First, I looked into what the other guy was running and told the business owner what was wrong. I explained the entire situation.
Then I made a controversial move. My CPL actually went UP to $30 on Meta and $55 on Google. I dramatically increased their cost per lead even though their goal is a very low CPL.
But here's why I did that—I have no ways of verifying lead quality. These guys don't have lifecycle stages setup. They don't have a CRM with pipeline stages. They don't report on good leads in any meaningful way. They only get leads.
So I have no offline conversions to pass back to Facebook or Google, even though this is something I talk about in every single case study. This is one of the rare cases where I don't even have this option.
What I had to go off of was basically what I know, based on my experience, generates good quality leads. And I had to double down on that.
For example, I decided that no Facebook lead forms would ever run. Even though we now have an option to implement one-time passcode to verify phone numbers, I still see not very high quality leads from Facebook lead forms.
On Google Ads, I obviously turned off the display campaign. Only running pure Google search campaigns now. No display network, no search partners, nothing like that. Only Google searches.
I built a landing page that does qualify those leads—a multi-step form with like 9 or 10 questions. Even though this might be hurting my CPL a little bit, because I don't have any visibility over lead quality, I have to pay this cost. And as long as I'm beating my KPIs, which I am (slightly at least), I'm happy to add more scrutiny to my forms to make sure these numbers are working.
No one is measuring lead quality, so it's up to me to make sure those leads are not only cheap but also high quality without getting any sort of feedback from the client about lead quality.
On the Facebook side, I'm constantly hiring actors, introducing new creative concepts every single week, putting out more static image ads, running AB tests on the landing page, and testing different ad copy.
On the Google side, I'm constantly adding negative keywords, changing bidding strategy over time, and testing broad against exact against phrase. Broad won, as you probably expected, because broad wins most of the time when it comes to mass markets or big markets.