Just How Badly Did AOL's Screwy Traffic Numbers Drop? [Update]

Last Updated May 9, 2011 2:58 PM EDT

When AOL released its quarterly earnings last week, some nosing into the 10-Q showed that the division with Huffington Post was down a bit in unique readers. But it turns out that things had gone far worse for AOL than the numbers suggested. The only problem is figuring out exactly how bad things have been.

Some back-and-forth with the company uncovered an eyebrow-raiser. A year-over-year comparison on unique visitors to AOL was misleading, making last quarter look better than it was. Without HuffPo, AOL's media-related traffic was down significantly.

To make things worse, total Q1 unique visitors varies by nearly 20 million depending on which financial report you look at. It sure smacks of something suspicious -- perhaps something intended to mask a general deterioration of the company's historic business.
Enter the cavalry
It was the acquisition of Huffington Post and AOL's reorganized structure, replacing the old AOL Media division with Huffington Post Media Group (HPMG), that raised questions:

We utilize unique visitor numbers to evaluate the performance of AOL Properties. Following the acquisition of The Huffington Post on March 4, 2011, AOL aligned all of its content under the newly formed HPMG, which is a subset of AOL Properties and excludes Mail, Instant Messaging and Ventures. As a result of this realignment and to reflect how management views the business, we are disclosing domestic average monthly unique visitors to HPMG for all periods presented and will no longer disclose domestic average monthly unique visitors to AOL Media. The primary differences between HPMG and AOL Media are that HPMG includes The Huffington Post, AOL Search and Local.
AOL's most recent 10-Q noted that, as measured by comScore's Media Metrix 360 service, HPMG dropped 3 million unique visitors in the first quarter, from 103 million in 2010 to 100 million this year.

It depends on your definition of addition
There are, however, a few problems with this reporting:

  • Even though AOL had access to Huffington Post traffic numbers, it only included them for March of 2011 and didn't report what they were, so there was no clear comparison to show how AOL's own media properties did. (I've requested this information from AOL.)
  • There is no indication of where AOL Search traffic was accounted for in the past. The only direct reference to AOL Search in AOL's 2010 10-K is an explanation of what it is. However, the description of AOL Properties suggests that the company either included search results in that division or else didn't report them in any form.
  • The Local group included not only Patch, but mapping service MapQuest and a local entertainment information service called City's Best. Again, it's difficult to tell whether unique visitor numbers appeared rolled up under the AOL Properties category or not.
When you look at all the unique visitor numbers based on Media Metrix 360 that AOL has reported over the last year, the results are peculiar and don't seem to hold up. Here's a table of them (click to enlarge):


Except for a dip in the third quarter of 2010, the numbers are completely flat. That alone seems odd. Next, notice that AOL Media numbers slid and then rebounded a bit by the end of the year. You might explain that by a combination of a general loss of visitors with a seasonal increase.

Who's on first?
But then everything gets completely screwy with the HPMG numbers. Although a real look at the company would require the year-over-year comparison to include equivalent HuffPo number to get a true comparison, that's not what happened, according to what AOL told BNET. The company only included HuffPo visitor numbers for March of 2011, and none at all for 2010.

HuffPo has a significant readership, so let's filter out that result. According to comScore, the HuffPo Media Metrix 360 number for January 2011 was 27.6 million, and the number for February 2011 was 24.7 million. HuffPo came off the published listings in March, as it had become part of AOL.

The two months give a monthly average of 26.15 million unique visitors as measured by comScore. (Remember that the actual number could be different, as there is no easy way to independently verify comScore's methodology.) That would have added roughly 8.7 million per month in unique visitors to the HPMG 2011Q1 numbers (one average month of visitors split among the three months of the quarter). Pull them out, and suddenly HPMG average monthly visitors drop to 91.3 million.

Pay no attention to the managers behind the curtain
However, remember that AOL also added added local and search to the old AOL Media numbers to get the HPMG figures. The difference between them in 2010Q1 was 17 million. So, do we subtract that from the 91.3 million to get an equivalent 74.3 million for the first quarter of this year? That would be a 13.6 percent drop. If you remove only the Huffington Post influence, you're still left with an 11.4 percent drop in unique visitors.

Now look at that chart again. AOL Properties traffic was largely flat. AOL Media was declining slightly, not by the lurching drop that HPMG showed in a year-over-year comparison without the HuffPost boost. So where do the HPMG numbers come from? How can AOL have 384 million unique visitors between AOL Properties, AOL Media, and AOL Ad Network in the first quarter of 2010, but then have 401 million between AOL Properties, HPMG, and AOL Ad Media? It comes across as AOL bolstering results with numbers that it pulled from its corporate backside.

[Update: Here is AOL's explanation. First, AOL Media was always a subset of AOL Properties and there is significant crossover in audience, so subtracting one from the other doesn't necessarily mean anything.

According to comScore, Huffington Post's numbers in March were 31.1 million. According to AOL sources, even before the acquisition, there was about a 50 percent cross-over in unique traffic between AOL and HuffPo, so the 31.1 million probably represents more like 15 million incremental users. A third of that (to get an average for the quarter) would be 5 million per month. AOL only included HuffPo numbers for March 2011, which makes things difficult without that additional information.

In 2010, the difference between AOL Properties and AOL Media unique visitors were AOL Mail and AIM. Again, though, there was duplication in visitors.

The difference between HMPG between 2010 and 2011 without Huffington Post was closer to 95 million, which meant a traffic drop of closer to 7.8 percent. AOL sources say that their internal numbers suggest a lower percentage drop. I'd have to see more details to say more.]

Related:

Image: Flickr user monkeyatlarge, CC 2.0.
  • Erik Sherman On Twitter» On Facebook»

    Erik Sherman is a widely published writer and editor who also does select ghosting and corporate work. The views expressed in this column belong to Sherman and do not represent the views of CBS Interactive. Follow him on Twitter at @ErikSherman or on Facebook.