Custom audiences

Custom Audiences let a company target ads using its own data: customer lists, website visitors, app users or specific behaviours. Unlike broad demographic or interest-based targeting, Custom Audiences reach exactly the people the company wants to reach.

This is particularly valuable for B2B, niche products or when the audience is well defined. Instead of paying to reach “men aged 25-45 interested in technology”, the advertiser can target ads at specific companies, job roles or people who have bought before.

Custom Audiences mean less wastage, higher relevance and often a better return on ad spend than broad targeting.

Why Custom Audiences work

Precise targeting reduces wastage

Broad targeting reaches many people, but most of them are irrelevant.

Example

Broad targeting (B2B company)

“CTOs at companies with 100+ employees in Sweden”

Problem

The ad reaches thousands of CTOs, but only a fraction are potential customers for that specific product.

Custom Audience

Upload a list of the people who are relevant. The list must be large enough to meet the platform’s minimum.

Result

Less wastage and a larger share of the budget going to the right audience.

The right people mean higher relevance

Ads that reach exactly the right people perform better on three measures.

The company’s own knowledge becomes audiences

With Custom Audiences, that information can be used for advertising.

How B2B benefits from Custom Audiences

In B2B, the audience is often small and specific. Custom Audiences are perfect for that.

Example

A company selling enterprise software to large manufacturers in the Nordics knows exactly which 200 companies are relevant.

With Custom Audiences, ads can be targeted at decision-makers at exactly those companies. Less of the budget goes to small businesses or the wrong industries.

There are several types of Custom Audiences

Customer Match starts from customer lists

Upload a list of email addresses, phone numbers or postal addresses. The platform matches the details against user accounts and shows ads to those particular people.

Uses

Existing customers (upsell/cross-sell)

Upload a customer list. Show ads for new products or upgrades.

Inactive customers (reactivation)

Upload a list of customers who have not bought anything for six months or more. Show ads with special offers designed to win them back.

Leads that have not converted

Upload a list of them from the CRM system. Nurture them with ads that build trust.

Event attendees

Upload a list from the webinar or event. Follow up with relevant ads.

Website Custom Audiences are based on website visitors

Audiences based on who has visited specific pages on the website.

Uses

Visited specific product pages

Show ads for exactly those products.

Visited the pricing page but did not buy

Show ads with a discount or case studies that build trust.

Read a blog post on a specific topic

Show ads related to that topic (they showed interest).

Spent a long time on the site

Use more aggressive remarketing.

Technically

Requires a pixel or tag on the website (the same as for remarketing).

Engagement Custom Audiences are based on social media interaction

Audiences based on interaction with content on social media.

Uses

Nurture engaged audiences

Follow up with more in-depth content or offers.

Video viewers

Show purchase ads to people who watched at least 75 per cent of the product demo video.

App Activity Audiences are based on app users

Audiences based on in-app behaviour, for companies with mobile apps.

Requires: An implemented SDK for app tracking.

Lookalike Audiences find similar audiences

The platform analyses a Custom Audience (existing customers, for example) and finds new people who resemble them.

Availability: Meta has Lookalike Audiences. At Google, Lookalike segments are available in Demand Gen campaigns, and campaigns with optimised targeting can use customer lists. Other platforms have their own variants, so check what is available when planning the campaign.

Uses

Find new customers who resemble existing ones

If current customers are CTOs at tech companies with 100 to 500 employees, Lookalike finds more people like them.

Increase reach while maintaining quality

Instead of broad targeting (“all CTOs”), Lookalike reaches people who resemble buyers.

Best for: B2C and B2B with larger audiences. Less valuable for very niche B2B companies, where the audience is already known.

Which platforms suit Custom Audiences?

Google Customer Match gives wide reach

What it is:
Upload customer lists (email addresses, phone numbers) and show ads in Google Search, Shopping, Gmail and YouTube. Within the European Economic Area (EEA), the lists cannot be used on Google’s partner websites and partner apps.

Uses: B2B and B2C. Particularly strong for upselling to existing customers.

Facebook Custom Audiences are strong for B2C

What it is:
Upload lists or create audiences from website visitors, app users or engagement.

Uses: Mainly B2C, though it works for some B2B companies, depending on the audience.

LinkedIn Matched Audiences are best suited to B2B

What it is:
Upload customer lists, create audiences from website visitors or target based on companies and contacts from LinkedIn.

Uses: B2B, especially enterprise sales. Less relevant for B2C.

Programmatic platforms (DSPs) suit larger companies

For larger companies: demand-side platforms such as The Trade Desk.

Advantage

Cross-platform Custom Audiences, advanced data integration.

Disadvantage

Requires large budgets and technical expertise.

How to use Custom Audiences strategically

Upselling to existing customers

Audience

Customers who have bought product A.

Ad

“Now that you have [Product A], add [Product B] for even better results.”

Result

A higher conversion rate (they already trust the brand) and a higher AOV (average order value).

Reactivating inactive customers

Audience

Customers who have not bought anything for six months or more.

Ad

“We miss you! Here is 20 per cent off to welcome you back.”

Result

It is cheaper to win back customers than to acquire new ones.

Nurturing leads

Audience

Leads in the CRM that have not converted yet.

Ad

Content that builds trust, such as case studies, white papers and customer reviews.

Result

The ads move leads from awareness through consideration to decision.

Targeting selected companies with account-based marketing (ABM)

Audience:
Specific target companies (target accounts).

Example

An ad for decision-makers at Acme Corp:
“Acme Corp: see how companies like yours solve [problem] with our solution.”

Result:
Extremely high relevance. The cost per lead is much higher, but so is the close rate.

Excluding customers who have already bought

Audience

People who have already bought.

Strategy

Upload a customer list as an exclusion from acquisition campaigns.

Why

Do not waste budget trying to sell to people who have already bought (unless it is an upsell).

Using purchase history for cross-selling

Audience:
Customers segmented by what they have bought.

Result:
Relevant, personal marketing that increases customer lifetime value.

How to automate with data integration

How the CRM connects to the ad platforms

To automate Custom Audiences, the CRM can be integrated with ad platforms.

Advantage:
No manual uploads, and the lists are always up to date.

Uses

Marketing automation can drive Custom Audiences

Platforms such as HubSpot, Marketo and ActiveCampaign can update Custom Audiences based on behaviour.

Example

Advantage:
Dynamic audiences that update based on behaviour in real time.

How to comply with the GDPR when handling data

Custom Audiences require a legal basis under the GDPR

Custom Audiences are built on personal data (email addresses, phone numbers). Anyone uploading a list needs a legal basis under the GDPR. The European Data Protection Board (EDPB) identifies two possible bases for targeted advertising on social media: consent or legitimate interest. Legitimate interest requires a balancing test, and consent is the safest route. Google requires consent where the law or Google’s policy demands it. The platform and the advertiser may be joint controllers.

Consequence:
Some customers cannot be included in Custom Audiences. Respect opt-outs.

How to manage data security

This is what you should do

Hashing

Email addresses and phone numbers must be hashed, either by you before uploading or automatically by the platform. Hashing is pseudonymisation, so the data is still personal data.

Restricted access

Only authorised staff have access to customer lists.

Regular clean-up

Remove old and outdated data from lists.

How to handle opt-outs

How to adapt creative for Custom Audiences

Tailor the message to the audience

With Custom Audiences, the message can be personalised.

Example

Existing customers

“Thank you for being a customer. Here is something exclusive for you.”

Inactive customers

“We miss you. Welcome back with 20 per cent off.”

Specific industry (ABM)

“Tech companies like yours solve [problem] with our solution.”

Advantage

Higher relevance means higher engagement.

Use social proof

For Custom Audiences who already know the brand, social proof works well.

Why:
It builds trust further.

Case studies and testimonials build trust in B2B

For B2B Custom Audiences (especially ABM), case studies work.

Why:
B2B decisions require trust. Case studies from similar companies reduce the risk.

How to measure and optimise Custom Audiences

How to track KPIs for Custom Audiences

Match rate:
The proportion of the uploaded list that the platform matches.

Usually some addresses do not match any account on the platform, so the list shrinks during matching.

CTR

Expected to be higher than with broad targeting (a more relevant audience).

Conversion rate

Expected to be higher than for new audiences (they already know the brand).

CPA

Expected to be lower (less wastage).

ROAS

Expected to be higher (better-quality traffic).

Compare the segments

Which Custom Audiences perform best?

Optimise:
Increase the budget for the best-performing audiences.

Keep customer lists clean and up to date

Problem:
Old lists with inactive email addresses reduce match rate and performance.

How to avoid common mistakes with Custom Audiences

Mistake 1: The lists are too small

Problem

The advertiser uploads a list of 50 email addresses to Google Customer Match (which requires at least 100 active members).

Result

The ads are not shown.

Solution

Check the platform’s minimum requirements. Combine lists if necessary.

Mistake 2: The lists are not segmented

Problem

The advertiser uploads an “all customers” list and shows the same ad to everyone.

Result

Low relevance (new customers and customers of five years have different needs).

Solution

Segment into new customers, active customers, inactive customers and high-value customers. A different message for each segment.

Mistake 3: The GDPR is forgotten

Problem

The advertiser uploads lists without checking the legal basis.

Result

A GDPR breach, potential fines, damage to the brand.

Solution

Only include people covered by a documented legal basis.

Mistake 4: Converted customers are not excluded

Problem

Acquisition ads are shown to people who have already bought.

Result

It wastes budget and confuses customers.

Solution

Exclude existing customers from acquisition campaigns (or move them to upsell campaigns).

Mistake 5: The lists are static

Problem

The advertiser uploads a list once and never updates it.

Result

The lists become outdated. New customers are missing, and old customers who should be excluded are still there.

Solution

Automate via a CRM integration or update manually every month.

How Custom Audiences fit into a complete strategy

Combine with layered targeting

Combine Custom Audiences with other targeting methods.

Example

Result:
Only CTOs at these specific companies see the ad. The advertising becomes extremely focused.

How Custom Audiences are used across the whole funnel

Top of funnel (awareness)

Lookalike Audiences (find new people who resemble customers).

Middle of funnel (consideration)

Website Custom Audiences (remarketing to visitors).

Bottom of funnel (conversion)

Customer Match with prospects from the CRM (nurturing leads).

Post-purchase

Customer Match with existing customers (upsell/cross-sell).

How we go from data to results

Custom Audiences require careful data handling but can deliver good profitability once the basics are in place.

The process has eight steps

Contact us about Custom Audiences

Carnaby helps you build a Custom Audiences strategy, from data segmentation to platform selection and creative adaptation.

Contact us →