The Analytics Setup Every Marketing Team Needs Before Running Paid Ads

Running paid ads without proper measurement infrastructure is the fastest way to spend budget without learning anything. You will get traffic. You might get conversions. But you will not know which ads drove them, whether the conversions are the ones that matter, or how the paid channel interacts with your organic funnel. Six weeks later you will have a spend report and no idea whether to scale or cut.

"The most expensive mistake in paid advertising is not creative that does not convert — it is measurement infrastructure that cannot tell you what is and is not converting. Before you spend your first dollar on paid, build the system that will tell you whether the dollar worked."

— Neil Patel, Co-founder, NP Digital and Ubersuggest, in a 2023 keynote on paid acquisition fundamentals

UTM taxonomy first

Every paid ad needs UTMs. Not just campaign-level UTMs, but ad-set and ad-level UTMs that let you distinguish performance by creative, audience, and placement. Build your UTM taxonomy before you write your first ad: source, medium, campaign name, content, and term where relevant.

The naming convention matters more than you think. If one team member uses "google-ads" as source and another uses "google_ads," your analytics platform will split the data across two rows and you will never see consolidated performance. Document the convention, use a URL builder that enforces it, and audit UTMs before any campaign goes live.

Conversion events

Define what a conversion is before you set up tracking, not after. For most B2B marketing teams, the conversion hierarchy looks like: micro-conversions such as email subscription or content download, then pipeline conversions such as demo request or free trial signup, then revenue conversions such as closed deal or paid subscription. Set up tracking for all three levels.

Paid platforms optimize toward the conversion event you tell them to optimize for. If you only track demo requests, the algorithm will find more demo requesters, some of whom will never buy. If you can pass closed deal data back to the platform via CRM integration or offline conversion upload, the algorithm can optimize toward the outcomes that actually matter to revenue.

Attribution window decisions

The default attribution windows in paid platforms are set to maximize reported conversion numbers, not to reflect your actual buying cycle. Set your attribution windows to match your sales cycle. For long B2B cycles, use 90-day or 180-day click windows where the platform allows it. Use view-through attribution sparingly as it typically overstates the impact of display and video.

Baseline metrics before launch

Document your baseline before any paid campaign starts: organic traffic volume, organic conversion rate, direct traffic volume, email subscriber growth rate from organic. Without baselines, you cannot distinguish the paid channel's actual contribution from organic trends that were already happening.

The minimum viable measurement stack

You need: a web analytics platform with UTM tracking and conversion event capture, UTMs on every ad with a documented naming convention, conversion events firing on the actual conversion action (not the thank-you page load), and a weekly reporting cadence that reviews spend, conversions, cost per conversion, and conversion rate by channel and campaign. Everything else is optimization that comes after you have the basics working cleanly.

📊By the numbers