In-market segments, Customer Match, Meta lookalikes and retargeting split by behaviour: the four targeting levers that consistently separate efficient spend from wasted budget in e-commerce PPC.
By Jack Goldsmith, Founder & Performance Marketer, Social Surge · 30 July 2026
What are the most effective audience targeting methods for online retail campaigns?
For Google Ads, in-market audiences layered onto Shopping and Search campaigns, and Customer Match lists built from your CRM data, deliver the strongest intent signal. On Meta, lookalike audiences seeded from high-value purchasers and retargeting segments split by engagement depth consistently outperform broad interest stacking. Across both platforms, audience exclusions (removing existing customers from acquisition campaigns) are the most overlooked lever for reducing wasted spend.
The most effective audience targeting methods for online retail campaigns depend on the platform and where a buyer sits in the funnel, but across the accounts we manage, four signals consistently separate efficient spend from wasted budget: intent-based audiences on Google, first-party data matching, lookalike seeding on Meta, and retargeting segmented by behaviour rather than just the fact of a site visit.
Google's in-market audiences group users based on their recent search and browsing behaviour: people who have been actively comparing products in a category over the past few weeks. According to Google Ads Help, in-market audiences can be applied to Search, Shopping, Display and Performance Max campaigns as observation or targeting layers.
In the accounts we manage, the starting point is always to apply in-market audiences to Shopping campaigns in observation mode, not targeting mode. Observation keeps your full reach while tracking which audience segments convert at better efficiency. Once a segment has 30 or more conversions, you have enough data to apply bid adjustments and push budget toward the higher-intent groups without arbitrarily restricting reach elsewhere.
Practically: if you sell cycling gear, the "Sports & Fitness: Cycling Enthusiasts" in-market segment typically shows a measurably lower CPA than your account average within four to six weeks of observation. Most accounts have this data sitting unused. It takes about ten minutes to add the audiences and four weeks to have something actionable.
Customer Match lets you upload your customer email list and Google matches it to signed-in Google users, creating a targetable (or excludable) audience from people who already have a relationship with your brand. Google requires at least 1,000 matched users for the list to become active in Search and Shopping, which typically means uploading 2,000-4,000 addresses depending on your match rate.
For online retailers, two uses deliver the most immediate return:
Data must be collected and shared in line with your privacy policy and UK GDPR obligations. If you are unsure about compliance, review your consent capture at checkout before uploading any list.
Meta's targeting approach for e-commerce has shifted considerably. Advantage+ audiences (Meta's AI-driven broad targeting) now outperforms manually stacked interest sets in most of the retail accounts we run. Meta's purchase signal is extensive enough that it can find buyers without being told "interested in cycling AND aged 35-54 AND owns an iPhone." Adding too many interest constraints shrinks the pool so far that the algorithm cannot optimise efficiently, which tends to push CPMs up and conversion rates down.
Where audience targeting still adds genuine value on Meta is in lookalike audiences seeded from your best customers. Build a 1-3% lookalike from purchasers whose AOV sits above your store average, not from your full buyer list. Seed quality matters more than seed size: a 500-person list of your top customers produces a better lookalike than a 10,000-person list that includes every one-off, low-value buyer.
Interest stacking still has a role for very specialist products. Fly fishing reels, gravel cycling frames and similar high-specificity items have enough niche identity that tight interest audiences can reduce early-funnel wasted spend before the algorithm builds sufficient purchase signal. For most mainstream sporting goods and outdoor products, Advantage+ audiences paired with strong creative will outperform.
See our guide to Meta Ads management for e-commerce for more on campaign structure and creative testing.
Retargeting everyone who visited your site is the lowest-quality version of this tactic. Someone who spent four minutes on a product page and added to cart is not the same signal as someone who bounced from the homepage in three seconds. Treating both identically wastes budget on people who were never close to buying and underspends on people who nearly did.
The segmentation structure that works in practice across the e-commerce accounts we manage:
Exclude recent purchasers from all retargeting tiers except dedicated retention campaigns. A customer who bought yesterday does not need a cart abandonment ad, and serving one erodes trust rather than driving revenue.
For the full retargeting structure including cross-channel coordination between Google and Meta, see our post on retargeting strategies that boost online store conversions.
Most discussion of audience targeting focuses on who to reach. Exclusions matter equally and are audited far less often. Three exclusions that apply in almost every e-commerce account:
A quick reference based on how we structure accounts at Social Surge:
Getting this right across Google and Meta simultaneously is one of the clearest differentiators between accounts that compound their returns and accounts that plateau. If your current setup has not been audited at the audience layer recently, our free PPC audit covers targeting structure, exclusion gaps and segmentation as part of the full review.
You can see the kind of results this approach produces in our case studies, and our pricing page has full service detail if you want to understand what managed PPC looks like for a business at your stage.
Yes. You can apply in-market audiences to Shopping campaigns in observation mode, which lets you track performance by audience segment without restricting reach. Once you have enough conversion data per segment, typically 30 or more conversions, use bid adjustments to push more budget toward high-converting groups.
Google requires at least 1,000 matched users for a Customer Match list to become active in Search and Shopping campaigns. For most retailers, this means uploading several thousand email addresses, since match rates typically run at 40-60% depending on how the data was collected.
For most e-commerce products, yes. Meta's Advantage+ audience uses its full pool of purchase data to find buyers without being constrained by interest selections. Interest layering adds value mainly for very niche products where the algorithm lacks sufficient signal to find buyers on its own.
Audience exclusions. Most accounts focus entirely on who to target and never audit who they are paying to reach repeatedly. Excluding existing customers from acquisition campaigns and recent purchasers from retargeting often reduces wasted spend faster than any new audience addition.
Book a free PPC audit, full account review, wasted-spend analysis and a clear growth plan, delivered in plain English.
Book Your Free Audit