitchen

Weeks or months between first visit and purchase, often with a showroom visit somewhere in the middle. Retargeting keeps the brand present without demanding a decision.

Local And Professional Services

Small audiences again, and the conversion is often a call instead of a form. Suppression is harder to automate here and worth doing manually.

THE NUMBERS

What Actually Changed With Tracking

Three facts that matter for retargeting in 2026, two of which are widely reported wrongly. Each carries its date.

Google confirmed in April 2025 that it would not remove third party cookies from Chrome, reversing the plan it had announced years earlier.

Google shut down the Privacy Sandbox advertising interfaces in October 2025, ending the replacement it had been building.

Safari and Firefox have blocked third party cookies by default for years, so a meaningful share of any audience was never reachable that way.

HOW WE WORK

Segment, Suppress, Cap, Then Test

Four stages, and the first three are mostly about advertising to fewer people. That is where the improvement usually comes from.

Segment
Split the audience by what someone actually did and how recently. Cart abandoners and blog readers stop sharing a campaign and start getting different messages.
Suppress
Remove existing customers, recent converters, and anyone who has told you no. This alone often frees a meaningful share of the budget in the first month.
Cap
Set frequency limits across platforms rather than inside each one, so nobody is seeing the same advert from two accounts that cannot see each other.
Test
Hold back part of the audience, show them nothing, and compare. The gap is what retargeting added, and it is the figure we report and stand behind.

See What The Current Setup Is Doing

The audit comes first and costs nothing. It reads the audiences, the overlap, the frequency, and what the reported numbers would survive.

Start With The Free Audit

FREE RETARGETING AUDIT

Get A Free Retargeting Audit

We look at how the audiences are built, where they overlap, and what windows are set. Then whether existing customers are suppressed, how often one person is reached, and what the reported return would look like with a holdout behind it. The write up says what to cut, what to separate, and what the current numbers can fairly claim. There is no charge, the findings are yours whatever you decide, and the same audit sits inside our free marketing audit.

Your retargeting auditFree

Audience segments and how they overlap
Membership windows against your actual sale cycle
Customer and converter suppression
Frequency across platforms rather than inside one
Feed health where dynamic ads are running
What the reported return would survive

1
Tell us what you are running
The website, which platforms retargeting runs on, roughly what you spend, and how long a typical purchase takes.
2
We look
Read only access to the ad accounts and analytics goes deepest. Nothing gets changed, and an NDA is available on request.
3
You get the scorecard and a call
A written scorecard ranked by impact, then a call to walk through it.

Ready To Find Out What Retargeting Is Really Adding?

There is no fee and no obligation, and everything the audit finds belongs to you.

Get My Free Retargeting Audit

PRICING

How We Price Retargeting Work

Media management is billed as a retainer plus a share of ad spend, sized after a conversation about the account. We are open to other formats where one fits better. Retargeting has the lowest entry point of anything here. The audiences are small and warm, so the spend does not need to be large to do its job. That cuts both ways: it is also the easiest budget in an account to quietly overspend.

Where retargeting runs alongside paid search or paid social with us, it usually sits inside that work. The audiences and the accounts are already shared. As a standalone service it is quoted on its own. Creative refreshes are agreed before they start.

Send the account and tell us how long a typical purchase takes. Those two things shape the recommendation more than the budget does.

Get A Quote

What Clients Say About Working With Komplex

The Komplex team helped us reach new audiences and drive higher conversions.

Strategies that delivered results beyond our expectations.

Their expertise helped us scale efficiently, attract more customers, and drive measurable growth.

Komplex played a key role in growing our user base and optimizing our marketing efforts.

Common Questions About Retargeting And Remarketing

What is the difference between retargeting and remarketing?

Nothing meaningful. The words are used interchangeably. There is a loose convention that remarketing came from Google’s product naming and from email follow up, while retargeting meant display advertising. Nobody applies it consistently. If an agency tells you they are fundamentally different services, ask them to explain how.

Does retargeting still work now that cookies did not go away?

Yes, and the panic was overdone in both directions. Google announced in April 2025 that Chrome would keep third party cookies, then shut down its replacement interfaces that October. Safari and Firefox have blocked them by default for years, so part of your audience was never reachable that way. Nothing about that changes the sensible strategy, which is to lean on lists you own.

Are view through conversions real?

They are real events and they are not proof. A view through conversion counts someone who saw an advert, did not click, and converted later. Nobody can tell you whether the advert caused it. On retargeting audiences, where most people were already engaged, the majority of these conversions would probably have happened anyway. We report them. They do not carry the argument.

How do you know retargeting is actually adding sales?

By holding some of the audience back and showing them nothing. Whatever the advertised group does above the held back group is what retargeting added. It costs a little reach and it produces a number you can defend. Almost every account we look at has never done this, which is why the reported returns look so good.

What is the minimum budget?

Low, because the audiences are small. That is the appeal and also the trap. A small warm audience can absorb far more money than it should and keep reporting well. That is why capping and suppression matter more here than budget size. We would rather size it correctly than encourage you to spend more.

How do you avoid annoying people with frequency?

Caps set across platforms rather than inside each one, suppression of anyone who already converted, tighter membership windows, and creative rotation before the ads become wallpaper. Uncapped retargeting is the fastest way to make people dislike a brand they were otherwise fine with.

Which platforms do you run retargeting on?

Google Ads, covering display, YouTube, and search remarketing lists, and Meta for Instagram and Facebook. Both are accounts you already own. Where a client runs paid social with us, the Meta side is planned inside that work so the two are not bidding against each other.

Can you retarget without a big audience?

Yes, and small audiences are often the most valuable ones. The platforms enforce minimum audience sizes before a list can be used, so there is a floor. Below it, the answer is usually to widen the window slightly or to combine two related behaviours rather than to give up on the idea.

What is dynamic product retargeting?

Adverts that show the specific products someone viewed, generated automatically from your product feed. It works well for ecommerce with a large catalogue and it is completely dependent on feed quality. A feed with wrong prices or missing images will produce adverts that actively cost you sales.

How do you suppress existing customers?

By uploading customer lists and excluding them, and by using purchase events to remove people automatically. Where the conversion happens off site, such as a phone call, we check it manually. Suppression is unglamorous and it is usually the single biggest saving available in a retargeting account.

How long should a retargeting window be?

As long as your sale actually takes, which is rarely the default. Something bought in two days does not need a thirty day window. A kitchen bought over three months needs longer than any default offers. We set the window from your data rather than from the platform suggestion.

Does retargeting work for B2B?

It can, with care. The audiences are small, the cycles are long, and the risk of showing the same few people an advert hundreds of times is high. Done well it keeps you present through a long evaluation. Done carelessly it becomes an expensive way to irritate a buying committee.

How does this relate to our paid search account?

Search remarketing lists live inside paid search and adjust bidding for people who already visited, which is a different mechanic from showing them a display advert. Both use the same audiences. When one lead runs both, the audiences get built once and used properly rather than duplicated with slightly different settings.

Do you work with companies outside Austin, Texas?

Yes. Komplex is based in Austin with a presence in New York City and Reykjavik, and this work happens remotely by nature. Campaigns across countries get separate audiences and separate suppression lists. A customer in one market should not be paying to be advertised to in another.

Send The Account And We Will Tell You What Retargeting Is Really Adding

Read only access is enough. Back comes the audience overlap, the frequency, the suppression gaps, and an honest view of what the reported return would survive.

Get My Free Audit