My day is not always about software. Sometimes it is about business consulting, but it took a lot of software and a few months to get to the point of being able to do some management coaching.

A customer is running google ads which are not cheap to run, and for each ad seen on the web they generate a unique tracking code called a gclid for that visitor. That gets them to the website as a “lead. Sales then takes that lead and converts them  into a customer through direct contact such as with phone calls handled by the call center.

The customer was supposed to report back to google the gclid tracking code of each sold customer. Google Ad’s then take that returned gclid and tries to find the customer more people like that to target the customers advertisements to. So the gclids flow from Google, through the customer website and sales process and back to google so google can go look for more people like the one that just became a customer.

But for months the customer has not been able to provide google with closed loop reporting of gclids.

Now if google does not receive gclids from the customer, they will continue to send just random traffic to the customers website that may or may not turn into a customer.

My question is, this is just 1 online advertiser with a gclid feedback issue. How many more are out there with the same issues? Must be thousands.

In other words, you want people to respond to google ads that are likely to convert into sales rather than random visitors that might not even have a need or interest in your product or service. But without the gclids flowing back to google, that is what you get. Random traffic which is costly for the customer. So we had to get the gclids flowing back to google.

Before I came along they had some work done for them that caused a page redirect. Google was properly creating GCLIDS and attaching them to the visitor when they sent them to the customer site. But the redirect in the page was causing the gclids to be stripped from the association with that particular visitor.

This went on for months and lots of back and forth before I came along.

I actually got in there and troubleshoot the the issue and discovered it was the redirect causing the issue. I could see it with my own eyes once I studied all their code and figured out what it was supposed to be doing. The Regex in their .htaccess file was supposed to be copying the URL as received through the redirect, but a higher up master server was over riding the lower level server .htaccess file. That was one issue I got fixed. Took a while to track that down and figure out what the code was doing and why.

I also discovered that the google tag manager code was not properly coded to work with this customer’s forms and I was also able to demonstrate and fix the issue with the forms not being able to accept the GCLID even if everything else worked fine.

The redirect took a couple of months to sort out. Had permissions and budgets to get approval for to solve that issue.

Then I had to redesign the forms to be able to talk to the google tag manager code. I also had to modify the google tag manager Javascript code itself to talk to specific fields on the customers forms.

I also had to write some custom PHP code on their server to be able to read and capture the GCLIDS and I had to code up an API to talk back to google tag manager and google ads using CURL.

It was a long way in coming but finally, finally got all the engineering side issues worked out and debugged.

So about 2 weeks ago we finally got good gclids for sold customers flowing back to Google.

Well what do you do with the new customer data.

The GCLID is just a hash code. It means nothing to anyone. But they are traceable to specific customers.

So with good customer data I fired up excel and ran statistics the past 2 days on the known customers that responded to our paid google ads. The whole point being, how can we tune the paid advertisements to work even better than they currently are.

I have an engineering degree. I have used math my whole life. And number crunching just comes naturally to me.

So after crunching the numbers, throwing out the outliers, and making a few observations of the data, I was able to point out to the boss where they were wasting money with their ad targeting and what demographic they should do more targeting of as well as what business hours they should be keeping in the call center. Not what a software guy is supposed to normally be doing, but I am not your ordinary software guy. After helping launch 24 successful startups, I have management skills as well.

So we will see what happens.

We can see about 80 people responded to ads on 1 specific day, to me this means about 80 people wanted to become a customer because they responded to the ad and filled out the form.  They already went through 2 gates. They are red hot leads. But only about 10 actually became customers. Why only 10? Why not 80? What happened?

So now I am wondering what is happening down stream of the form. Would like to see what happens with the sales process to see why only 10 were sold when I would expect 80 to be sold.

I guess we will know in a few months. Always something new to learn.