I’m working with a new client who runs Google Ads for lead generation using HubSpot forms. The problem I’m facing is that when the HubSpot events were originally set up (leads, MQLs, SQLs), they all got the same default conversion value assigned to them.
This is causing issues because I need Google to understand that SQLs are higher quality leads compared to regular leads or MQLs. Right now Google treats them all the same since they have identical conversion values.
I’ve looked everywhere in HubSpot but there’s no way to edit the conversion value of existing events. The only solution I can think of is to delete these events completely and create new ones with proper values.
My main concern is what happens to my current campaigns if I do this. Will deleting and recreating the HubSpot events mess up the machine learning that Google has already built up? I don’t want to hurt campaign performance while trying to fix this tracking issue.
Has anyone dealt with this before? What’s the best approach here?
Just recreate them - I know it’s scary but I’ve done this twice and ML recovery is quicker than expected. Export your historical reports first to keep that data. Here’s what most people miss: update your bid strategies at the same time. Don’t leave them optimizing for deleted conversion actions. My campaigns took 2-3 weeks to hit baseline again, but performance actually got better because Google finally understood the lead quality differences.
You’re spot on - HubSpot won’t let you edit conversion values once events are created. Hit the same wall with a SaaS client last year. Here’s what worked: Don’t delete the old events right away. Set up new events with correct conversion values and run both versions side by side for 3 weeks. Google’s algorithm gets time to learn the new data without losing all that historical context. The trick is slowly moving budget toward campaigns that optimize for the new events instead of flipping a switch overnight. Performance stayed steady during the switch, and lead quality got way better once Google figured out the value differences between lead types. Took about a month total, but beat the hell out of starting from scratch and dealing with that learning period reset.
Had this exact same issue 6 months ago with an e-commerce client. Yeah, deleting and recreating events resets Google’s machine learning, but it’s not as bad as you’d think. I created the new events with proper values first, then slowly moved conversion tracking over 2 weeks instead of all at once. Kept the old events running so Google could still optimize while learning the new setup. Performance dropped a bit for 10-14 days but bounced back once the algorithm figured out the new conversion values. Time this when you can handle some temporary dips. Make sure your new conversion values actually match what each lead type is worth - I used our average customer lifetime values for the ratios. The better optimization from proper conversion weighting was totally worth the short-term hassle.