A couple of different issues have come to light in recent months and they highlight just how messy and murky online advertising is becoming. The first issue has to do with how most marketing companies now incorporate some from of Google pay-per-click into their advertising solution. Many times however, the customer is completely unaware of this. The other issue I will highlight in this posting is how improper tracking of the effectiveness of your online advertising can very easily lead you to the absolute wrong conclusion.

Let’s start with the latter issue of tracking your different online advertising campaigns. A favorite way to do this is to use call-tracking (Figure 1 below). Basically you get hooked up with a call-tracking company that can create multiple unique phone numbers for your various different advertising campaigns. You can create a unique number for your newspaper ads, your yellow pages ad, your business cards, the side of your service van, your website and various different online advertising methods like search engine pay-per-click, Dexknows, Angie’s List, Yelp ect..  All of these unique numbers ring through to your business and can be saved and tracked in a historical record.  With all the different ways to advertise and each with a different cost, this strategy is a reasonably sound strategy to assess the cost-effectiveness of your different advertising campaigns, provided you are careful not to over-interpret your results and you understand the potential pitfalls.

There are a couple very easy ways to demonstrate where the call-tracking results can break down and be misinterpreted. If you just take your final results at face value and start to make critical advertising decisions ONLY based on your call-tracking data you could very easily make the false conclusion that your best advertising campaign is actually your worst advertising campaign. So here is one easy example to highlight where call-tracking can easily lead you to the wrong conclusion about one form of advertising.

Many of your new customers will very likely find out your business name from one form of advertising, and then come back later to your website to do further research on your services. The initial connection between your business and that new customer might have been, for example, Angie’s List but in the end it will be recorded as your main website because that is the number they end up using latter. So this becomes an easy example of a conversion not being properly assigned to the correct initial advertising campaign. If you are using call-tracking, much of your directory ads (like Dexknows or Yelp) will also all have a unique phone numbers. These directory ads will also generally have a link back to your business website. The Yelp ad may have the phone number 612-233-4040 and the website 612-233-4042. So when the customer finds you through the Yelp ad but then decides to click on the link to go to your website to do a little more research about your business, the correct source for the conversion has now been lost.  Your call-tracking will record that this conversion came from your main website phone number and not Yelp. There are dozens of examples in terms of online advertising where this could happen. And actually is it the various forms of online advertising that tend to be miscategorized by call-tracking. More traditional means of advertising like door-knockers or the side of your van tend to be more correctly captured in your call-tracking data. But even with that, recent stats have indicated that 89% of all consumers use the Internet to help research their business decisions. So the lines between online advertising and traditional advertising are also getting blurred.

A more subtle example of online advertising campaigns not getting the proper attribution in the call-tracking reports are the very popular search engine pay-per-click campaigns. We provide a real life example below where a client came to us asking about why their call-tracking data was suggesting that their pay-per-click campaign was their least effective campaign. We were very surprised by this because throughout our 13 years of specializing in SEO and search engine pay-per-click advertising 9 times out of 10 the two most cost-effective online forms of advertising for a business will be SEO and PPC. Upon closer examination of our client’s data we quickly found the problem.

So if you look at the call-tracking data above the cost per lead for the client’s PPC is recorded as $81.44 cost per lead. At first glance this looks very high and gives you the impression that their PPC campaign is not very effective. But as we will show, this interpretation would be wrong because the tracking of their PPC campaign is grossly inaccurate for at least two different reasons.  The first one is that even though they are using a unique call tracking number for their PPC ads on Google, that number only shows up in about 44% of the ad displays. So this means that 56% of the time their PPC ad on Google shows up without any tracking number at all. This is completely a function of how Google displays their ads. We have analyzed millions of impressions on Google Adwords and this breakdown between a call-extension showing up with an ad versus not showing up is pretty consistent.

So just with this first issue our client’s true cost per lead for PPC immediately improves to $36 per lead instead of $81.

The second issue with pay-per-click ads is very similar to what I described with the Angie’s List and directory ads, as soon as the user clicks on the PPC ad they go to the company website which is now displaying a different unique phone number.

So in both cases the conversion would be recorded in the “Main-Website-Online” category on the Call Source Report like what is seen in figure 2 (above) and not the real source which was their PPC campaign.

So really, a few of the calls that are recorded in the “Main-Website-Online” category are almost guaranteed to be coming from PPC. So even one more lead that has been miscategorized in the “Main-Website-Online” would now drop their cost per lead for their PPC campaign to $25 / lead.  This now makes their PPC campaign their most effective advertising campaign as I would have predicted.

If you are going to use call-source tracking then you need to have secondary data to confirm and support what the call-source tracking report is implying AND we would recommend that PPC campaigns always be grouped with the Main-Website category when calculating effectiveness. So making that the case for this report you would group Category B and F together which produced 5 leads at a cost of $16.20  per lead. That is a much more accurate way to measure contributions from both the Main website and their Google Pay-per-click campaign and you won’t make the mistake of misinterpreting your best advertising campaign as your worst.

Stay tuned for the second part of this two part series “When Free is not Really Free” which we will be posting in the next couple of days. Hope you found this helpful in your future assessments of your various advertising campaigns and as always if you have any questions or need SEO or PPC advice do not hesitate to call us at  1-800-975-5695.

Eric SEO and SEM