When marketing and sales stop trusting HubSpot’s source reports, the usual reaction is to change the attribution model. On the audit below, the model wasn’t the problem. The data going into it was. Here are five checks I run, with what each one found on an audit for a multi-location estate law firm running Google Ads, Local Services Ads, Meta, CallRail, an intake chat tool and a long tail of legal directories.
1. Reconcile spend, platform by platform, day by day
Pull spend from each ad platform and from HubSpot for the same window, and compare them by day and by campaign.
On this audit, over a 30-day window, Google Ads matched to within 0.04% ($7,186.75 on the platform against $7,183.94 in HubSpot) and Meta to within 0.02% ($7,715.66 against $7,714.47). That’s good news: it means the syncs that existed were working, so any problem was elsewhere.
2. Look for channels that aren’t syncing at all
The same comparison shows what’s missing. Local Services Ads spend wasn’t reaching HubSpot in that portal at all. At one location, $2,354.69 of real spend with 16 charged leads showed in HubSpot only as $0 placeholders. Leads came in through CallRail with the call attached but not the cost, so they couldn’t be attributed.
3. Compare HubSpot’s source label with the UTM data
For a sample of contacts, put HubSpot’s original source property next to the UTM values that came with them.
Roughly 21% of “paid search” contacts (at least 83 of 393) were actually organic leads from Google Business Profile. HubSpot’s source property had labelled them as paid; the UTM data was right. That single difference makes paid search look better than it is in every report built on top of it.
4. Count leads on both sides
Compare the number of leads each platform says it delivered with the number that exist in HubSpot. At one location, 5 of 17 leads never reached HubSpot, and 4 of them had already been charged for.
5. Check what your offline conversions fire on
If you send conversions from HubSpot back to an ad platform, look at the exact trigger and the value it sends.
Here, the offline conversion import to Google Ads fired on a trigger that matched nearly every contact on entry, and it sent a flat default value instead of the real deal value. The bidding algorithm was effectively optimizing for raw lead volume, not qualified clients. That explained a complaint about cheap, low-quality leads that had been blamed on something else.
What to fix first
The audit led to these fixes:
- One set of source properties that every channel writes into, documented in a field map with one row per property and one column per source.
- Priority rules for which signal wins: UTM data over intake capture, over a manual “how did you hear about us” answer, over direct. The rules also say when a later signal may overwrite an earlier one.
- Tracking templates at the account level, so campaign and keyword data arrive filled in.
- The missing spend sync for Local Services Ads.
- Offline conversions on real deal stages, carrying the real deal value.
A quick self-check
| Question | If the answer is no |
|---|---|
| Does each platform’s spend match HubSpot within a fraction of a percent? | A sync is broken or missing |
| Does every paid channel appear in HubSpot at all? | Some spend is invisible |
| Do source labels agree with UTMs on a sample? | Your channel reports are skewed |
| Do lead counts match on both sides? | Leads are being lost in capture |
| Do offline conversions fire on qualified stages with real values? | Your ad platform is optimizing for the wrong thing |
If you can’t answer these for your own portal, that’s what a HubSpot audit is for.
Need this done for you?
A technical review of your portal, data, and integrations, with a written list of what to fix first.
HubSpot AuditWritten by Ali Ismail, who runs TechPad and builds HubSpot integrations, migrations and automation.