Measuring Paid Advertising Across Platforms
Add up the conversions reported by each ad platform and the total will usually be higher than the number of customers you actually gained. Each platform counts with its own rules and none of them can see the whole path a customer took. This guide explains where the overlap comes from and how to build a view of results you can trust.
On This Page
One Customer, Several Claims
Each platform uses an attribution window, a period after someone clicks or sees an ad during which a conversion is credited to that ad. If a customer watched a TikTok ad on Monday, clicked a Meta ad on Wednesday and searched your name on Google before booking on Friday, all three platforms can report that booking as theirs.
Windows also change. In October 2025 Meta announced that its 7-day and 28-day view-through attribution windows would stop returning data in its reporting API from January 12, 2026, which lowered reported conversions for many advertisers without any change in real sales.
UTM Tags
UTM parameters are short tags added to the end of an ad's link, such as utm_source=meta&utm_medium=paid_social&utm_campaign=spring_offer. Analytics tools like Google Analytics read them to show which campaign sent each visitor. Google notes the values are case sensitive, so Meta and meta show up as two different sources. Agree on one naming convention and use it on every ad.
Call Tracking
For many service businesses, the phone call is the conversion. Call tracking assigns a unique forwarding number to each channel or campaign and records which number was dialed. Google Ads can also count calls from call assets and calls from your website. Set a minimum call length, such as 60 seconds, so hang-ups and wrong numbers do not count as leads.
Offline Conversions
An ad platform only knows what you tell it. If a form lead turns into a $12,000 kitchen remodel three weeks later, the platform sees the form and nothing after it. Offline conversion imports send that outcome back. Google Ads accepts uploads of qualified leads and sales, and Meta's Conversions API accepts CRM events. Feeding real outcomes back helps each platform find more people like your best customers.
A Single Source of Truth
Pick one system as the record of customers, usually your CRM, booking software or point of sale, and judge platforms against it.
- Record the lead source on every new customer, using UTM data, the tracking number or a required how did you hear about us field.
- Compare platform-reported conversions with CRM customers by channel each month.
- Track blended cost per customer, which is total ad spend divided by total new customers from all paid channels.
Blended numbers cannot be gamed by overlapping attribution, which makes them the most honest check on whether advertising is growing the business.
Testing What Ads Add
The strongest evidence comes from experiments. A holdout test shows ads to one group and withholds them from a similar group, then compares results. A geographic test pauses a channel in some cities or ZIP codes and keeps it running in others. Meta and Google both offer lift studies for eligible accounts. Research on large Facebook advertising experiments found that the observational methods most advertisers rely on often over or understate the true effect.
In Practice
Consider a roofing company running Google Search and Meta ads. Google reports 30 conversions and Meta reports 25 in the same month, yet the CRM shows 38 new inspection bookings from paid channels. Every booking carries a lead source, so the owner can see Google produced 24 and Meta 14, and compare $9,000 in total spend against 38 bookings for a blended cost of about $237 per booking.
Attribution settings and measurement tools change often. Confirm details in the sources below before relying on them.
Key Terms
- UTM parameters
- Tags added to a link that identify the source, medium and campaign of a visit.
- Call tracking
- Unique phone numbers per channel that show which ad produced a call.
- Offline conversion import
- Sending sales or qualified leads from your own records back to an ad platform.
- Blended cost per customer
- Total paid media spend divided by all new customers from paid channels.
- Holdout test
- An experiment that withholds ads from a comparison group to measure what the ads added.
Sources
- URL builders, collect campaign data with custom URLs, Google Analytics Help support.google.com
- Ads Insights API metric availability updates, Meta for Developers developers.facebook.com
- Gordon, Zettelmeyer, Bhargava and Chapsky, "A Comparison of Approaches to Advertising Measurement: Evidence from Big Field Experiments at Facebook," Marketing Science, 2019, 10.1287/mksc.2018.1135 doi.org
- Google Ads Help support.google.com
Have Our Team Handle This
Our team runs this every day. Book a call and we will plan it for your business.