Why you really need to be doubtful of Chitika’s market-share reports

banner 468x60
banner 160x600

I do not have idea how efficient Chitika is within their character because an online advertisement network. But they’ve been a smashing achievement at getting the media to uncritically report their increasingly frequent “analysis reports.”

For the previous years, Chitika has been churning out confident reports chock full of information plus feeding those to a hungry technical hit. The approach has paid off for this Massachusetts-based online advertisement network, that has been inside company for almost a decade. Their reports have been approvingly quoted inside a Who’s Who of tech blogs, including TechCrunch, AllThingsD, the Los Angeles Times, plus even ZDNet. Those magazines have reprinted Chitika’s sweeping conclusions regarding the adoption rate of iOS versions, relative marketplace shares of Android plus iOS equipment, plus adoption rates for pills, to name merely a some.

Earlier now, the own Adrian Kingsley-Hughes passed along a Chitika report which advertised “the iPad’s Internet use share fell with a dramatic 7.1 percentage points following December 25.” And according to Techmeme 8 different tech websites reported the same story, citing which amount because an unquestioned truth.

Really? In what universe does which form of shift inside marketplace share result virtually instant? Hint: not this 1.

I love data-driven journalism over simply regarding anybody, however, Chitika’s numbers don’t pass the sniff test. I’ve prevented composing stories based about Chitika’s countless hit releases exactly considering I couldn’t discover any evidence which the numbers were reliable, trustworthy, consistent, or meaningful. On the contrary, the numbers have frequently been only laughably inconsistent.

For illustration, last May Chitika breathlessly reported which OS X use was up 2.58% month over month (from 11.44% to 14.02%), whilst Windows use was down 2.99%. There was no cause for the dramatic shift; it simply occurred. However 1 month later Windows use improved dramatically whilst Mac use was down greatly. Meanwhile, inside the month of July, Chitika reported which half of all Linux consumers abruptly stopped utilizing their preferred working program, with Linux share dropping from 2.05% to 1.12%.

Really? Really? Those kinds of extreme variations recommend which the underlying information is regarding because exact because a daily horoscope.

This morning I fired off a note to Chitika asking when they might offer details regarding their methodology. I got an immediate reaction from a click contact that pointed me to a page containing this terse description:

Chitika Insights is the independent analysis arm of online advertisement network Chitika. Our team of passionate plus innovative Data Solutions Engineers provide a series of reports every week uncovering specific styles inside Internet use. Research research are performed by tapping into Chitika’s extensive network composed of over 250,000 websites, totaling hundreds of millions of advertisement impressions every week. The test size chosen for all Chitika Insights research is big plus diversified, guaranteeing an impartial test plus permitting you to draw better conclusions because to what exactly is happening found on the Internet over North America.

That’s not what I was interested in. As any statistician may tell we, merely having a big test size doesn’t imply we receive valid conclusions. Garbage inside, trash out. Doubling or tripling the amount of information merely makes for a bigger pile of trash.

I replied with a request for more details regarding the information set plus regarding the skills of the individuals assembling these reports. (Is there a trained statistician inside the home?) Chitika has not yet replied.

Meanwhile, a small poking about at Chitika’s url makes me more doubtful of the standard of the company’s analysis. The Publishers page, for illustration, where a website owner may signal as much as run Chitika advertisements, contains reviews from supposedly happy clients. I refreshed the page repeatedly plus accumulated a list of 25 websites providing glowing reviews regarding their achievement with Chitika’s advertisement network.

Of those 20 websites, 10% lead to 404 or 401 mistakes. Another 20% led to websites that have been parked with a registrar plus serving up nothing yet spammy hyperlinks. Another 20% were redirected to domains alternative than the authentic, or went to crappy content farms or make-money-fast websites.

Here, for illustration, is a Chitika testimonial from 1 supposedly pleased customer:

chitika-oneinchgroup-testimonial

But this might be what we see in the event you see which website:

chitika-oneinchgroup-website

Yes, which site is parked at a domain host, plus there’s not a Chitika advertisement inside sight. It’s a pattern I saw over plus over again whenever I checked out those reviews.

In my research, half of the internet websites which Chitika promotes because achievement stories either don’t exist or exist only to serve advertisements. The remaining 50% appeared mostly weak plus sad. Many websites hadn’t been up-to-date inside months or years, plus just a few looked like they represented severe continuous companies.

As a possible advertiser, I wouldn’t be impressed. As a journalist, I question whether the same sloppiness displayed about this marketing page extends to the company’s analysis.

And then there’s this FTC information launch from March 14, 2011:

The FTC reached a settlement with online advertising firm Chitika, Inc. which ends the company’s allegedly deceptive practice of monitoring consumers’ online escapades even following they have selected to choose from online monitoring about Chitika’s url.

[...]

The FTC alleged which inside its confidentiality plan the firm states it collects information regarding consumers’ needs, however, enables customers to choose from having cookies located about their browsers plus getting targeted advertisements. The confidentiality plan involves an “Opt-Out” switch. Consumers that visit it activate a content which states, “We are opted out.”

According to the FTC problem, from at minimum May 2008 by February 2010, Chitika’s opt-out lasted just 10 days. After which time, Chitika placed monitoring cookies about browsers of customers whom had opted out plus targeted advertisements for them again. The FTC charged Chitika’s claims regarding its opt-out system were deceptive plus violated federal law.

Put those pieces together plus we don’t receive a pic of the organization whose information ought to be trusted about its face. Yet big technical information websites (including ZDNet) regularly do precisely which, plus inside the procedure they provide positive publicity to a business which refuses to need it.

Maybe the upcoming time we read a report based about information from Chitika, we can send the reporter a link to the post plus ask when they’ve got a 2nd source.

banner 468x15
author