It’s no secret that there are differences in the way Clicky and Google Analytics measure metrics and dimensions. This blog deals with the dimension of location, specifically the question of why random cities sometimes show up in your “cities” report.

It can be frustrating to open up a location report and see a city in that report that you may not be targeting. In the specific case I talk about below, I noticed that many of my clients in Michigan listed Philadelphia, Pennsylvania as the top city for the same month. How could this be? The client’s paid search wasn’t targeting Pennsylvania, and there’s no content on the site that would draw any visitors from Pennsylvania.

This is how my investigation unraveled, and how I found the answer:

Step 1 – Check your paid search targeting.

This should be your first step, because if you’re handling a lot of large campaigns, it’s possible that a new campaign was set up with incorrect geo-targeting. (Of course, the paid search team at Launch Digital Marketing made no mistakes—our geo-targeting was spot on.)

Step 2 – Check the “Visitor” log.

Check the “Visitor” log in Clicky for the time frame in question. Add a filter: City (choose the city in question). This will allow you to view only the visits from the city in question. In this view, you’re able to see information about each individual visit.

Step 3 – Identify patterns.

Look closely, and look for patterns. The pattern I first noticed was that all of the users in the Philadelphia, PA filter were using AT&T Wireless as their Internet Service Provider (or ISP). Compared to my “all traffic” data, this seemed to be a lot of traffic coming from AT&T Wireless.

All Visitors (notice the differences in ISPs—Sprint, Comcast, AT&T, Time Warner, etc):

All Visitors_ISP

Compare this to the Philadelphia, PA filter (notice that all of the users have the same ISP—AT&T Wireless):


Now I’m really confused. Is AT&T the only ISP in Philadelphia? Why are so many AT&T users from Philadelphia checking out a car dealer association in Detroit? Is AT&T messing with me? That’s crazy. Think again, Miller.

Step 4 – Drill down on individual users.

Back to the drawing board. One of the most useful tools in Clicky (and one that Google Analytics doesn’t offer) is the ability to drill down to the individual user level. You can see in the screenshot below that there there’s information attached to each visit—an ISP, an action, a referral source. If I click on the link tied to the ISP, I’m given even more information about a user:

Visit Information

Here, you can see the user’s IP, locale, ISP, and platform. Here is where I started questioning the validity of the locale. How could all of these people be using AT&T in Philadelphia?

Step 5 – Look up individual IP addresses.

Look up a unique user’s IP address. In reality, I looked up ~50 users’ IP addresses to check their locale against what Clicky reported—never just look up one IP address when trying to prove a reporting tool wrong. You need a good sample size.

I looked up the IP addresses using I prefer this website because it gives information from a number of different IP lookups. While a few reported Philadelphia as the location (likely the same IP address lookup the Clicky dashboard is pulling from), the overwhelming majority of the other IP address lookup sites reported that users were coming from Fraser, Michigan and surrounding areas. This was a bit of a breakthrough—the users weren’t physically located in Philadelphia, Pennsylvania.

So… why was Clicky reporting that these users were located in Philadelphia?

Step 6 – Check where the relevant IPs are registered.

I did some digging on the IP addresses in question using ARIN, or the American Registry for Internet Numbers (which Clicky very conveniently provides the link to):

Visit Information

I found that the same company was responsible for all of the IP addresses. This company (and its main point of contact and headquarters) were all registered to—you guessed it—Philadelphia.

Step 7

I investigated the company responsible for the IP addresses: Wireless Data Service Provider Corporation ( AH-HAH—straight from the homepage: “WDSPCo – promoting Wireless Data through the leasing, distribution, management of IP blocks to commercial Cellular corporations in the United States and abroad.”

So there you have it. A company (WDSPCo) owns bulk IP addresses, and they distribute and lease these IPs to large ISPs like AT&T Wireless (this is where noticing that initial pattern comes into play—noticing that all of the users had the same ISP lead to further investigation of the IPs and their ties to the ISP).

AT&T, in turn, uses these leased IPs for their wireless customers. And those wireless customers could be located in Michigan, or California, or Seattle, but because WDSPCo (the owner and leaser of the addresses) is located in Philadelphia, all of your visits will be tracked to Philly in the Clicky dashboard.

The real lesson here is learning to look at multiple sources for an answer. Just because Clicky uses one tool or website to pull geo-location data for IP addresses doesn’t mean that it’s necessarily the most reliable or up-to-date tool or website. When I used, I could easily see that the users in question weren’t physically located in Philadelphia.

So to wrap up, here are some tips for a quick investigation when you notice odd cities in your traffic reports by location:

  • Check the targeting in your paid campaigns!
  • Look up individual IP addresses using a third-party source. Clicky is not immune to reporting errors.
  • If you’re still curious and want a more definite answer to provide to your client, continue investigating!
  • Look up the location of the questionable IP’s provider—if it’s a small provider (i.e., something called “WDSPCo” and not Sprint, AT&T, etc.), keep digging!

Meet Megan Miller

Megan Miller is a Digital Media Manager at Launch Digital Marketing. She graduated from the University of Illinois at Chicago, where she studied English with a concentration in Media Studies. Megan lives in Chicago and enjoys reading, writing, and spending time with her two cats.