What Does a Consultant Check When Auditing Your Paid Media Mix?
Updated: 3 hours ago
Short answer: A paid media mix audit checks whether your budget is in the right channels, in the right amounts, for the right jobs. I start with tracking, because bad data makes every other finding suspect, then look at how platforms overlap and double count, whether each channel is past the point of diminishing returns, and whether awareness spend is being undercounted. Inside each channel I review targeting, bidding, creative and landing pages. The output is a recommended budget split with the reasoning behind it, plus a list of fixes ranked by impact.
Most companies build their media mix one channel at a time. Google Ads came first, then Meta, then someone suggested LinkedIn or YouTube, and now there are five accounts, two agencies and nobody looking at how the pieces fit. Each channel's report says it is doing well. The total doesn't add up.
I have worked with more than 100 companies between $5 million and $500 million in revenue, and I built Upwork's first enterprise marketing mix model across an $88 million annual media portfolio. I run Google Ads accounts myself, so I look at these audits as someone who has to make the changes work. Here is what I check.
1. Is the tracking accurate?
This always comes first. I check whether conversions are defined the same way in every platform, whether any are double firing, whether offline conversions or CRM data flow back to the platforms, and whether GA4 roughly matches what your sales or order system says. I often find conversion actions counting page views or form starts as leads, or purchases being counted twice.
If the tracking is off, the platforms optimize toward the wrong thing and every report you read is skewed. Fixing it is often the single biggest improvement in the whole audit.
The cost of bad tracking compounds. In Disruptive Advertising's review of 2,167 Google Ads accounts, 42 percent had no conversion tracking at all, and every 10 percent increase in wasted spend raised cost per conversion by 44 to 72 percent. A $200 lead can become a $300 lead before anyone notices.
2. Are the platforms taking credit for the same customers?
Almost always, yes. Each platform takes credit for the same customer. A buyer who saw a Meta ad, clicked a Google search ad and then opened an email can be claimed by all three. I compare total platform-reported conversions to actual orders or closed deals. When the platforms claim well above your real total, I know the reported ROAS on some channels is inflated.
This shows up in the research too. In 15 large advertising experiments at Facebook, published in Marketing Science in 2019, standard measurement methods were often far off, and in about half of the studies the estimated purchase lift was off by a factor of three. The MMM firm Analytic Partners reports that last-click attribution overstates the role of clickable channels by 2 to 10 times on average.
Branded search and retargeting are the usual suspects. They look efficient because they reach people who were already going to buy. I check how much of their volume is incremental by looking at what happens when spend changes and at brand search trends over time.
The best-known test of this came from eBay. In a large field experiment, eBay turned off its brand search ads and kept 99.5 percent of that click traffic through its free listings.
3. Is awareness spend being undercounted?
Upper-funnel channels like YouTube, connected TV, podcasts and prospecting on social rarely get the last click, so click-based reports undervalue them. Companies cut those channels, see no immediate drop, and then watch branded search and direct traffic decline a few months later.
Nielsen estimates that a brand loses about 2 percent of future revenue for each quarter it stops advertising, and that recovery can take three to five years.
For accounts with enough history, a marketing mix model is the best way to estimate what awareness spend actually contributes. For smaller accounts, I look at lagged relationships between awareness spend and branded search or direct traffic. My post on marketing mix modeling vs. attribution explains why the two methods give different answers.
4. Is each channel past the point of diminishing returns?
Every channel gets less efficient as you spend more. The first dollars reach your best prospects, and later dollars reach people who are less likely to buy. I look at how cost per acquisition has moved as spend changed in each channel. Often one channel is saturated while another still has room, and moving budget between them improves results without spending more. I cover this in How to Scale Paid Media Without Blowing Up Your CAC.
5. Does each channel have a clear job?
Every channel should have a defined role: capture existing demand, create new demand, retarget, or support sales. I check whether the targeting, creative and success metrics match that role. A prospecting campaign judged on last-click ROAS will always look bad. A branded search campaign judged as a growth driver will always look good. Mismatched roles lead to bad budget decisions.
6. What do I look for inside each channel?
In Google Ads, I review search terms, match types, negative keywords, bidding strategy, Performance Max settings and asset quality, and whether low-volume keywords that bring in your best customers are being cut by automation. In Meta, I look at audience structure, creative fatigue, frequency and how Advantage+ campaigns are set up. In LinkedIn, I look at targeting precision and cost per qualified lead, which matters more than cost per lead in B2B. I have separate posts on Performance Max audits, Meta Ads audits and LinkedIn Ads audits for B2B.
I also check landing pages. A lot of what looks like a media problem is a page that loads slowly, asks for too much, or doesn't match the ad.
7. Are outside factors explaining the results?
Before I blame a channel or an agency, I check seasonality, competitor activity, pricing changes, promotions, product launches and, for some categories, weather. A drop in conversions every summer is not necessarily a performance problem. A sudden rise in cost per click may mean a competitor entered the auction. Tools and AI can flag the numbers, and I use AI and Supermetrics to speed up the data pulls, but separating these causes takes someone who knows your market.
8. Are the fees and incentives set up right?
I look at how your agencies are paid. A percentage-of-spend fee rewards spending more whether or not it is working. I also check who is actually working on the account. Many companies are paying senior rates for work done by junior staff. I wrote about this in Is Your Agency Wasting Your Ad Budget? An audit is also a good step before switching agencies, so you know what to ask the next one to fix.
9. What does a paid media mix audit cost?
A Google Ads audit starts at $3,500. A multi-channel paid media audit starts at $5,500, scaled by the number of channels, campaigns and how far back we look. A full Marketing Performance Audit, which adds positioning, organic channels and unit economics, starts at $15,000. An audit plus a marketing mix model starts at $20,000. If you want me to manage paid search afterward, that is a flat fee starting at $2,750 a month plus a modest setup fee, never a percentage of spend. Hourly work is $475. See the pricing page for details.
Weigh that against what the audit can move. McKinsey estimates that 15 to 20 percent of marketing spend can be freed up through better measurement. If you spend $50,000 a month on paid media, that is $90,000 to $120,000 a year. A $5,500 audit that finds even a fraction of it pays for itself many times over, and the savings repeat every year.
Next step
If you mostly run Google Ads, the free Google Ads Efficiency Scorecard will show where the biggest gaps are. You can also read more about my marketing audits and Google Ads audit, or book a free 30-minute call to talk through your mix.
Related: How Much Does a Google Ads Audit Cost? | Should You Audit Before Switching Agencies? | Performance Max Audit | Marketing Mix Modeling vs. Attribution
Comments