Determining Website Traffic Tool Review: Compete.com

January 19, 2009 · Comments

Jason Falls

Jason Falls

Jason Falls

Determining a website’s traffic is a tricky thing. To get completely accurate numbers, you need to contact the website owner and ask them for a traffic report. Since many website owners won’t volunteer such information, for a variety of reasons, some of which are valid, you’re left hoping someone has built a cool tool to save you.

Maybe you’re a public relations professional looking for high-trafficked sites for outreach. Perhaps you’re a media planner and buyer and want to get a hold on which blogs and other online media sites should be in your client’s plan. Whatever the reason, finding a tool to help you research a site’s traffic can be a Godsend.

As with most Internet research needs, there are a variety of companies that provide solutions, some good. Others not so. There are three free tools available to anyone: Compete.com, Alexa.com and Quantcast.com. Today we’ll look at Compete.com. Soon, we’ll cover the other two. The analysis of the paid solutions (ComScore, Nielsen, Hitwise) will come whenever they give me a free preview to try their tools. (Hint, hint, boys.)

Compete.
Image via Wikipedia

Compete.com is a panel-based measurement service. This means they take a sample of Internet use statistics (called click streams) from users that have opted in to volunteer that information. Their data sample is from over 2 million U.S. consumers. This includes click stream data from a number of Internet Service Providers (ISPs) which anonymize the data of their users, who have agreed to terms which allow the ISP to sell the data to firms like Compete. The sample also includes Compete toolbar users: People who have gone through the trouble of downloading a toolbar plug-in from Compete specifically to volunteer their web surfing data for measurement. There is an inherent bias in this portion of their data because it will skew geek. Generalizing a bit, only computer nerds and webmasters are going to download this toolbar. Compete’s third data point is what they call a proprietary panel. (Because what’s a good sales pitch without working in the word “proprietary?”) This panel is a group of users actively engaged with Compete to provide data, respond to surveys, etc.

Once all the data of web surfing activity, sites, etc., is collected, it is normalized, or weighed and skewed to correct biases like the geek-heavy toolbar users. The goal of the normalization process, which by the way is conducted and overseen by people with Ph.d’s in this sort of number crunching, is to have each consumer measured represent the average American consumer. The normalized data is then used to project metrics that are an estimate of how the average U.S. consumer behaves online.

The key thing to know here is that even though Compete.com has data from over 2 million users (more than ComScore and Nielsen, to my knowledge, which use samples of 200,000-300,000) they are still projecting results, a process I believe is inherently flawed.

Still, Compete allows users to compare these projected traffic numbers of up to three websites for free. While the numbers are inconsistent and far from accurate (see below), the tool can be useful for site comparisons.

Paid Solution

Smartly, Compete doesn’t offer just a free tool and that’s it. For subscription rates ranging from $199 per month to $499 per month, plus custom pricing for enterprise offerings, you can access additional site analytics reports (like visits per person and daily reach, which do not appear in the free version) and a trio of added value pieces to make you think you’re getting something for your money.

The Search Analytics suite offers tools to find sites that get the most traffic from certain keywords, keywords that drive most search traffic to a given site and comparisons of search keywords for sites so you can judge yourself based on your competitors.

There’s a Referral Analytics toolset that allows you to see a ranking of what sites point to yours, what sites your website sends traffic to and a comparison tool to see referral and destination traffic for any two websites.

The final piece of the paid component is ranked lists of the top websites. You can get the top 200 websites with any paid plan. Stepping up a notch on the plan gradient gets you the top 1,000 and then the top 15,000.

Frankly, I don’t think there are enough tools for $200 per month. Sure, those there are worth something, but similar information can be found online and probably for free. There are also some keyword and search specific tools out there that are much more robust than what Compete offers. Still, the toolset is useful and Compete at least provides everyone with the traffic comparisons, giving something to the greater web community. That should count for something.

Aaron Prebluda, Compete’s director of market development and to whom I’m indebted for his kindness in answering questions, told me, “Our paid product provides competitive intelligence and behavioral data that is actionable.” To the letter, that’s true. But companies that use Compete’s paid solutions will have to be savvy enough to know what actions to take. Seeing a chart that shows you the top search keywords that drive traffic to your site is one thing. Knowing how to analyze that data and use it to improve search results for certain keywords is something very different. What that means is the paid solution is only going to be worth your while if you have an SEO specialist on staff or on retainer to do something with it. (If you need one, let me know. Kat and David are really good.)

The Unfortunate Discovery

All that said, I decided to test Compete’s data against sites for which I have access to Google Analytics. Honestly, it appears Compete.com isn’t accurate at all. Here’s a comparison of Compete’s numbers and those of several sites whose Google Analytics reports I can access. (I’ve removed the names of the sites as I don’t own them and am not sure if the owners want their actual traffic numbers public.)

Compete Traffic Comparison Chart

As you can see from this small sampling, there’s a big difference in actual and what Compete’s data and normalization produces. For the record, I did notice on one seasonal website a mirror of the peaks and valleys of the web traffic, but the metrics were way off.

What this tells us is that Compete’s statistical patterns might mirror reality, but the numbers are based on fuzzy math and unreliable. Folks from Compete may balk at that statement, saying I only looked at three websites. But isn’t taking a sampling of data and projecting it on the entire population kinda what they’re guilty of, too?

I would recommend Compete as a tool to compare the traffic of websites. And if you’re interested in a fairly economical keyword and search engine research tools, at least in comparison to a lot of solutions out there, it’s worth considering even if the price tag seems a bit disproportionate to the value you receive. Still, the primary focus of this series of reviews is to determine website traffic. Frankly, Compete.com’s vast inconsistencies with the web analytics I trust to be true makes me hesitant to do much more with the tool.

What’s your experience with Compete? Is there more to their paid solution I’ve missed? If you haven’t tried it, do so now and come back to report your thoughts on the tool. And what do you think about the comparison to Google Analytics numbers? Run sites you know the actual traffic for and compare your numbers to Compete’s. Let us know how far they are off. The comments are yours.

Reblog this post [with Zemanta]

Tagged as: Alexa.com, Compete.com, determining website traffic, Quantcast.com, web analytics, web traffic

Did you enjoy this blog post? If so, then why not:
Leave Comment Below | Subscribe To This Blog | Sign Up For Our Newsletter |


  • Compete has been about 5-10% shy on SeeClickFix.com but very very close to our analytics numbers.
    Compete counts our widget hits which alexa definitely does not.

    I'll check out quantcast and let you know.
  • Thanks, Ben. Good to get some real examples!
  • garygranneman
    Compete.com reported that schoolofmusic.com had only 179 visits in September 09. We monitor visits, and our software determines IP addresses, city, state, country, OS, date and time of visit, and more. This is not some fantasy. We counted over 3,000 unique visits. Since we are localized the difference in the numbers is significant in our relationships with others. Quantcast, which monitors our site, showed 2300 visits in the last 30 days. All I get from compete.com is an explanation of their, obviously flawed, methodology. Apparently, we get a low rating because few of their audience members are interested in music lessons. Most of them are probably not parents and most lessons are taken by children. We had more hits reported on Google Maps alone than Compete shows. Is there really a point in using them for anything? Does anyone actually believe their numbers? For some relevant pngs and other information, please visit http://www.ripoffs.org/competecomaccuracy.html.
  • Thanks for this Gary. The problem with Compete and Alexa (and
    Quantcast for those who don't embed their code) is they predict what
    they think a website's traffic might be based on extrapolating data
    from a variety of sources. This is the key problem I've always had
    with predictive audience equations. They're guesses ... even educated
    ones ... but inherently flawed. I doubt you could sue ... I'm sure
    there's something in the fine print that says they aren't responsible
    for bad numbers ... but you certainly raise a great point about how
    reliable they are.

    Thanks for sharing your numbers and perspective.
  • Alexa is one of the most known names in the web circuit. http://www.alexa.com/siteinfo/fortunehotels.in Alexa's numerous features aimed at the benefit of the users have contributed immensely to its rapid expansion and ever growing popularity.
  • Well then have someone from Alexa return an email or a Tweet so I can do a better analysis of their tool. As such, it appears to be a geek-centric browser download driven extrapolation of skewed numbers. Just my take. But thanks for chiming in.
  • LaurenCandito
    Thanks for the article Jason - I have used Compete before and I was not very impressed with the results. I appreciate knowing someone else shares my opinion.

    Lauren
  • You're welcome Lauren. They have their uses and I wouldn't totally walk away from using them, but they have their flaws as well.
  • I run a new company called Cequity . We have a blog at http://blog.cequitysolutions.com/

    What is the best way to analyze Blog traffic? Also any idea about a tool called Hubspot that I have just started using.
  • Hubspot is a good place to start. Love what they do and know they do good work.
  • aprebluda
    Aaron from Compete here with a few points:

    - Compete takes out any traffic originating from outside of the U.S. – in other words, if a user comes from India, Canada, or any other non-U.S. country to a site, this visit will not be reflected in our numbers. Additionally, we count people only (i.e. eliminate all bots, spiders, and other automated site crawlers)
    - While Compete is panel-based, local analytics providers track user activity by setting cookies through some element on the web page (i.e. a pixel, an iframe). So, if the user deletes their cookies and clears their cache, they will appear as a new visitor the next time they access the site. Since our approach is not cookie-based, we will only count that person once, which is a reason why the numbers never match perfectly
    - It is tough to compare traffic numbers for very small sites (i.e. under 5000 UVs/month) as there is always going to be more noise with these sites due to fewer actual data points in Compete's panel. On top of that, if the sites are not consumer sites (like jasonfalls.com) they are not going to be as well-represented in the data
    - An important clarification regarding Compete PRO (our paid product) is that the data is based on actual consumer BEHAVIOR (in other words, from a Search perspective, what actual keywords consumers are typing into Google, Yahoo, etc). A smart online marketer understands the value of this behavioral data and can use it to compete more effectively
  • Much thanks for the clarifications, Aaron. Good to know Compete is watching and responding. Hopefully, the feedback here is helpful. I and my audience thank you for your time on the front end giving me information and the effort to come back and make sure you're well represented.
  • Cliche warning: you get what you pay for.... Alexa & Compete are only *okay* even for some sort of competitive analysis. Hope your shout out gets you something "free" from the others. You may need to get in line (another cliche) though...
  • I would concur, Barbara. But keep in mind, the point of the series was to review the free tools out there. I don't know how much I trust the paid ones, either, but it's what we've got.

    Thanks for commenting.
  • I don't buy too much into the data, but I like that they allow users to export the data into .csv files, which I can then use the data to run a quick analysis. They allow you to export 12 months of backdated data. Btw, looks like the skewed numbers are helping you out a bit ;).
  • Good thoughts. Thanks for the input. Exportable data certainly makes sites more useful.

    Thanks for commenting!
  • I recently did similar comparisons on a few sites, too, Jason, and my conclusions were the same as yours. I compared sites with a little and a lot of traffic, and Compete was way off on both. Way too low on the 'mass appeal' site, and way too high on the 'appeals to tech crowd' site.
  • And for those of you who don't know, Matt McGee is SmallBusinessSEM.com and damn smart. Thanks Matt. Good to see you here again. Long time no see in person. Hope that fixes itself soon.
  • Maybe the Google Analytics numbers are way off and the compete.com numbers are a better estimate. I've heard that google's numbers don't always reflect the most accurate data.
  • Not sure if I'd go that far, Craig. I trust Google more than Compete since Compete's model is based on assumption and guesstimates. Google, though I could be wrong, has no reason to not report the information its counter finds, which I have always trusted. Again, I'd like to see some evidence to the contrary.
  • This is really useful stuff, Jason. Much appreciated. My problem is that I've just launched a site that I think is pretty interesting. People are visiting it. Occasionally they're blogging or twittering about it. But the numbers (according to Google Analytics) are pretty small.
    That brings me to my problem. I know that the metrics I look at for my site depend in large part on why I have a site in the first place . . . what it is that I'm trying to achieve. But a part of what I'm trying to achieve is to get some people reading my site and ideally, sticking around to have a dialog about it.
    How can I tell what numbers I OUGHT to be getting; irrespective of any further ROI related metrics?
  • Ah the new blogger, low traffic syndrome. (Kidding). What SHOULD you be getting? Probably what you are. Keep in mind it takes time to build a consistent flow of traffic to a non-commercial, corporate or media site. Without advertising support or mainstream media play, you're relying on word-of-mouth. That, my friend, is a slow burn. I was less than 100 uniques a day for a LONG time on SME before traffic started to pick up. The longer I blogged, the better I blogged, the more I networked, attended conferences, etc., the better the traffic got. I'm now between 700-1100 uniques each day, which I think is good for a niche blog. Add to that the over 3,000 RSS subscribers (roughly since I don't take the time to login to Feedburner and subtract out the click throughs) and I've got a pretty good sized audience. It will continue to grow, too, as will yours. It just takes consistency and time.
  • Jason -- Compete is relatively new to this space, with Alexa being around for much longer. Any of these free services should *only* be used for trend analysis. Most rely on a community of web users who happen to install a plugin to track traffic. Unfortunately there isn't anything reliable unless you have access to the server/traffic logs. I've seen people apply some weird calculations to "account for the discrepancies", but it's all BS.
  • I'm starting to be in agreement with your assessment Matt. It seems that trends and relative comparisons are the useful points for these types of services. Compete reports their paid tools are really more for comparative analysis, which I agree with. They also claim theirs is the best at tracking behavior online. They do have a powerful tool, but I agree it's best used without focusing on the numbers.
  • Great analysis and review, Jason. Greg and I were having a conversation about competitors sites and your post has cleared up some of the questions I had in my mind. At this point, I personally think that looking at the competition is secondary to building and implementing a strategy around our own site(s).
  • True Chris, but knowing the competition is important, too. Still you're right. Focus on your own house and keeping it in order and you accomplish a lot of what you need to do.
  • Just a quick question you've probably taken care of, but I didn't notice in your post - when using the Google Analytics figures, I take it they're U.S. only traffic? Otherwise the international traffic could explain why there's such a large discrepancy in some cases?

    As the sites I run and work with are UK/European, I only use Compete as a way to get a rough comparison - as I do with most tools as analytics are never 100% accurate, and also don't transfer from one service to another with any reliability!
  • Fred
    another free place to go for stats is http://www.webtraffic24.com
  • Good question Dan. U.S. visitors was 766,000+ so the numbers are still way off. Compete indicated there may be widgets or cookies or something in the way that might skew things, but the site in question has none of those that I can tell.

    Good advice on the reliability of these. I'm also at a skeptical point with the accuracy, knowing they can't be 100-percent right. Still, I'd expect closer.
blog comments powered by Disqus