Loading tutorials…
Loading tutorials…
Remarketing is the lowest-friction conversion lift in the platform — but only when the audiences are built cleanly. Stale lists, leaky audience definitions, and missing exclusions kill the lift. Here is the setup specialists actually use.
Who this is forOwners with existing Google Ads campaigns and at least 30 days of site traffic, ready to add a remarketing layer. If your account has been running for 60+ days without remarketing, you are leaving meaningful revenue on the table.
What you'll need
Step 1
Tools → Audience Manager → Your data sources. Verify Google Ads tag, GA4 link, and Customer Match upload are all "Active" with recent data.
Click Tools (wrench icon) → Shared Library → Audience Manager → Your data sources.
You should see at least three sources listed: "Google Ads tag," "Google Analytics 4 (Web)," and (optionally) "Customer Match."
Google Ads tag status: should be "Tag is active and collecting data." If it says "Tag inactive" or "No recent users," your remarketing tag is not firing. Go fix the tag installation before building audiences.
GA4 source: should show recent users and the GA4 property name. If it shows "No linked GA4 property," return to Tools → Setup → Linked accounts and link GA4 with audience sharing enabled.
Customer Match: only available if you have uploaded a customer list. We will set this up in Step 3.
Click into "Google Ads tag" → check that "Remarketing" is set to ON. Some accounts have remarketing disabled by default in Tag settings — this kills audience building before it starts.
Step 2
GA4 → Admin → Audiences. Build audiences in tiers: All visitors (540-day), Engaged visitors (90-day), High-intent (30-day), Converters (90-day, for exclusion).
Most remarketing setups have a single audience: 'all site visitors, 540 days.' This is too broad to be useful. Build a tier structure instead.
Open GA4 → Admin (gear icon, bottom left) → Audiences.
TIER 1 — All site visitors (540-day): create an audience named "All visitors 540d." Definition: include users whose "session_start" event occurred in the last 540 days. This is your top-of-funnel awareness audience.
TIER 2 — Engaged visitors (90-day): "Engaged 90d." Definition: include users who triggered "user_engagement" event in the last 90 days AND visited 2+ pages. This filters out one-page bounces.
TIER 3 — High-intent visitors (30-day): "High intent 30d." Definition: include users who viewed a product page, added to cart, or visited a pricing page in the last 30 days. Specific to your site's intent signals.
TIER 4 — Cart abandoners (14-day): "Cart abandoners 14d." Definition: triggered "add_to_cart" but NOT "purchase" in the last 14 days. Use both add_to_cart and purchase as inclusion/exclusion events.
TIER 5 — Converters (90-day, for exclusion): "Converters 90d." Definition: triggered "purchase" or your lead conversion event in the last 90 days. You will USE this as an exclusion on most remarketing campaigns so you stop spending on people who already bought.
Save each audience. They take 24-48 hours to populate fully in Google Ads. Most audiences need ~1,000 users before they are usable for ad targeting.
Step 3
Tools → Shared Library → Audience Manager → Segments → "+" → Customer list. Upload existing customer emails (hashed automatically). 5K minimum for some uses.
Click Tools (wrench icon) → Shared Library → Audience Manager → Segments. Click the blue '+' button → 'Customer list.'
Customer Match lets you upload a CSV of customer emails (and/or phone numbers, addresses, mobile device IDs). Google hashes them locally before upload — your raw data does not leave your browser.
Name the list specifically: "Customers - All Time" or "Customers - 90d" or "VIP Customers - LTV>$500." Each segmentation needs its own list.
Upload format: CSV with one column per data type (Email, Phone, etc.). Each row is one customer. Headers in row 1. UTF-8 encoding.
Match rates vary: typical match rate is 40-70% (Google can only match emails to people with Google accounts using that email). 80%+ match rates usually mean enterprise-quality customer data.
Minimum sizes: Search remarketing (RLSA) needs 1,000 active users. Display/YouTube needs 100 active users. If your customer list is below 1K, RLSA will not be usable — you will need to combine it with site audiences.
Refresh customer lists at least monthly. New customers get added, churned customers get removed. Most teams automate this via the Google Ads API or a CSV export from their CRM.
Step 4
Open your Search campaign → Audiences → "+ Add audience" → Observation mode. Add your audience tiers. Then set bid adjustments per audience.
RLSA = Remarketing Lists for Search Ads. You apply audience lists to existing Search campaigns to bid differently when an audience member searches for your keywords.
In your Search campaign → Audiences (left navigation) → "+ Edit audience segments" → Observation (NOT Targeting).
CRITICAL: pick OBSERVATION mode, not Targeting. Targeting restricts the campaign to only show ads to people in the audience — usually not what you want. Observation lets the campaign run normally but tracks performance by audience.
Add your audience tiers: "Customers - All Time," "High intent 30d," "Cart abandoners 14d," and "Converters 90d."
Save. Wait 14 days to collect performance data per audience.
After 14 days, review the Audiences tab. Look at conversion rate per audience. Apply bid adjustments: +25-50% bid on high-intent audiences with proven conversion lift, -50-90% bid on converted audiences if you do NOT want repeat-purchase, or use converters as an exclusion entirely.
Common RLSA bid structure: Cart abandoners +40%, High intent +25%, Engaged visitors +10%, Converters -90% (effectively excluded) or excluded.
Step 5
Tools → Setup → Business data → "+ Data source" → Display & Video 360 / Google Ads dynamic ads feed. Connect your Merchant Center product feed.
Dynamic remarketing shows visitors the specific products they viewed on your site — not generic brand ads. For e-com, this drives 2-3x the CTR and conversion rate of static remarketing.
Prerequisite: a Google Merchant Center account with a verified product feed.
In Google Ads, click Tools → Setup → Business data. Add a new data source pointing to your Merchant Center feed.
Edit your Google tag to send "ecomm_pagetype," "ecomm_prodid," and "ecomm_totalvalue" parameters on product pages, cart pages, and confirmation pages.
Easiest path is via GTM: install the "Google Ads Remarketing Tag" template, set Conversion ID to your account ID, and add the dynamic parameters as variables that pull from your dataLayer.
In Google Ads, create a new Display campaign (Campaigns → "+ New campaign" → Sales/Leads → Display).
Set the campaign type to use "Dynamic ads" and connect your Merchant Center data source.
Build a responsive display ad. The product images and details will populate dynamically based on which products each viewer looked at.
Target the campaign to your remarketing audiences: Cart abandoners, High intent 30d, etc. Use Audience Manager combinations for tight targeting.
Step 6
Audience Manager → audience combinations. Exclude Converters from most remarketing campaigns. Set frequency caps. Add brand-safety exclusions on Display.
EXCLUDE CONVERTERS on most remarketing campaigns: most accounts do not want to keep showing ads to people who just bought (unless you sell repeat-purchase products with short repurchase cycles).
In your remarketing campaign → Audiences → Edit → scroll to "Exclusions" → add "Converters 90d" as an exclusion.
For repeat-purchase products (consumables, subscriptions), use a shorter conversion exclusion window — e.g., 7 days for consumables you expect to be repurchased monthly.
FREQUENCY CAPPING: Campaign settings → Additional settings → "Frequency capping." For Display remarketing, cap at 3-5 impressions per user per day. Without caps, the same visitor can see your ad 20+ times in a day — annoying and budget-wasteful.
BRAND SAFETY (Display remarketing only): Campaign settings → Content → "Content suitability." Pick "Limited inventory" for safest placement quality. Add specific website exclusions for sites you do not want your brand near (Tools → Shared Library → Placement exclusions).
DEVICE EXCLUSIONS: if your remarketing is on Display, consider excluding "TV screens" inventory unless you specifically want it. Apps-only is also worth considering if mobile-app inventory has historically not converted.
Step 7
Launch the remarketing campaign. After 7 days, review per-audience CTR and conversion rate. Refresh audience definitions monthly. Refresh customer lists weekly to monthly.
Publish the campaign. Remarketing campaigns usually start serving within 2-4 hours of approval since the audiences are already built.
Day 7 review: check CTR per audience and conversion rate per audience. Healthy display remarketing CTR is 0.5-1.5%; healthy RLSA CTR matches or exceeds your non-remarketing Search CTR.
If any audience has CTR below 0.2% on display, it is probably overserving — pause it and rebuild with tighter recency or higher intent thresholds.
Monthly: review audience sizes. Audiences shrinking month-over-month either reflect declining site traffic or audience hygiene issues. Audiences ballooning may be too broad.
Refresh Customer Match lists at the cadence your business changes: weekly for fast-growing SaaS, monthly for most e-com, quarterly for established service businesses.
Quarterly: review audience definitions in GA4. As your site adds/removes pages or as customer behavior shifts, audience definitions need updating to stay accurate.
Common mistakes
One generic "all visitors 540-day" audience
What goes wrong: You spend equal money trying to convert someone who bounced off your homepage 17 months ago and someone who abandoned a cart yesterday. The yesterday person converts; the 17-month person does not. Mixing them in one audience drowns the signal.
How to avoid: Build a tiered audience tree (All visitors / Engaged / High intent / Cart abandoners / Converters). Bid differently per tier. Exclude Converters from most campaigns.
Not excluding converters from remarketing
What goes wrong: You keep showing ads to people who already bought. Annoys customers (especially for one-time purchases). Wastes 20-40% of remarketing budget. Inflates ROAS reports because the 'remarketing-attributed' conversions include repeat buyers who would have come back organically.
How to avoid: Always add "Converters 90d" (or appropriate window) as an exclusion on remarketing campaigns. Exception: legitimate repeat-purchase products with a clear repurchase cycle, where you actually want to re-engage at the right interval.
No frequency capping on display remarketing
What goes wrong: Without caps, the same visitor can see your ad 15-30 times per day across the Display Network. They start associating your brand with intrusive advertising. CTR drops, brand sentiment drops, budget gets eaten by uninterested impressions.
How to avoid: Campaign settings → Additional settings → Frequency capping → set to 3-5 impressions per user per day for most display remarketing.
RLSA in Targeting mode instead of Observation
What goes wrong: Targeting mode restricts Search ads to only show to people in your audience. This kills all new-customer acquisition — you can only convert people who already visited your site. New campaigns silently get 80% smaller.
How to avoid: Always use Observation mode for RLSA unless you have a deliberate strategy reason for Targeting. Observation lets the campaign serve normally while letting you bid adjust by audience.
Audiences too small to serve
What goes wrong: Some audience minimums: 1,000 active users for RLSA, 100 active users for Display/YouTube, 1,000+ for Customer Match in most cases. If your audience is below the threshold, it does not show up as usable in campaigns and you silently lose remarketing coverage.
How to avoid: Check Audience Manager → Segments → membership column. Any audience under the minimum needs either a broader definition (longer recency window, looser criteria) or more time to accumulate.
Static remarketing on e-com when dynamic remarketing is available
What goes wrong: Static remarketing shows the same generic brand image to everyone. Dynamic remarketing shows the specific product the viewer looked at. For e-com, dynamic typically doubles CTR and triples conversion rate. Running static when you could run dynamic leaves significant revenue on the table.
How to avoid: Set up Merchant Center, connect the feed to Google Ads, install the dynamic remarketing tag parameters, and run dynamic remarketing campaigns. It is 4-6 hours of one-time setup that pays back monthly.
Recap
Done — what's next
How to create a Performance Max campaign (without losing control)
Read the next tutorial
Hand it off
Remarketing audience hygiene is one of the most underrated places where specialist work pays back. The setup is technical but the ongoing audience curation — keeping recency tiers fresh, exclusions current, dynamic feeds clean — is where DIY accounts quietly drift. EverestX Google Ads specialists who manage remarketing typically charge $200-600/mo at $14-16/hr, depending on audience complexity.
See specialist rates
Customer Match uses customer data you upload (emails, phones) — it does not require visitors to come to your site. Site remarketing uses the Google Ads tag to track site visitors and put them in audiences. Most accounts use both: Customer Match for retention and lookalikes, site remarketing for prospecting and cart-abandonment.
Match the window to your buying cycle. E-com impulse purchases: 7-30 days. SaaS evaluations: 30-90 days. B2B with long sales cycles: 90-180 days. Real estate or auto: 180-540 days. Beyond the buying cycle, conversion rates drop sharply and you are paying to nag uninterested people.
Both, if budget allows. Display remarketing is cheaper per impression and good for static brand presence. YouTube remarketing is more engaging (video format) and works especially well for high-consideration purchases. Start with Display, add YouTube when you have a 30-second creative ready.
Below 1,000 monthly visitors, site remarketing audiences are too small for RLSA (1,000 minimum) and barely usable for Display (100 minimum). Focus on Customer Match if you have an existing customer list — Customer Match has lower minimums and does not depend on site traffic.
Yes. Your privacy policy must disclose that you use Google Ads remarketing tracking, link to Google's ads policy, and explain how users can opt out (link to Google's Ads Settings). Without this disclosure, you may violate Google Ads policy (and possibly GDPR/CCPA depending on jurisdiction).
Yes — both are part of the Display Network and dynamic remarketing reaches both placements. PMax campaigns also reach Gmail and Discover inventory automatically. You do not configure these separately; they come included with Display/PMax campaigns targeting your remarketing audiences.
Google Ads
Performance Max is Google's most opaque ad product — and for accounts with the right ingredients, also one of the highest-ROAS. The catch: every default Google offers is tuned for Google's revenue, not yours. Here is the build that gives you back control.
Google Ads
Linking these two correctly is what makes Google's whole stack actually function as a stack. Done wrong, you'll have two systems that disagree about reality for years. Here's the right setup.
Google Ads
Conversions firing in your store but Google reporting zero? You're not alone. This walks through the right setup path for your stack — including the GA4 import workflow that 80% of DIY tutorials skip.
Google Ads
CPC creeping up while conversion rate stays flat means you're paying more for the same outcome. These six tactics drop CPC 20-40% on most accounts. Some take 1 hour. Some take 1 month. All matter.
Google Ads
DIY Google Ads is a great idea — until it isn't. This is the honest framework: when the cost of self-managing exceeds the cost of hiring, and how to tell which side you're on.