The proverb ”you get what you give” is extremely important in the world of digital marketing. With the way that bidding algorithms run in platforms like Google Ads, Microsoft Advertising, and Meta, each user is analyzed from every possible angle when determining what the bid in that particular instance should be.
For instance, Google Ads looks at millions of user signals – everything from their gender and location to which site is their favorite for new recipes – to determine the likelihood that the particular individual searching your keyword at that moment is going to turn into a customer. If the likelihood is low, the algorithm might bid a couple of cents, or not even offer a bid. If the likelihood is high enough, the bid might come in just under a typical cost per acquisition (CPA) for the campaign.
Algorithms Are Powerful — But Only as Smart as Your Data
These bidding algorithms are very powerful tools, but if the advertiser is not providing accurate information about what a real conversion is, all that marketing power is just wasted.
Ecommerce vs. Lead Generation: A Clear vs. Blurred Conversion
This is an easier task for ecommerce marketers versus the lead generation crowd. If a visitor buys a $50 tool off a website, that’s a very clear conversion action, as each dollar of revenue is easily measured. But what if a lead gen prospect picks up the phone and calls, or what if a user fills out a form? Not every phone call and every form submission is created equal.
The Problem: Unvetted Lead Conversions
One of the first issues I see clients have with lead conversions is that the leads fed to your ad platform are not vetted. If you have the mark of a phone conversion set to any call over a minute, and it takes that long for a caller to prove that they are not looking for a product or service that the business offers, that call still gets sent to the platform as a conversion. And the algorithm adds that user’s signals to the positive list when determining future bids. The same is true when a spam form submission gets logged in the platform as a conversion.
The Solution: Vet Leads Before Sending Them
The key is to vet your leads before sending them to the marketing platforms. If you are using a call tracking service, instead of setting a call duration as the trigger for a conversion, take the time to listen to the calls. Mark the good calls and only send the relevant calls as conversions.
I had a client make this kind of switch with his call conversions recently, and the results have been outstanding. Not only did he realize that some of the long calls were irrelevant despite the long duration, but he also found that a lot of the shorter calls were also qualified leads. As a result of providing better conversion data to Google Ads, this client is seeing a higher rate of quality calls coming in lately.
Tracking and Filtering Form Leads Properly
The same thing applies to form leads. However, to really track quality form leads properly, you need to be able to capture the GCLID (Google Click Identifier) or MSCLKID (Microsoft Click ID) and then upload only the quality submissions. Read our previous blog by Jeff Kramer about offline conversion tracking for more information.
If you tell the platform that a spam lead is equal to every other form lead, the algorithm will see them as equal, and if an easy source of spam leads is found by the algorithm, overall account lead quality could take a large downturn. This is especially true with Performance Max campaigns, as they seek to find leads throughout the internet.
Don’t Focus Only on the Final Booking
With lead gen accounts, you don’t want to focus solely on the end goal. I had a private practice doctor whose team would mark each lead as good or bad. However, they only marked the leads that completed a booking on the call. While it is good to tell the platform the users who booked, there were not that many users who booked on the initial call. They were only sending Google a handful of conversions a week. As a result, the number of leads dwindled, especially as they reached their slow season late in the summer.
At that point in time, the doctor decided to start reviewing the calls himself. Most every call was relevant. The calls that did not complete bookings usually didn’t book due to pricing, scheduling, or having to speak to a friend or family member before setting up an appointment. Once all the relevant calls were being pushed into Google Ads, calls increased, appointments increased, and the doctor recently increased his budget in order to fill up the remainder of his surgical calendar.
Value Every Meaningful Milestone
The ideal structure for feeding the algorithms the best data for bidding purposes is to set checkpoints along the sale journey and value these checkpoints at increasingly higher values. For instance, one of my B2B clients captures leads from advertising buyers. When a lead comes in and has been vetted as a real person working for a real company, we place a value of $10 on the lead.
Even though we have more important checkpoints in the sales funnel, we like to put a little value on every relevant lead. If we didn’t do that, the algorithm would see these relevant leads as just as valuable as someone who visited the site for five seconds and then left without taking any additional actions.
If the user attends a demo meeting, it’s worth an additional $25. If they move on and close a deal, we add another $250 to $500 of conversion value based on the size of the deal. And if it’s a highly desirable lead, we add $100 even if they do not close a deal, just because we want the algorithm to place more value on similar users.
Feed the Machine the Right Data
The key is to show the algorithm an appropriate value when there is a good lead, while also not giving any value to irrelevant or spam leads. The platforms are only as smart as the data you give them. If you send junk, they will find more junk. If you send clean, qualified, properly valued conversions, they will find more of those.
The algorithm is not guessing. It is learning. Every conversion you upload teaches it who to bid on next. So be intentional. Vet the calls. Filter the forms. Assign real value to real milestones. Stop rewarding spam. Stop under-valuing good prospects. Because in digital marketing, you truly do get what you give. Feed the machine the right data, and it will reward you with better leads, stronger performance, and scalable growth.
