4 ms·

Sorry for asking so many questions, but I don't know much about this and you seem knowledgeable. > However, for search and display ads, which drives a huge amo

by tpxl 4y ago

Sorry for asking so many questions, but I don't know much about this and you seem knowledgeable. > However, for search and display ads, which drives a huge amount of traffic to EU businesses, there's no effective way to run ads without conversion tracking. Obviously you can just throw money at Google/Facebook, but without conversion tracking, the average return on advertising spend will be 1-10% of what it would be otherwise and you will pay for a lot of junk traffic. 1. Can you not track conversion in aggregate by running X number of ads and checking how much extra traffic you got? It's obviously noisy and you can't run multiple campaigns at the same time, but conceptually I don't see why it wouldn't work? 2. I'm assuming you put links in your ads, so if you run ads for lander.example.com, can you not put the URL in google/facebook as lander.example.com?src=g and src=fb? Or, if you can't use query params, g.lander.example.com and fb.lander.example.com? Tracking stats for those and seeing how much ads you ran should give you conversion, similar to 1. but better?

closewith 4y ago
No problem. > 1. Can you not track conversion in aggregate by running X number of ads and checking how much extra traffic you got? It's obviously noisy and you can't run multiple campaigns at the same time, but conceptually I don't see why it wouldn't work? You can and many companies will have a first-party (or partly first-party, feeding into PowerBI or Tableau or something) system for this kind of reporting completely independent of the ad networks (to independently verify the numbers the ad networks claim, to include other data that a company would never share with Google, and for fraud detection). However, without conversion tracking, Google Ads' "Smart Bidding" and "Remarketing" targeting features don't work. For a lot of businesses, being unable to, for example, stop advertising to existing customers or to use a certain bidding strategy means that the Return On Advertising Spend (ROAS - the big metric for enterprise marketing departments) goes below 1. > 2. I'm assuming you put links in your ads, so if you run ads for lander.example.com, can you not put the URL in google/facebook as lander.example.com?src=g and src=fb? Or, if you can't use query params, g.lander.example.com and fb.lander.example.com? Tracking stats for those and seeing how much ads you ran should give you conversion, similar to 1. but better? Companies often do this, with Google Analytics-compatible UTM parameters, custom source IDs, and very often unique marketing landing pages for each ad campaign. As above, though, reliable reporting of conversions/revenue isn't the difficult part. It's being able to reliably target the ads/campaigns without conversion tracking. Right now, targeted search and display advertising can be 10-100x more effective (in terms of Return On Advertising Spend) than keyword- or context-only targeting. That's not an advantage a company can forgo unless the entire market is forced to simultaneously.

by tpxl 4y ago

closewith 4y ago

No problem. > 1. Can you not track conversion in aggregate by running X number of ads and checking how much extra traffic you got? It's obviously noisy and you can't run multiple campaigns at the same time, but conceptually I don't see why it wouldn't work? You can and many companies will have a first-party (or partly first-party, feeding into PowerBI or Tableau or something) system for this kind of reporting completely independent of the ad networks (to independently verify the numbers the ad networks claim, to include other data that a company would never share with Google, and for fraud detection). However, without conversion tracking, Google Ads' "Smart Bidding" and "Remarketing" targeting features don't work. For a lot of businesses, being unable to, for example, stop advertising to existing customers or to use a certain bidding strategy means that the Return On Advertising Spend (ROAS - the big metric for enterprise marketing departments) goes below 1. > 2. I'm assuming you put links in your ads, so if you run ads for lander.example.com, can you not put the URL in google/facebook as lander.example.com?src=g and src=fb? Or, if you can't use query params, g.lander.example.com and fb.lander.example.com? Tracking stats for those and seeing how much ads you ran should give you conversion, similar to 1. but better? Companies often do this, with Google Analytics-compatible UTM parameters, custom source IDs, and very often unique marketing landing pages for each ad campaign. As above, though, reliable reporting of conversions/revenue isn't the difficult part. It's being able to reliably target the ads/campaigns without conversion tracking. Right now, targeted search and display advertising can be 10-100x more effective (in terms of Return On Advertising Spend) than keyword- or context-only targeting. That's not an advantage a company can forgo unless the entire market is forced to simultaneously.