I’ll bet you have a few things that are always on your mind as you move through your day. Me too. One is something I’ve mentioned before: Statistics Lie.
It’s like the conversation about “Perception is Reality“. Except that in the case of statistics, the goal is to intentionally manipulate the perception that other people have of realities surrounding your stories. Reality doesn’t develop as a result of perception, it’s manipulated.
I thought about this today when I was looking at the statistics provided by Shareaholic about how often readers here send what they find to other places. Shareaholic is the company that helps put all those little icons at the bottom of the posts in Answer Guy Central Changes, and they let me see how often you share our work—and where.
Here’s what Shareaholic says:
- In the last 30 days, Shareaholic helped share content from Answer Guy Central 61 times
- That makes us the 676th most-popular site among Shareaholic users
- There are over 415,000 Shareaholic users
- The guy who’s most popular has almost 6,000 shares in the same 30 day period that Shareaholic reports us as having 61.
If I wanted to spin these statistics in a positive way, I’d tell you that according to Shareaholic, Answer Guy Central is more influential than 99.8% of all web sites (676 against 415,000). Or better yet, I’d leave out the “according to Shareaholic” part and just quote the statistics as reality: Answer Guy Central is more influential than 99.8% of all web sites. Wow! Look at Me!
Here’s another example of how statistics can be made to lie:
Hubspot reports that our marketing of Answer Guy Central is more effective than 98% of all web sites’ marketing efforts. Hubspot also reports that Answer Guy Central has more traffic than over 96% of all web sites.
It all sounds pretty impressive, doesn’t it? I hope so, because Perception is Reality.
But here’s an alternate perception of the same reality: 61 engagements via Shareaholic is pretty lame when you compare it to 6,000 (that guy at the top is 100 times better than we are) and while being in the top 3.84% of all web sites sounds great, the fact that almost 1.2 million sites get more traffic that we do is pretty awful—especially when you consider that the site reporting that statistic to Hubspot doesn’t bother keeping details on any sites outside the first 100,000.
Yikes.
Then, there’s the centrist position: based on the statistics we gather ourselves and those being reported to us by Hubspot and Shareaholic, I’m pleased with the results we see from our marketing efforts. We do this all using long-tail marketing techniques and Search Engine Optimization, using the same successful marketing techniques we apply for our clients.
So am I saying that you should ignore or mistrust what Shareaholic tells you, or what comes out of Hubspot’s tools? No. We use Shareaholic because it gives our readers an option to be social, and we’re planning to keep using Shareaholic. And while we don’t use and generally recommend against Hubspot, we believe that the information they provide has value.
Just remember that statistics lie. And THAT perception is everybody’s reality.
Repost This Article
Statistics do not and cannot lie. Statistics are only a manipulation of data. Should the method of manipulation be based on a shaky premise then…
These posts of yours are poor…and i’ve perused a few. How you’ve been in business for 7 years seems to be a testament to the stupidity of small business owners. It’s certainly not your business acumen.
Be sure, i’ll never be back. I’d hope that you leave this comment for the future reference of others, but i find that doubtful. Doesn’t appear that many others visit anyway, at least judging by all of the comments they leave.
Nice personal attack, lacking any substance. I hope you appreciate that I’m taking you seriously enough to let it stay here, contrary to your prediction. And that I think you’ve given me something new to write about (thanks!).
I do wonder why you posted here twice, minutes apart using two different email addresses that I presume are fake (here’s the other one: http://answerguy.com/2010/11/02/college-atheletes-paid-sports-agents-business-change/comment-page-1/#comment-4298).
To answer your question: the people who read what I write here do so specifically because they’re smart enough to think and form their own opinions. I could pander to the less smart and blow most of these articles out by about three times the length, but … I guess you just aren’t in the target demographic.
I can live with that, by the way.
We’re definitely listening
Let me know if I can answer any questions.
Jay
—
Quick Links:
Homepage: http://shareaholic.com
Twitter: http://twitter.com/shareaholic
Facebook: http://facebook.com/shareaholic
Get Help: http://www.shareaholic.com/help
Thanks for saying “hi” Jay.
You know, it’s funny; I almost missed the reference … I DID ask if you were listening, didn’t I?
Ah, rhetoric. I’m glad you were, and I hope it’s clear that I don’t in any way blame you guys, Hubspot, or anyone else for the meat of that story! I point out the talking points and the readers and our clients take it from there.
And for my nickel, I think you’ve made it clear that Shareaholic, by listening, understands the place in and importance of that place which you occupy in the communications ecosystem.
[...] You need to trust the data being presented by those tools, and as I pointed out last week, statistics lie. And Hubspot doesn’t even create any of these statistics; they just aggregate the information [...]
[...] people. And talking about Answer Guy Central with one of those people. And I told him about the way statistics lie, using our traffic numbers as reported here as an [...]
Statistics do not “lie”; they simply are. Statistics are often misused however, either intentionally or through poor understanding.
Your phrase “Statistics lie” is more misleading than the statistics you seek to pick apart.
Martin, you aren’t wrong about that stuff, and I acknowledge that my rhetoric can be read that way. Perhaps a more accurate way to put it would be “statistics can be molded to say almost anything, so watch out for that”. Of course, that’s a bit clumsier than “statistics lie”.
In assuming that my readers are smart enough to see the difference, I’ve stepped on your sensibilities, and I apologize for that!
But … you can bet I’ll do it again
[...] as the information becomes more overwhelming, but Oppenheimer’s article reminded me that it’s easy to twist information to say what we want it to even when the conclusions make no sense and are morally [...]
[...] Hubspot? A great service, but a bad idea for most of its potential customers. And the statistics that Hubspot provides? Not especially trustworthy unless your perception/reailty curv…. [...]
[...] is the same: reviews need to be REAL, or they’re useless (or will be used in some sort of statistics-bending kind of way that makes them [...]
[...] Statistics Lie. [...]
[...] as we know, statistics lie, and can be misinterpreted even by the people who put them forth. So lets look at some raw [...]
[...] the numbers together any way you like; as we know, statistics lie. Ask whatever questions or you think are right about the audience for blue humor in general or Ron [...]
[...] the face of new statistics showing that Google’s grip on the search engine market is getting even stronger, the folks [...]
[...] almost every restaurant ends up with three or three-and-a-half stars on a scale of one to five, rendering the information useless. Or worse, fake reviews get posted, either very low by competitors or people with some other ax to [...]
[...] Sounds weird, right? Where do I get the idea that Facebook Timeline being released after a two-month delay is a chance to talk about SEO? Or Long-Tail Marketing? Or Statistics? [...]
[...] than the reality you thought you were shooting at. And as I’ve said many times right here, statistics are the ultimate example of perception and reality not really having much to do with ea…, at [...]
There are three types of lies, Lies, Damn Lies, and Statistics.
Data doesn’t lie, but statistics does.
Thanks, Michael. But let’s give attribution where it’s due, shall we? After all this IS the Internet!
[...] for being honest in their marketing material? Well, no, of course not; statistics lie, and perception is reality, right? But by making up information, they’ve crossed the line, [...]
[...] net gain while Windows has lost share. That’s fine, and speaks to how easy it is to make statistics say whatever you want them to say. What I see, though, is that Microsoft has managed, even as everything their business was built on [...]
[...] Statistics Lie. No, really, statistics aren’t meaningful. I mean, seriously; statistics say that people don’t trust Facebook . . . but also don’t care, and use it anyway. [...]
[...] matrices, Hubspot Grader doesn’t “mean” anything; it’s an example of statistics lying, and a snapshot of a particular point in time compared to a specific set of criteria. You have to [...]
[...] raw data equates to 2.54 pages per visit, and Google Analytics’ data comes to 2.55, which are statistically identical. So the two data sources are seeing something that matches, but other things (most things) that [...]