Biznology Blog: December 2006
« November 2006 | Main | January 2007 »
December 28, 2006
The Useless Web Metric of the Day
I once read about a railroad company that found that most accidents involved the caboose, so they decided to leave off the caboose from then on, for safety reasons.
I was reminded of that story when reading Avinash Kaushik's great post today on the uselessness of the "top exit page" report. I've had the same thought myself many times (like whenever someone shows me that report), but I never bothered to write a blog entry about it. Avinash is right on all counts. He says that when you have a very well-structured experience (your customer dropped an item in the cart and fainted dead away before the confirmation page), then you want to use multivariate testing to fix the pages where they exit. But 98% of your visitors exit and fixing most exit pages will just make them exit somewhere else—fixing that page is like leaving off the caboose. If the caboose wasn't the last car of the train, then another one would be.
Posted by mikemoran at 2:47 PM | Comments (0) | TrackBack
December 20, 2006
What's Google's real market share?
Rich Skrenta did some quick calculations based on search referrals from a few large sites—his research shows Google's real market share might be as high as 70%. Ironically, yesterday was the day that Nielsen/NetRatings clocked in with its numbers, with Google still listed as barely below 50%. What gives?
As someone who has been quoting Comscore numbers that show Google's share of searches in the 40s, I've long known that this number may have no relation to the number of referrals drawn to a particular site. When I worked at ibm.com, we routinely saw our search referrals from Google exceed 70% of our totals, and I've spoken to other companies with the same experience.
I've always theorized that savvy Web searchers are more likely to use Google, because when they formed their search habits a few years ago, Google was the trendy techie search engine. Yahoo! used Google but geeks wanted to show off they knew the real deal. And the Google toolbar effectively locked in the choice among this group. So, I thought that IBM's customers might strongly prefer Google, but more consumer-oriented sites might skew higher to Yahoo! and MSN.
But when I look at Rich's list, I don't think of the USA Today as a geek mecca. That's exactly the kind of site that I would have expected Yahoo! referrals to be high on. But it's not. Rich believes that Comscore and Nielsen and the rest simply have it wrong. Rich doesn't have scientific evidence that proves his assertion, but it certainly does make you think.
I wonder whether Comscore or Nelsen will comment.
Posted by mikemoran at 8:13 AM | Comments (1) | TrackBack
December 19, 2006
Multivariate testing in action
I've written before about multivariate testing, the best way of optimizing your Web site to drive conversions. I was lucky today to speak with the president of ReallyGreatRate.com, Matt Schaub, who told me his story of how he's improved his Web marketing through multivariate testing. If you're not using this technique, read on to find out how it works.
ReallyGreatRate.com is an online financial lead generation company that connects lenders to people looking for student loans, debt consolidation loans, and home mortgages. The company does all of its marketing online, passing loan applicants to up to four lenders (getting compensated for each valid lead passed). You can imagine how valuable it is to increase the number of leads passed, which requires convincing more people to complete the online application form. 
But how do you do that? Everyone has an opinion of what should be changed. Multivariate testing boils those opinions down to testable sizes. Matt started with the submission form you see above. He worked with Optimost, a leading multivariate testing vendor, to identify 13 separate areas on the page that could be modified, and came up with seven or eight different treatments for each area.
For example, the masthead at the top of the page was one such area. Matt worked with Optimost to come up with several variations, such as changing the location of the image to the left side, changing the words in the tag line, and many others.
Optimost ran several rounds of testing to cycle through each variant, identifying the ones that seemed to outperform the rest—the ones that correlated with more loan applicants pressing that "Submit" button. Within four months, a winning design was chosen. 
It doesn't look all that different, does it? It's a bit less cluttered on the left side, removing some things that proved more distracting than helpful, and the most noticed text in the middle of the screen has been reworded to emphasize starting the process. But it looks very similar to what Matt already had.
Here's what wasn't so similar: Form completion went up 116%! A new form that didn't look all that much different than the old one had dramatically different results. Now, no expert could look at these two forms and know for sure which would test better. And, remember, thousands of different variants were tested, which no expert could do either.
So now Matt is happy he's got the best one, right? Wrong. "I'm starting a new round of testing this week," he told me. "I'm going to constantly change my form. You have to! What works now may not work next month."
When I tell people to "Do it wrong quickly," this is what I am driving at. Experiment. Measure. Keep what is working and change all of it, working or not. Matt advised all Web marketers, "Never get complacent."
Posted by mikemoran at 12:01 AM | Comments (0) | TrackBack
December 18, 2006
Thank you, Avinash
If you're interested in Web metrics and you don't read Avinash Kaushik's blog, Occam's Razor, you don't know what you're missing. Many top bloggers do short posts several times a day, but Avinash prefers longer pieces written less frequently. If you try him out, I think you'll like his style, too. Sometimes he does interviews with folks in the industry—today's is with me. So I hope you check out Ten Minutes with Mike Moran. Thanks, Avinash.
Posted by mikemoran at 7:08 AM | Comments (0) | TrackBack
December 16, 2006
Tag Along with Me
So here's an odd form of Internet marketing. Bloggers have started a game of tag, where they name five things about themselves that few people know and then tag five other bloggers to do the same. I got tagged by Marshall Sponder, so here goes. It feels like a chain letter, but, gee, all the kids are doing it.
Five things you probably don't know (and maybe don't want to know) about me:
- I have four kids aged eight through fourteen, and my oldest (David) has Down Syndrome, so we are members of the disability community. My wife and I have an e-mail newsletter that helps parents set up a Life Plan to take care of their disabled family member when they no longer can.
- My wife, Linda, in addition to being the erstwhile editor of my books, is an author herself with a popular book on how to ditch diets and be at the right weight for your body. Several years ago, we both lost a lot of weight this way and we've kept it off. (OK, about ten pounds have crept back onto me from the 70 I lost.)
- I have been playing fantasy baseball for 25 years (even before the original Rotisseries baseball book was published—we made up our own game). I finished next to last in 2006, but I have won the league several times also, including 2004 and 2005.
- I'm a Chicago Cubs fan. Stop laughing. Any team can have a bad century.
- I'm not an avid biker—just taking trips around the area with the family (including pulling David in a large trailer behind my bike)—but each year I take a bike ride to Yankee Stadium from my home in Ridgewood, New Jersey and park it outside to watch a ballgame. I have a 1970s Schwinn bought at a garage sale for $10, so my bike lock literally costs more than the bike. The first year it took me seven hours round trip (lots of hills in New Jersey), but last year I had it down to three hours. This is as close as I get to feeling like an athlete.
Well, that's it about me. Let me try to find five people that I don't think have been tagged yet. I'm going to focus on blogs that I read from people who don't post every day (like me). I'll tag Mike Grehan, whose stories about travelling the world of search marketing conferences helps me stay connected with people I know even though I can't get to many of the conferences. I've always liked Alex Bosworth's blog, although he's been quiet recently. Everyone reads Danny Sullivan on search marketing, but I also read his personal blog, I can't have any list with one Cubs blog—my favorite is by Derek Smart, although he hasn't posted in a few weeks now (maybe he is stunned by the Cubs unusual off-season activity). And I have to tag my wife, Linda Moran, who has blogs on subjects as diverse as math instruction, normal eating, and parenting teens and tweens. I sent you to her home page to see more about her and pick what interests you.
Posted by mikemoran at 7:23 AM | Comments (2) | TrackBack
December 15, 2006
How much does "free" cost?
Wednesday, I talked about the new free search engine that Yahoo! and IBM have made available for download, We on the product team are gratified at the rave reviews it has garnered, but we weren't as clear about the hardware requirements as we should have been. Larry Dignan of ZDnet had an excellent post about how our free search engine still requires hardware, a point first made by Jason Berberich in a comment to a post on Nick Carr's blog. So I asked a few questions about what kind of hardware we really need.
Normally, people wouldn't spend a lot of time checking out the price of hardware for a piece of software, but this is a special situation, because our new search engine is inevitably being compared to Google Mini and Google Appliance. Those offerings are not free, but they do come bundled with a piece of hardware, so it's legitimate to ask about the cost of hardware required by our free software.
And, in our passionate focus on simplicity, we didn't provide the details on what hardware you needed for different size search needs, we just put in a minimum recommendation for testing and a recommended configuration for our maximum 500,000-document capacity. So it was perfectly reasonable for bloggers to conclude that we were recommending that configuration for any use of our search engine—we kinda said that.
But it's kinda not true. We were being conservative. Our recommended configuration is what you need to run a robust production implementation for half a million documents. So we decided to provide more information (which we should have done in the first place) about what kind of configurations we recommend for fewer documents. We updated our product information page earlier this evening (click the learn more link to see all the details).
So what does that mean for our hardware costs? Let start at the low end.
Our minimum hardware, which can handle up to 20,000 documents, requires a single 1.5GHz processor with a gigabyte of memory, and 80 free gigabytes of disk space. I'm sure that you could find many servers around that met these requirements, but, hey, I work for IBM! So I looked for the most inexpensive IBM server that would do the job and configured an IBM xSeries 100, Model 8486E2U, which sports an Intel Celeron D processor 326 at 2.53GHz, one gigabyte of memory, and 160-gigabyte disk drive for $843.
Now it's not fair to directly compare the Google Mini to our search engine, because they have different features that might be more appealing to some customers than others. Although we've been praised for how simple our product is to set up, some would argue that it's easier to plug in a Mini. Other folks who have a server lying around would claim our solution really is free. Still others would remind you that their Web server is maintained by a hosting company. They can't exactly stick a Mini into that hosted environment. So, different people have different opinions and different needs.
But for those people that have really have a choice, price is important. And Google Mini's lowest price is $1,995 (although that buys you 50,000-document capacity, not our 20,000). So, you could legitimately ask if Google's cost of ownership is lower than ours for 50,000 documents.
Luckily for us, it's not. For 100,000 documents, our newly-minted recommendation is for a single dual core processor of at least 2GHz, with one gigabyte of memory and 100 gigabytes of free disk space. I priced an IBM server (xSeries 100 Model 8486E7U) configured with a Pentium D processor 820 at 2.80GHz, one gigabyte of memory, with a 160-gigabyte disk drive for $1,083, less than Google's price for half the number of documents. Google's price for a 100,000-document system is $2,995.
We also looked at the high end of our capacity, the 500,000-document configuration. What would it set you back to buy a 3GHz dual processor server with two gigabytes of memory and 250 gigabytes of free disk space? I was able to configure an IBM System x3400 Model 797312U with one Intel Dual Core Processor Model 5050 and one Intel Dual Core Processor Model 5110, with two gigabytes of memory, and a 500-gigabyte disk drive. It cost $2,866, compared to $30,000 for the low-end Google Search Appliance that handles 500,000 documents.
To be fair, Jason believes that the hardware you get with a Google Search Appliance is priced at $8,000, so spending less than $3,000 might not be a fair comparison to the power you'd get. Still if you want to, you can go buy the $8,000 server to use with our free software and get some change back from your $30,000.
Now those who know me understand that I am no expert on computer servers. (They are laughing at this point because they know how much I am not an expert—you know who you are.) So I apologize in advance if I have made an error here. If you want to have mirrored disks and maybe some other features on your servers (which are probably excellent ideas for many uses), these prices would probably be higher. But my suspicion is that even with these extras, our free software helps you come out ahead. Larry calculated that a server meeting our 500,000-document requirement with mirrored disks still cost less than $5,000.
So, to sum up, the bloggers making this point about our total cost of ownership are absolutely correct that we should talk about hardware costs, too. And their analyses were based on exactly what we wrote on our Web site. But as they say in the NFL, "Upon further review," we were being overly pessimistic about what we require—in our zeal to be simple. Sometimes simple can be oversimplified, and we needed to correct that. Thanks to the bloggers who raised this point and gave us a chance to explain ourselves better—let's keep the conversation going.
Posted by mikemoran at 8:03 PM | Comments (0) | TrackBack
December 13, 2006
Is your favorite price for a site search engine "Free"?
Mine is. Amazingly, it's my favorite price for everything. I'm just funny that way.
That's why I was excited to take on a new role in IBM a weeks ago. After eight years at ibm.com, I have a new job as a product manager for IBM's OmniFind search products. I've been talking to the OmniFind team for several years, learning a lot from them and offering the occasional suggestion about their enterprise search products, OmniFind Enterprise Edition and OmniFind Discovery Edition, but when they asked me to join the team, they told me about a new product they were cooking up to go after the requirements of smaller businesses that can't afford the time or expense to put a good search engine on their Web sites, their intranets, or their departmentmental servers. And today, we can announce a free search engine, in partnership with Yahoo!, called the IBM OmniFind Yahoo! Edition.
The first thing you notice is that it's free. (Did I mention that?) It's not a trial. It's free. Just register your e-mail address with us and download away.
The second thing you notice is that it's not a toy. It handles up to 500,000 documents and provides excellent relevance, a configurable user interface, and many other features you'd expect in a search engine you plunk down thousands of dollars for. (If you need more than what this one does, then you can check out the other search engines that we sell.) This search engine can meet the needs of many sites.
But the main thing you'll notice is that it is simple. It isn't hard to install or configure and it just plain works. If you are running a huge Web site and have to customize every aspect of the search engine to satisfy your needs, this one is not for you. What you are doing is complex and it makes sense that setting things up might take some know-how. But if you have more modest needs, why not use something that doesn't require a degree in searchology to make it work? This one's for you.
I'll be talking about this more as time goes on because it's an interesting story. It's interesting that the search engine is built on the powerful open source Lucene package. It's interesting that IBM is contributing code back to the Lucene base. It's interesting that IBM has a community approach to support and requests for enhancements.
But what's really important are three things. It's simple. It works. It's free. Try it yourself and let me know what you think.
Posted by mikemoran at 9:19 AM | Comments (0) | TrackBack
December 5, 2006
More thoughts on unifying Web metrics
Marshall Sponder had some thoughtful feedback on my newsletter this month on unifying Web metrics. Marshall agreed that metrics are too hard to measure accurately right now, and added some more depth to how hard it is to know whether someone is qualified to see your promotional offer and how tough it is to figure out how many times your blog gets read.
After reading Marshall's concern about my statement that "Until something better comes along, we can make the simplifying assumption that the impressions of your blog posts in blog readers are equal to your number of subscribers," I realized that I had not been clear in the original post. What I was trying to say was that you can make the assumption that whatever number of subscribers you have is the best surrogate we have for blog impressions. The way I wrote it, it sounded like the reverse (use your blog impressions to estimate subscribers). What I was trying to say is that because so many blogs are read within RSS readers with no clicks to the blog Web site, the (admittedly poor) estimates we have of number of subscribers is the best guess at how many impressions each blog post gets. Yes, I know that it's wrong, but it's better than nothing. I suggested later in my newsletter a few ways that we could refine the estimates of all of these counts to be more accurate.
Thanks to Marshall for pointing out an area were I was not being clear. You can check out all of Marshall's thoughts at WebMetricsGuru.
Posted by mikemoran at 8:18 AM | Comments (0) | TrackBack
December 1, 2006
Crash Course in Search Marketing for $50
For those of you in the New York City area, I'll be teaching another one-night search marketing class on December 11 for the Learning Annex. You get the basics on what search marketing is worth to your business, how to succeed at organic search marketing, and what to look out for with paid search also, all in less than three hours. I'll answer your questions, too, and give away a free copy of our book to one lucky attendee (and have it on sale for the rest). I'll keep it all free of mumbo jumbo, just like our book—technical folks will understand the marketing stuff and vice versa. I hope I get a chance to meet you on December 11.
Posted by mikemoran at 4:04 PM | Comments (0) | TrackBack
The Unified Field Theory of Web Metrics
If Web metrics has always seemed like an overly complicated endeavor that left you scratching your head, don't feel singled out. The Web is still a baby, changing so fast that we're still learning how to measure it. But if all the newfangled metrics concepts and methodologies leave you wanting for some simplicity, read on. The Unified Field Theory of Web Metrics attempts to condense all Web activity into three concepts, and it's the subject of this month's Biznology newsletter.
Posted by mikemoran at 3:55 PM | Comments (0) | TrackBack
Subscribe to daily feed
Subscribe to daily e-mail
