LinkedIn Campaign Manager: A Setup Walkthrough for B2B Teams

LinkedIn Campaign Manager is not complicated, but it has three or four defaults that quietly spend your budget on things you did not intend, and one structural decision that is annoying to change later. This is the setup order that avoids both.

This is for a B2B team running LinkedIn ads for the first time, or inheriting an account that underperforms.

Install the Insight Tag Before You Spend Anything

The Insight Tag is LinkedIn's site-wide tracking script, and everything useful depends on it: conversion tracking, retargeting audiences, and the demographic reporting that tells you who is actually visiting.

Install it site-wide, not just on landing pages, and then wait. Retargeting audiences need time and traffic to populate; an audience that has been collecting for six weeks is usable, one collecting for three days is not.

This is the most common sequencing mistake: launching campaigns first and installing the tag later, which means the first month of spend generates no reusable audience.

Account Structure: One Campaign Group Per Objective

The hierarchy is account, campaign group, campaign, ad. Most teams overuse campaign groups and end up with fragmented budgets and no statistical significance anywhere.

A structure that works for a team spending under about $10,000 a month: one campaign group per objective (awareness, lead generation, retargeting) and within each, one campaign per audience, not per creative. Creative variants belong inside a campaign so LinkedIn can optimize between them.

Three campaigns with meaningful spend beats twelve with $200 each, where none accumulates enough data to learn from.

The Four Settings That Waste Budget

Audience Expansion, on by default. It lets LinkedIn add people "similar to" your audience, which defeats the entire reason you are paying LinkedIn's premium. Turn it off.

The LinkedIn Audience Network, also on by default. It serves your ads on third-party apps and sites outside LinkedIn, at lower cost and considerably lower quality. Turn it off until you have a working campaign, then test it separately so you can see its numbers in isolation.

Automated bidding, which is fine once you have conversion data and expensive before then. Start with manual cost-per-click bidding so you control the ceiling while you learn what a click is worth.

Over-stacked targeting. Each additional filter raises CPM and shrinks the audience, and below roughly 20,000 people costs climb sharply. Two or three filters, not six.

Matched Audiences Beat Native Targeting

The highest-return feature in Campaign Manager and the most underused. You can upload a list of company names or contact emails and target exactly those.

This outperforms LinkedIn's native targeting almost every time, because your list is built on something you know (a trigger, a filing, a filtered plan record) rather than something LinkedIn inferred from a profile.

Company lists match at a higher rate than contact lists, so upload companies when you can. Expect a match rate well below 100% and plan the audience size accordingly: a 500-company list may match to a few thousand targetable people, which is on the small side and will cost more per impression.

Conversion Tracking Worth Having

Define conversions at the point that matters to you, not at the point LinkedIn finds easiest to measure. A lead-gen form submission is a conversion LinkedIn can see. A qualified opportunity is the conversion you care about.

Set up the form submission in Campaign Manager because you need something for the platform to optimize against, and then track the real number downstream in your pipeline with the campaign tagged on the contact record.

Without that second half, Campaign Manager will report a healthy cost per lead forever while the leads qualify at 20%, and nobody will be able to argue either way when the budget review comes.

A Sane First Campaign

One campaign group, one campaign, a matched audience of 300 to 500 target companies, document ads with a genuinely useful artifact, manual bidding, expansion and the Audience Network both off. Roughly $3,000 over six weeks.

Measure meetings booked, not leads captured. If that produces nothing, the answer is not a larger budget. It is that this audience is reachable more cheaply another way, and the test has done its job.