Archive for 15 July 2011

Android versus iPhone: an analogy

If you could have either a Porsche or a Trans-Am, which would you choose?

If you could have either a Porsche or a Trans-Am that would turn into KITT (from Knight Rider) with the push of a series of buttons, which would you choose?

First, answer the second question.
Second, tell me whether it’s an appropriate analogy for the decision between iPhone and Android, in your opinion.


Yo dawg! I heard you like SiteCatalyst. . .

I’ve been asked a few times recently how a company can report on SiteCatalyst dashboard usage by its own employees. If you’re running an analytics practice, there are a few reasons why you might want to do this—for example, knowing which dashboards are most (and least) popular helps you manage your list of dashboards and optimize your dashboards by learning which features internal consumers prefer. (I know Adam Greco will like this! See his Idea Exchange submission here.)

You can set this up in just a few simple steps.

First, create a new report suite to capture dashboard usage. My RSID for this report suite is “gainescorpdashboards.”

Second, on each of your dashboards, click “Open Editor” (in v15, this is the “Layout” button). Under “User Content,” choose to add an image reportlet to your dashboard. Put it someplace inconspicuous, such as at the very bottom of the dashboard. The image URL should be a hard-coded beacon, like this one:[AQB]pageName=GainesCorp%20Main%20Dashboard&c1=Ben%20Gaines

(Set the protocol to https because SiteCatalyst uses https and that’s where you’re viewing the dashboard.)

Add a hard-coded beacon to a reportlet

As with many old-school mobile implementations, you’ll need to hard-code the values that you want to capture. In the example above, I’m keeping it really simple: page name is the name of the dashboard, and prop1 (c1) is the name of the dashboard author/owner. Another possibility is to use a list prop (or, in v15, a multi-value variable) to capture data about the various reportlets in each dashboard. Remember to URL encode!

Third, save that reportlet and save the dashboard changes.

Whenever a user loads that dashboard, a simple hard-coded beacon will be passed into the new report suite that you created. You’ll have pathing enabled on page names, so you can see how users move from dashboard to dashboard. The depth of the data isn’t anything to write home about, but (as I said above) an implementation like this can yield insights that help you better manage your use of dashboards and better serve your users.

For example, take a look at this Next Dashboard report, which would have been possible based on the example described above:

Next Dashboard Report

People are going to leave dashboards. That doesn’t necessarily concern me. However, it’s very interesting that 37% of my users are switching to a different dashboard at some point later on in their SiteCatalyst session. I can begin to ask what they’re getting out of “Fall 2011 Campaign Dashboard” and to explore whether something needs to be added to the main dashboard to better serve users. This is an extremely rudimentary example, but hopefully you can see the point.

There are a few limitations that I should mention.

  • We can only capture static data about the dashboard. This isn’t a JavaScript implementation, and the image reportlet doesn’t accept JavaScript. This means we can’t capture the username of the person viewing the dashboard.
  • This reportlet needs to be added to every dashboard individually. You can copy a dashboard, but you’ll still need to edit the reportlet to reflect the new dashboard. Some effort required.
  • This doesn’t work for distributed reports. A beacon won’t be fired when someone opens the dashboard in PDF form. Depending on your situation, that may be a non-starter or it may be completely irrelevant.
  • People may ask about that weird blank reportlet. Yep.

So there you have it. A (relatively) easy way to capture some rudimentary (but valuable) data about your company’s dashboard usage in SiteCatalyst. Don’t just assume people are using the dashboards you’re building! Hopefully doing some basic analysis on dashboard usage will prevent you from wasting time on pointless dashboards and allow you to optimize your dashboards to better serve your company’s needs.

Netflix, consumer surplus, and brand loyalty

Having lived through the occasional PR disaster in my previous life as a Community Manager at Omniture, I sympathized with the people responsible for managing social media for Netflix following the company’s rate hike for those who receive DVDs by mail. Today cannot have been a fun day.

It led me to think about the relationship between consumer surplus and brand perception. I’m not an economist, but one of my favorite principles of economics is the study of the difference between the value from a product or service and the actual price that customers pay for it.

I haven’t studied this, but I’d be willing to bet that Netflix (at least prior to the announcement earlier today) had a tremendously high consumer surplus. Until recently, I got unlimited streaming and one DVD at a time for something like $9.99. My family uses streaming constantly, and we’ve discovered some gems via DVD. All of this happened without us ever having to leave our home. High perceived value, low price.

Why did Netflix raise prices? There are likely several reasons. I like my friend’s suggestion: “Netflix needed to divorce mail and streaming so that it can compete with future streamers on their level.” Certainly possible.

When you change the price without changing the perceived value—which is exactly what Netflix did to so many of its customers today—you are inevitably going to price out some segment of your customer base. They slip below the line and no longer receive enough perceived value to justify the cost, and they drop your service. But as long as Netflix maintains enough customer surplus that the increase in prices more than offsets the loss of some business, it’s a good move—at least on paper, in the short term.

But even if customers keep paying, changing that relationship has an impact on their perception of your brand. I spoke with a former colleague who has been a Netflix customer for eight years. She’ll remain a customer (begrudgingly for now), but her loyalty to the brand is shot. Granted, that’s a sample size of one. However, the venom and bile that the public relations and social media teams at Netflix must be feeling today is certainly not limited to this one customer. My Twitter feed has been literally filled with justifiably angry complaints against Netflix because consumer surplus decreased for literally every user.

I’m not sure how you come back from this. Trust is hard to earn and easy to lose, and even if the coming weeks and months demonstrate added value—such as improved quality of service or a larger library of streaming content—when a brand alienates its loyal followers by screwing with the relationship between perceived value and price, it’s hard to win them back.

So, how could Netflix have prevented this? I would start by articulating the value to customers a little more clearly than this statement from Chief Service and Operations Officer Andy Rendich, found in the company’s press release today:

Netflix members love watching instantly, but we’ve come to recognize there is still a very large continuing demand for DVDs by mail. . . By better reflecting the underlying costs and offering our lowest prices ever for unlimited DVD, we hope to provide a great value to our current and future DVD-by-mail members.

Consumers don’t care what your costs are. You can’t sell a price hike to consumers by complaining about your costs. They care about the value they’re receiving for the price. If the issue is cost (and note that Netflix has a solid net profit margin), then your business model was flawed from the start. Netflix torpedoed its own brand equity by decreasing consumer surplus.

Can you think of any other individual companies that have thrown caution—and brand perception—to the wind by increasing prices without satisfying customers’ need to understand how they benefit from the change? (I can think of at least two. I used to love unlimited data on my cell phone.)

For now, Netflix can rest assured that many or most of its customers will likely still see enough value to justify the price. I certainly feel that $16 a month is a good deal; most of the complaints I’ve seen include the disappointed recognition that we’ll all have to pay at the new prices. We can’t live without our movies. Netflix basically has a monopoly, giving it some flexibility to play with consumer surplus.

Netflix is trying to cash in on its tremendous consumer surplus while it can. Is it a myopic strategy? Where a loyal customer base could have helped Netflix survive a future onslaught from Apple, Google, Amazon, or another challenger, they’ve sacrificed much of that loyalty today for a quick boost in revenue. The real danger is that, by impeaching its own brand, Netflix has created a sizable market of former loyalists who are now begging and pleading for a competitor to arise with greater consumer surplus than Netflix offers.

And I can’t imagine a more terrifying scenario for a brand manager than one where your customers are clamoring for the competition to succeed.

The move

(I thought this post would be easier to write.)

I don’t want to bury the lead, so I’ll just say it: At the end of this week, I will be leaving my post as Product Manager on the SiteCatalyst team at Adobe and taking a position as Manager of Research Analytics for ESPN. I’m tremendously excited, although I will miss many people, places, and things that my family and I have come to love during our time here in Utah, and specifically at Omniture/Adobe.

(Fortunately, the world is a lot smaller than it used to be. I’m still going to pester you, Jeff Jordan. I’m keeping your number in my phone. You’ve been warned. Oh, and Ambria? I’ll be giving out your e-mail address to everyone who wants to participate in one of your beta tests.)

The past five years at Omniture (now Adobe) have been an honor. I feel it’s important to mention that there is probably one company on the planet that could wrestle me away from Adobe at this time, and that is ESPN. In case my blog hasn’t made it clear already, I’m a sports nut who loves analytics and grew up in New England. ESPN combines all three of those things. (Plus, Bristol is only two hours from Fenway Park.) The point is that I am not making this move out of frustration, disenchantment, or fear about the future.

I don’t want anyone to think otherwise for even half a second.

My greatest concern is that people in the #omniture community that I helped build on Twitter will jump to foolhardy conclusions. That’s the downside of having been one of public faces of a brand on social media—when you leave, it never looks good. I know this because I’ve been in the “rush-to-judgment” camp before. For example, I wondered about Comcast when Frank Eliason left last year. How could Comcast have lost Frank? Things must be really bad for him over there.

How could Omniture have lost Ben? They didn’t. ESPN won me. There’s a huge difference.

I’m looking forward to writing often here as I begin to explore life as a daily practitioner of analytics. I performed analysis frequently as a Product Manager (and previously) at Adobe—as I hope you’d assume, we do use SiteCatalyst heavily to analyze and optimize SiteCatalyst—but I also spent a lot of time on other things. Fortunately, I’ve been taught well by mentors too numerous to name, and I hope to do them proud.

You can expect plenty of continued involvement in the analytics community, as well. In fact, I hope that I can participate in new and exciting ways, now that I won’t be a “second-class citizen” (as described—correctly, I think—by Jennifer Day on Emer Kirrane’s blog). On this site, I’m hoping to continue to write posts similar to those I’ve been publishing on the Omniture blog since 2009, discussing implementation, analysis, and more, as well as whatever else I decide is worth writing.

So, there you have it. If you’re ever in the Bristol, CT area, please drop me a line.