Performance
The performance marketing audit: 12 checks before you spend another dollar.
Most paid media problems aren't media problems. They're tracking, funnel and creative problems that show up as rising costs. Run these twelve checks before you increase a single budget.
When paid media performance slips, the instinct is to change something inside the ad platform: new bids, new audiences, a new campaign type. Sometimes that's right. More often, the cause sits outside the platform, in tracking, in the funnel, or in the creative. After more than a decade running and reviewing paid media for brands across Toronto and Canada, this is the checklist we work through before recommending any change in spend.
It's grouped into four areas. Work through them in order. There's no point optimizing creative against numbers you can't trust.
Tracking: can you trust the numbers?
1. Conversion tracking fires once, on the right event
Place a real test order or lead and watch what each platform records. Common faults: the purchase event firing on page reload, both a pixel and a tag manager tag sending the same conversion, or "lead" being counted on a button click rather than a completed submission. Compare platform-reported conversions against your back end (Shopify, your CRM, your order system) for the same period. A large gap is a finding, not a rounding error.
2. Server-side tracking and consent mode are set up properly
Browser restrictions and ad blockers mean a pure pixel setup misses conversions. Server-side options such as Meta's Conversions API and Google's enhanced conversions help recover signal. At the same time, consent has to be respected. Check that your consent banner actually controls what fires, that Google consent mode is configured rather than just installed, and that server-side events are deduplicated against browser events. Getting this wrong in either direction, over-collecting or under-reporting, is costly.
3. Attribution windows make sense for how people buy
Look at the attribution window and model each platform uses. A long view-through window can credit an ad for a purchase the customer was going to make anyway. A window that's too short can undercount a considered purchase with a long research phase. Align the windows with your actual buying cycle, and make sure everyone comparing channels knows which windows are in play.
Funnel: does the click turn into revenue?
4. You're measuring blended performance, not just platform ROAS
Each platform reports its own return, and the totals usually add up to more revenue than you actually earned. Track a blended view alongside them: MER (total revenue divided by total marketing spend) and blended CAC (total acquisition spend divided by new customers). If platform ROAS is climbing while MER is flat or falling, the platforms are likely claiming credit for sales that would have happened anyway.
5. Landing pages load fast on a phone
Test your top landing pages on a mid-range phone over a mobile connection, not on your office Wi-Fi. Use Google's PageSpeed Insights to check Core Web Vitals. Heavy hero videos, unoptimized images and stacks of third-party scripts are the usual culprits. Speed problems quietly inflate your cost per conversion on every channel at once.
6. The landing page matches the ad's promise
Click each of your top ads and read the page as a stranger would. Does the headline continue the ad's message? Is the offer the same? Is the next step obvious? Sending every campaign to the homepage is one of the most common and fixable leaks we see.
7. Lead capture is CASL-compliant and qualifies leads
Canada's anti-spam legislation sets rules for obtaining consent to send commercial electronic messages. If your forms pre-tick a marketing consent box, bury consent language, or don't record when and how consent was given, you have both a compliance risk and a weaker list. Check, too, whether your forms ask enough to separate good leads from poor ones. A cheap lead that never converts isn't cheap. This is general guidance, not legal advice, so involve counsel for anything material.
Channels: is the budget in the right place?
8. Campaigns aren't bidding against each other
Audience overlap is a bigger issue in Canadian accounts because the pools are smaller. Prospecting, retargeting and broad campaigns can end up chasing the same people, which raises costs and muddies results. Use the platforms' overlap tools, check that exclusions are in place, and consider whether fewer, broader campaigns would perform better than many narrow ones.
9. Branded search isn't flattering the account
Branded search campaigns often post excellent returns because the people searching your name were already coming. Report branded and non-branded search separately. Then decide, ideally with a controlled test, how much branded spend is protecting real revenue versus buying clicks you'd have earned for free.
10. Budget follows evidence, not habit
Lay out spend by channel and campaign against blended contribution, not platform-reported results alone. Ask what would happen if a channel's budget dropped by a meaningful amount for a few weeks. Holdout and geo tests, for example pausing spend in one region of Ontario and comparing it with a similar region, are imperfect but far more honest than last-click reports.
Creative: are people still paying attention?
11. Creative fatigue is being watched in small audiences
In a smaller market, the same people see the same ads sooner. Track frequency alongside click-through and conversion rates over time. When frequency rises and response falls, the creative is tired. Plan for more creative variety than a US playbook would suggest, and refresh on a schedule rather than waiting for results to collapse. For bilingual campaigns, make sure French creative is written for Quebec audiences, not translated as an afterthought.
12. You're testing ideas, not just variations
Swapping button colours and headline wording produces small, noisy results. Meaningful tests compare different ideas: a different problem, a different audience motivation, a different proof point. Keep a simple log of what you tested, what you expected and what happened, so learning accumulates instead of resetting every quarter.
What to do with the findings
Sort what you find into three groups. Fix now: broken tracking, compliance gaps, and anything that makes the numbers wrong. Test next: landing pages, budget shifts and creative ideas. Rethink: findings that point to a positioning or audience problem, which no amount of optimization will solve. That last group is a strategy question, and our piece on digital strategy vs. digital marketing explains why. If you're about to launch something new, pair this audit with our go-to-market framework.
If the audit reveals that nobody owns these decisions across channels, that's often a leadership gap rather than a skills gap. A fractional CMO or a strategy partner can help close it.
How Three Dots Brands approaches this
An audit like this is often where our work starts. It sits in the Think stage of our approach: diagnose before you prescribe. We review tracking, funnel, channels and creative, tell you plainly what's working and what isn't, then build the plan and help execute it. The person who audits your account is the person who presents the findings. For a quick first read, try our marketing diagnostic.
Frequently asked questions
What is a performance marketing audit?
A performance marketing audit is a structured review of your paid media and the systems around it: conversion tracking, attribution, landing pages, lead capture, budget allocation, audience setup and creative. The goal is to find where budget is being wasted or where the numbers can't be trusted before spending more.
How often should we audit our paid media?
A full audit typically makes sense once or twice a year, and whenever something major changes: a new agency, a new website, a tracking or consent change, or a sudden shift in costs. Lighter monthly checks on tracking accuracy, frequency and blended metrics catch most problems early.
What is the difference between ROAS and MER?
ROAS is revenue attributed by an individual ad platform divided by spend on that platform. MER, or marketing efficiency ratio, is total business revenue divided by total marketing spend. Platforms tend to over-credit themselves, so MER gives a more honest view of whether marketing as a whole is working.
Does CASL apply to lead generation forms?
If you plan to send commercial emails or texts to people who fill out your forms, Canada's anti-spam legislation generally applies. That usually means clear, unticked consent, identifying who is asking, and keeping a record of when and how consent was given. Speak with legal counsel for advice specific to your business.