The professional targeting data
Every campaign starts with a location, which Campaign Manager requires, and a choice of whether to reach members by their permanent location, their recent location or both. Everything after that is a professional attribute drawn from member profiles. LinkedIn organises the attributes into facets, and each facet can include or exclude values.
| Facet | Attributes | What it is good for | Watch out for |
|---|---|---|---|
| Company | Company name, industry, size, growth rate, category, followers, connections | Reaching named accounts or a type of employer | Size and industry are self-declared by the company page; small firms often sit in the wrong bands |
| Job experience | Job title, job function, seniority, skills, years of experience | Reaching the people who hold a role or a responsibility | Titles are matched narrowly and vary by company; functions and seniority are broader and more reliable |
| Education | Degrees, fields of study, schools | Niche professional audiences defined by qualification | Thin coverage outside well-known institutions |
| Demographics | Age, gender | Rarely needed for business buyers | Inferred rather than declared, and restricted in some regions |
| Interests and traits | Member interests, member groups, member traits (such as frequent travellers or job seekers) | Adding a topical signal to a professional audience | Interests are inferred from activity and are weaker than declared attributes |
| Audience templates | Pre-built combinations LinkedIn maintains | A quick start for common audiences such as decision makers or recent graduates | Opaque; check what is inside before relying on one |
Two facets carry most business-to-business targeting: company and job experience. The usual pattern is company attributes to describe the employer (industry and size, or a list of names) and job function plus seniority to describe the person. Job titles look precise and are tempting, but members write their own titles, companies invent new ones, and LinkedIn's title matching is narrow, so a title-only audience is often smaller than the real population it is meant to describe. Skills are the underused facet: a member who lists a skill has declared an interest in a topic, and skills broaden an audience in a way that stays relevant.
Building an audience: attributes, Matched Audiences and retargeting
Within a facet, selected values are combined with OR: a member in any of the selected industries qualifies. Across facets, the combination is AND when you use the option to narrow the audience further: the member must be in a selected industry and hold a selected function and sit at a selected seniority. Exclusions are applied last and remove members regardless of what else matches. Campaign Manager shows a forecast of the audience size and its make-up as you build, and that panel is the main tool for keeping the audience sensible. Audience expansion, which lets LinkedIn add members similar to those targeted, and the LinkedIn Audience Network, which extends delivery to partner sites and apps, are both options Campaign Manager may pre-select depending on the objective; check them at launch and leave them off until the core audience is proven.
Matched Audiences bring your own data into Campaign Manager. LinkedIn documents several sources:
- Website retargeting builds an audience from visitors recorded by the Insight Tag, defined by URL rules (everyone who visited the pricing page; everyone who visited any page). This is the LinkedIn form of RemarketingAdvertising to previous visitors or engagers., and it needs the tag to have been live for long enough to collect a population above the documented minimum.
- Contact lists are uploaded files of people, matched to members on the identifiers LinkedIn supports. In practice match ratesShare of the records in an uploaded customer or account list that an ad platform matched to its own users or companies. are well below the number of rows uploaded, because the email addresses in a CRM are work addresses and profiles are often registered with personal ones.
- Company lists are uploaded files of organisations, matched to company pages, and are the mechanism behind account-based programmes; the account-based marketing guide covers how to build and tier them.
- Engagement retargeting builds audiences from people who interacted with earlier campaigns: watched a video to a chosen point, opened or submitted a Lead Gen Form, engaged with a single image ad, attended an event, or interacted with the company page. These audiences are the glue between a top-of-funnel campaign and the one that follows it.
- Predictive audiences are generated by LinkedIn from a seed you supply (a contact list, a Lead Gen Form, a conversion), finding members who resemble the seed. LinkedIn retired its older Lookalike AudienceAudience similar to a seed list. product in favour of this, and the help centre describes which seeds are accepted and how the audience is sized.
The logic is the same as Meta's, which the custom audiences guide explains, with one difference that matters: LinkedIn's audiences build slowly because the member base is smaller and visits are less frequent. A retargeting pool that fills in a day on Meta can take weeks on LinkedIn, so plan the sequence before launching the first campaign, not after.
Audience size, exclusions and overlap
LinkedIn will not run a campaign whose audience falls below the documented minimum, and the forecast panel shows when you are close. The more common problem is the opposite: an audience built wide enough to launch that is then too vague to be worth the price of reaching it. Three checks keep the audience honest.
Does the forecast describe the people you meant? The panel shows the audience's distribution by function, seniority, industry and company size. If a campaign meant for finance decision-makers forecasts mostly entry level members, the function was too wide or seniority was not set.
Who should never see this ad? Exclude current customers by uploading them as a contact or company list so acquisition budget is not spent on retention. Exclude your own company and its employees. Exclude competitors by company name where the content would help them. Exclude people who have already converted on the campaign's form so they are not asked twice. Exclusions are also where lead quality starts: every unqualified person removed is a form submission that did not need to be filtered later.
Where do campaigns overlap? Two campaigns targeting overlapping audiences bid against each other and show both ads to the same members, raising frequency without adding reach. Where campaigns target by stage (a cold audience, an engaged audience, a converted audience) use exclusions so each member belongs to one stage at a time; a member who watched the awareness video should be removed from the awareness campaign and appear only in the consideration one. Frequency per member is reported in Campaign Manager and is the earliest sign that an audience is exhausted. The guide to structuring a LinkedIn Ads account plans these exclusions across the whole account, stage by stage, alongside campaign groups and names. Before launch, the LinkedIn Ads launch QA checklist confirms that exclusions are applied, matched lists have finished matching and audience expansion was set on purpose.
Narrowing an audience until it describes the buyer
A payroll software company targets the UK with job function "Human Resources" and company size 51–200 employees, and launches a Lead Gen Form campaign at £120 a day. The forecast shows an audience of about 410,000 members. After three weeks the campaign has produced 94 leads at £27 each, of which sales accepts 9: a cost per qualified lead of £280, against a target of £200. Reviewing the leads, most are HR assistants and recruiters. The campaign is rebuilt with seniority restricted to manager, director and VP, skills "payroll" and "HR operations" added, the current customer list excluded, and the audience falls to about 38,000. Over the next three weeks it produces 41 leads at £58 each, of which sales accepts 19: a cost per qualified lead of £125. The headline cost per lead doubled; the cost that mattered fell by more than half.
Lead Gen Forms: fields, routing and consent
A Lead Gen Form attaches to a Sponsored Content or sponsored messaging ad under the lead generation objective. When the member clicks the call to action, the form opens inside LinkedIn with fields pre-filled from their profile, and submitting it takes a couple of taps. Each form has a name and language, a headline and a short description that restate the offer, a set of profile fields, optional custom questions, optional consent checkboxes, a required link to your privacy policy, and a confirmation message with a call to action that can link to the downloaded asset, a page or a thank-you message.
Fields. LinkedIn documents the profile fields a form can request: name, email, phone, location fields, job title, company, company size, industry, function, seniority, education fields, and a work email field that, unlike the standard email field, is not simply the profile's primary address and in practice is usually typed by the member. Fewer fields means more submissions. The rule is to ask only for what the next step needs: a nurture sequence needs an email and a role; a sales call needs a work email, a company and a phone number. Custom questions, which LinkedIn limits in number, are where qualification happens; a single question such as "How many people does your company pay each month?" or "Which system do you use today?" separates buyers from browsers. Hidden fieldsA form field the visitor never sees that records values such as source, campaign and landing page, so each lead arrives in the CRM with its origin. can carry campaign identifiers through to the CRM so each lead's origin is known.
Consent. The privacy policy link is required, and custom checkboxes handle marketing consent where the law or your own policy requires an opt-in. Keep the checkbox unticked by default and worded plainly; a consented lead can be nurtured, an unconsented one can only be called about the thing they asked for.
Routing. Submitted leads are visible and downloadable in Campaign Manager for a documented retention period, after which LinkedIn removes them, so a manual download routine is a liability. LinkedIn documents native integrations with several CRM and marketing automation platforms and a generic route through automation tools, and the integration should be configured and tested with a live test lead before the campaign spends. A lead that sits in a spreadsheet for a week has gone cold; one that reaches the CRM in minutes can be called while the member remembers the ad.
Lead quality over lead volume
The form's pre-fill is a gift and a trap. Completion rates on Lead Gen Forms run far higher than on landing pages because there is almost nothing to type, and the ad platform's conversion rate (CVRShare of visits or clicks that convert.) and cost per lead both look excellent. The leads then reach sales, who find personal email addresses, job titles that do not match the offer, and people who tapped submit to get a document and have no intention of buying. The programme has optimised for the metric the platform reports and not for the one the business pays for. The glossary entry on CPA in LinkedIn Ads explains why the platform's cost per lead counts form submissions rather than buyers.
The correction is to define the programme's cost per acquisition (CPACost of each conversion.) one stage further along than the form: cost per sales-accepted lead, or cost per qualified opportunity, calculated from the CRM rather than from Campaign Manager. The CPA calculator turns spend and accepted leads into that figure, and the comparison between campaigns, formats and audiences should be made on it. The result is often that the campaign with the highest cost per lead has the lowest cost per opportunity, because a precise audience and a qualifying question filter out the people who would have been filtered out later at greater expense.
Three levers raise quality without killing volume: the audience (seniority and skills, exclusions of customers and juniors), the form (the work email field and one real question), and the offer (content that only a buyer would want; a procurement checklist attracts procurement people, a free template attracts everyone). Pull them in that order, because the audience change affects every campaign at once and the offer change needs new creative. The same choice between native forms and landing pages, judged on qualified rate, runs across every network in the guide to paid social for lead generation. To make that comparison as a controlled test, the native lead form test splits the audience between a Lead Gen Form and a landing page and reads both on cost per sales-qualified lead from the CRM.
Feeding leads into the CRM and nurture
A LinkedIn lead is the start of a sequence, not the end of a campaign. The integration into the CRM should create or update a record with the form fields, the campaign and ad that produced it, the consent status and the time of submission. From there two routes run in parallel. Leads that meet the qualification rule (right seniority, right company size, work email, a positive answer to the custom question) are routed to sales with a service level for first contact measured in hours. The rest enter a nurture sequence that offers more of what they came for and watches for the behaviour that marks a buyer: a visit to the pricing page, a second download, a reply. The lifecycle marketing guide covers how that sequence is designed; the LinkedIn-specific point is that the same people can be placed in an engagement retargeting audience so the nurture runs in the feed as well as in email.
Close the loop in the other direction too. Upload won customers as an exclusion list so they stop seeing acquisition ads. Upload closed-lost accounts as a separate audience for a re-engagement campaign later. And report customer acquisition cost (CACCost to acquire one new customer.) by campaign from the CRM, which is the only figure that lets LinkedIn be compared with other channels on equal terms; the LinkedIn Ads hub explains why the platform's own cost figures cannot be compared with Meta's or Google's directly.
Professional targeting comes at a price, and the LinkedIn Ads versus Meta Ads comparison helps you judge when that precision justifies the cost and when broader audiences perform better.
Before adding qualifying questions to a form, see the lead form length experiment, which shows how to measure the trade-off between lead volume and lead quality properly.
When the goal is leads, the recommended LinkedIn lead generation setup turns these targeting and form choices into one campaign plan, from the offer to same-day CRM sync.
When a campaign built on these facets spends little or nothing, the diagnosis page for a LinkedIn campaign that is not spending reads the forecast, the bid range and the delivery status to find the constraint.
When sales rejects most form leads, the troubleshooting page for low-quality Lead Gen Form leads splits acceptance by audience, form and offer to find where quality leaks.
Choosing between a Lead Gen Form and a landing page by offer, audience and sales capacity is the subject of the comparison of native lead gen forms and landing pages.
Common mistakes
- Targeting job titles alone. Titles are matched narrowly and written inconsistently; function plus seniority plus skills describes the same people more completely.
- Stacking facets until the audience is tiny. Every AND halves the audience; the forecast panel shows when precision has become absence.
- Leaving audience expansion and the Audience Network on by default. Both spend budget outside the audience you designed; prove the core first.
- No exclusions. Customers, employees, competitors and recent converters all see acquisition ads and cost money or leak content.
- Judging the form by submissions. Cost per lead rewards the easiest form and the broadest audience; cost per accepted lead rewards the right ones.
- Downloading leads by hand. Leads expire from Campaign Manager after the documented period and go cold in days; integrate before launch and test with a live lead.
- One audience for every stage. Cold, engaged and converted members need separate campaigns with exclusions between them, or frequency climbs and the sequence never happens.
