Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Google Ads & PPC › Display Campaigns

Display Targeting Options

Display targeting explained: audience segments, your data, content targeting, optimised targeting, and how to layer them for reach or precision.
Guideintermediate11 min readUpdated

How display targeting is organised

Google's own documentation for Display campaigns separates targeting into audiences and content, and the distinction is worth holding onto because the two behave differently in delivery. Audience targeting follows the person: if someone is in a chosen segment, the ad is eligible to show to them on any page in the Google Display Network (GDN) that the campaign's other settings allow. Content targeting follows the page: the ad is eligible on pages about a topic, on named placements, or on pages whose content matches chosen keywords, whoever is reading them.

Within a campaign, targeting is set at the ad group level. Several targeting methods in the same ad group narrow delivery, because an impression has to satisfy all of them; the same methods in separate ad groups widen it, because each ad group bids on its own. Demographic settings (age, gender, parental status, household income where available) are a third, lighter layer that can be applied to any ad group. On top of all of this sits optimised targeting, which is enabled by default on new Display campaigns and which treats the chosen segments as a starting point rather than a boundary.

The practical consequence is that "what is this campaign targeting" is rarely answered by the targeting tab alone. It is answered by the targeting tab, the optimised targeting setting, the exclusions, and the placement report read together.

Audience segments: in-market, affinity, life events, custom

Google builds audience segments from behaviour across its properties and the network. They are inferences about groups of people, and Google does not document how any individual is assigned to one.

In-market segments group people whose recent behaviour suggests they are actively considering a category of product or service. They are the closest thing display has to intent, and the natural first choice for consideration and prospecting campaigns that need to find new buyers rather than remind old ones.

Affinity segments group people by durable interests and lifestyles. They are much larger and looser, which suits awareness campaigns that want cheap Reach among a broadly relevant group, and suits direct response poorly.

Life events group people around moments such as moving house, getting married or graduating. They are useful where the product follows the event, and small enough that they are usually combined with other segments rather than run alone.

Detailed demographics go beyond age and gender to attributes such as parental stage, education or home ownership, as Google infers them.

Custom segments are built by the advertiser from interests or keywords, from URLs of sites people browse, or from apps people use. For Display campaigns, Google documents that keyword-based custom segments are matched on interests and purchase intentions; the stronger "people who searched these terms" behaviour is available on some campaign types and not others, and the YouTube audience targeting guide covers where search-term custom segments apply. A custom segment built from competitor and category URLs is often the best prospecting audience a display campaign has, because it describes the person's reading rather than Google's inferred category.

Your data: remarketing lists and customer match

"Your data" segments are built from people who have interacted with the business, and they are the part of display that most reliably pays back. The lists come from the Google tag or Google Analytics (site visitors, specific page viewers, converters, cart abandoners), from app events, from YouTube channel interactions where the channel is linked, and from customer match, where the advertiser uploads hashed contact data and Google matches it to signed-in users.

Remarketing on display is a precision tool with its own failure modes. Lists need enough members before Google will serve against them, and small lists with large budgets produce the high Frequency that annoys people and wastes money, so a frequency cap belongs on every remarketing campaign. Membership duration should match the purchase cycle: a few days for an impulse product, months for a considered one. Converters should normally be excluded from acquisition remarketing and moved into their own retention list with a different message. Customer match lists are the foundation for both: an existing-customer list to exclude from prospecting, and a lapsed-customer list to reactivate. The remarketing expertise page describes how the author structures these lists across campaigns. In search campaigns the same lists modify keywords rather than replace them, which the guide to audience segments in search campaigns covers.

Content targeting: topics, placements, keywords

Content targeting puts the ad where the subject is, regardless of who is reading.

Topics are Google's classification of pages into subject categories. They are coarse, large and cheap to reach, and they are useful when the product is tied to a subject (cycling kit on cycling content) more than to a person's profile.

Placements are named sites, pages, apps, YouTube channels or videos. Managed placements give the advertiser the most control over context and the least reach, and they are the right choice when brand context matters more than scale or when a few publishers reach the whole audience. Placement bids compete with everyone else buying that inventory through Google, so a good placement is not guaranteed to be affordable.

Keywords on display match the ad to pages whose content relates to the words, and Google documents that keyword matching on display is broader than on search. They are best treated as a content signal to combine with an audience, not as the search-style precision tool their name suggests.

Content targeting is also where brand suitability lives: the pages a brand will not appear on are excluded by the same mechanisms (topics, placements) that select the pages it will.

Optimised targeting and targeting expansion: what they do

Optimised targeting is enabled by default on Display campaigns. Google documents that it looks beyond the selected audience segments to find people likely to convert, using signals such as the keywords on the landing page and the characteristics of people who already converted. The chosen segments act as a starting hint rather than a fence. Google also documents that exclusions are still respected, so a negative audience or excluded placement holds even when optimised targeting is on. Targeting expansion was the earlier, narrower form of the same idea, which widened reach to similar audiences while keeping the chosen segments as the core; Google has consolidated towards optimised targeting over time, and the exact options shown in the interface depend on the campaign type and bidding strategy.

Whether to leave it on is a question about the campaign's job. For a campaign bidding on conversions with solid tracking and enough conversions a month to learn from, it generally finds converters the chosen segments would have missed, and the chosen segments stop describing where the money goes. For a remarketing campaign, a tightly briefed awareness buy, or any campaign whose point is to reach a specific audience, it should be off. In practice, the most common surprise in a display account is an audience report that shows most spend under "optimised targeting" or "expansion" rather than under any segment the advertiser selected.

Layering: reach vs precision

Each targeting method added to an ad group removes everyone it does not describe. That is the trade: the more precisely an ad group describes its audience, the fewer impressions it can win and the higher the cost per thousand impressions (CPM) tends to be, because the campaign is competing for a scarcer set of people. The reach and frequency calculator makes the trade visible before launch: a precise audience is small, so the same budget buys frequency rather than reach.

ObjectiveTargeting shapeOptimised targetingExpect
Awareness among a broad groupOne affinity or topic layer per ad group; demographics only if the product demands itOn, if bidding on conversions; otherwise offLarge reach, low frequency, low cost per impression
Prospecting for new buyersIn-market or custom segment per ad group; no content layerOn, with conversion tracking in placeModerate reach, the system expanding over time
Prospecting with contextIn-market or custom segment plus a topic or placement layer in the same ad groupOffSmall reach, higher cost per impression, cleaner placements
RemarketingYour-data list per ad group; frequency cap; converters excludedOffReach bounded by list size; frequency is the lever
Brand-safe reachManaged placements onlyOffReach limited to the chosen publishers; context fully controlled

Two rules keep layering honest. Put one idea per ad group, so the reports can show which audience worked. And layer within an ad group only when the extra precision is worth the lost reach, which is usually when context matters or when the audience segment alone has already proved too loose. When a layered campaign leaves its budget unspent, the diagnosis for a Display or video campaign that is not spending works through layering, list size, frequency caps, exclusions, bidding and approvals in order.

Illustrative example — not a real result

A boutique hotel group layers for precision, then loosens

A hotel group with four properties launches a display campaign with a single ad group targeting an in-market "hotels and accommodations" segment, optimised targeting on, and a £150 daily budget. After two weeks the placement report shows 61% of spend on mobile game apps and a cost per thousand impressions of £1.40, with 2 booking conversions. The team excludes app categories, splits the campaign into three ad groups — the in-market segment alone, a custom segment built from twelve competitor and travel-guide URLs, and the in-market segment layered with a travel topic — and turns optimised targeting off on the layered group. Over the next month the layered group reaches 38,000 people at £4.10 per thousand impressions and produces 11 bookings; the custom segment group reaches 120,000 people at £2.30 and produces 9; the in-market-only group is paused. The campaign is relaunched around the two surviving ad groups, with the custom segment carrying most of the budget because its cost per booking is lowest.

Exclusions and brand safety

Exclusions are the half of targeting that protects the budget. They apply on top of whatever targeting is chosen, and Google documents that they are respected even when optimised targeting is expanding delivery.

  • Audience exclusions: existing customers from prospecting campaigns, recent converters from remarketing, and any list that should not be shown the message (job applicants, staff).
  • Placement exclusions: named sites, apps and channels, built from the placement report. Account-level exclusion lists mean each campaign inherits the same list, which is how a mature account avoids rediscovering the same poor placements campaign by campaign.
  • App category and mobile app exclusions: where mobile apps make up a large share of low-quality impressions, excluding app categories wholesale is often the right first move.
  • Topic exclusions: subject areas the brand should not appear beside.
  • Content exclusions and inventory type: Google's own sensitive-content categories and the inventory tiers that trade reach for stricter standards; the exact categories are listed in the Google Ads help pages and change over time.

The display hub sets out why placement hygiene is the single most valuable routine in display management. The targeting guide's contribution is the order of operations: exclusions first, then targeting, then review. When display later moves to Display & Video 360, the exclusion lists built here become channels and an inclusion list there, as the tutorial on moving display from Google Ads to DV360 explains.

Reading the where-ads-showed and audience reports

Two reports tell the truth about a display campaign. The "where ads showed" (placement) report lists every site, app and channel that received impressions, with cost and conversions. Read it in the first days of a campaign to catch obvious waste, then on a schedule; sort by cost, then by impressions with no engagement, then by anything whose name is unfamiliar. Exclusions come straight from it.

The audience report shows spend and results by segment, and, crucially, how much went to optimised targeting or expansion rather than to the chosen segments. If that share is most of the budget, the campaign is no longer the campaign that was planned, which is acceptable only if the results are better. Demographic reports show whether delivery skews away from the intended group, and the reach and frequency view shows whether the budget is buying new people or repeating impressions to the same ones. What people see once the targeting has found them is the subject of the responsive display ads guide; the author's working method for both halves is on the Google Display Network expertise page.

When the placement report shows money going to apps and sites nobody chose, the low-quality placements diagnosis works from that report to decide whether to exclude, restrict or change what the bidding pursues.

Common mistakes

  • Leaving optimised targeting on by default and never noticing that it has taken most of the spend.
  • Stacking several targeting methods in one ad group until the audience is too small to deliver, then concluding display has no reach.
  • Treating display keywords like search keywords. They describe page content, broadly; they do not capture intent.
  • No frequency cap on remarketing. Small list, large budget, the same person every hour.
  • Never excluding converters from acquisition remarketing, so the budget chases people who already bought.
  • Ignoring the placement report. Exclusions are built from it; without the review, the list never grows.
  • One campaign for prospecting and remarketing, so budget, bids and reporting for two very different jobs are blended together.
  • Setting membership duration by default. A list that keeps people for the maximum period long after the purchase decision has passed is reach spent on the wrong moment.

Frequently asked questions

Should optimised targeting be left on?

It depends on what the campaign is for. For a conversion-bid campaign with reliable conversion tracking and enough volume, optimised targeting usually finds converters the chosen segments would have missed. For a remarketing campaign, a tightly scoped awareness buy, or any campaign where the audience itself is the point, switch it off, because it will spend beyond the list and the reports will no longer describe the audience you intended.

What is the difference between an in-market segment and an affinity segment?

In-market segments group people Google judges to be actively researching or comparing a type of product or service, based on recent behaviour, so they are closer to a purchase. Affinity segments group people by durable interests and habits, which makes them larger and better suited to awareness than to direct response. Google documents both as inferred groupings, not declared ones.

Why do my ads show on sites that have nothing to do with my targeting?

Audience targeting follows the person, not the page, so a person in a chosen segment will see the ad on whatever site they happen to be reading, including ones unrelated to the product. If the pages matter, add content targeting, apply placement and topic exclusions, and review the placement report regularly. Optimised targeting or targeting expansion can also widen delivery beyond the chosen segments unless it is turned off.

Terms used on this page

Tools to try

Frameworks

Where this page is used

Also related

Sources

  1. About targeting for Display campaigns — Google Ads Help
  2. About audience targeting — Google Ads Help
  3. Google Ads Help — Google Ads Help

Related