3 builders who never spoke to each other

Cheap traffic is not good traffic

Three operators independently found the cheapest clicks were the worst ones. One platform reliably produced the cheapest visits to a site and almost no buyers; a community platform's ads bought clicks cheaply whose value was correspondingly low; and across a newsletter's channels, open, click and referral rates varied by a factor of two to three, which is the argument for paying more per subscriber where they engage. Judge on conversions, and if conversion tracking is not in place yet, use average session duration as a stand-in, because traffic that leaves in seconds is the wrong traffic.

The method

1. 01

   Measure your ads on conversions rather than clicks, because a platform can deliver very cheap traffic that has no intent to buy.

2. 02

   Reddit ads bought clicks cheaply but the clicks were worth correspondingly little, so cheap traffic is not automatically good traffic.

3. 03

   Pay more per subscriber for an expensive channel when the people it sends you engage better, instead of routing budget to whatever is cheapest.

4. 04

   If conversion tracking is not set up yet, use average session duration as a stand-in: traffic that leaves within a few seconds is the wrong traffic.

What it returned

• Facebook reliably produced the cheapest clicks to the site yet almost no buyers

• Zue ran Reddit ads for placecard.me and found cost per click low but value per click low as well

• Across Morning Brew's channels including Facebook, Instagram and influencer campaigns, open, click and referral rates differed by as much as two to three times.

Sources

• A story about my dad and PPC adsHarry Dry · Marketing Examples · And finally · undated
• The Secrets Behind Morning Brew's Growth to 2.5 Million Newsletter SubscribersJenny Rothenberg · sparkloop.app · 2. Invest in High-Quality Acquisition Channels · undated
• Week 14 RetrospectiveCory Zue · coryzue.com · Comments · 2017-06-26