Web Analytics is Recession Proof?

For the past few weeks I have been thinking about the economy and trying to reconcile two seemingly contradictory observations:

1. The economy sucks, and it doesn’t seem likely to improve anytime very soon
2. The web analytics sector is reportedly recession-proof and, in fact, predicted to grow in 2009

While I hardly need to provide any proof of the first observation, evidence for the latter has been emerging from a variety of voices in our community for the past few months. Case in point:

* In January, the Web Analytics Association reported that nearly 70% of companies planned to continue to invest in new tools and staff throughout 2008
* Back in July, Corry Prohens from IQ Workforce wrote in a guest post in this blog that “74% of practitioners expect that spending on web analytics will increase at their company during the recession”
* In September, my good friend Jim Sterne surveyed the web analytics crowd and found that 87% of practitioners plan to maintain or increase their budgets for web analytics tools and services in the face of the current economy, and that nearly twice as many respondents indicated they planned to increase budgets for web analytics (21%) as planned to decrease same (13%)
* Last week, the fine folks at E-consultancy splashed all across the news with the headline “Web Analytics: A Silver Lining in the Recession Cloud” by reporting that web analytics was poised to grow in 2008 by 12%

In the E-consultancy report, the organization’s head of research Linus Gregoriadis was quoted as saying: “The profile of Web analytics continues to grow as it becomes more integral to business decision-making and organisational strategy. The credit crunch is putting the spotlight on analytics as organisations work harder to understand where they are getting the best return on investment and where real value is being added.”

Recently, Josh James, the CEO of Omniture said something similar during the Q&A portion of the company’s Q3 earnings call in response to a question about whether businesses saw web analytics as discretionary:

“Every dollar that a marketer has, I think everyone has in every organization is under pressure right now and certainly marketing spend is where CFOs like to look and see if they can cut. But, what we’ve seen with our customers is their online channels are the ones that are performing the best. Their online channels are the ones that are giving the most direct impact within that quarter that spend is also taking place.

In terms of the way that they think about Omniture, even if they cut let’s say 10% of their marketing spend, they’re going to use us to a) identify the 10% they’re going to cut and b) use us to optimize the other 90% to try to get back up to the same results as they had with the 100% the year before. These kinds of times actually drive usage of our product.

When things are good it’s a lot easier when you want more sales just to throw more money at the top of funnel and to generate more leads and go through the process. When things get bad people try to focus on of everyone that’s already coming to our store, what can we do to keep them more attracted? What can we do to get them to look at other things? What can we do to get them to read additional articles? All of those behaviors drive uses of our product.”

All of this sounds absolutely spectacular. Except for one thing …

I’m not sure I believe any of it.

I think that we are collectively starting to suffer from the echo chamber effect, essentially reiterating that web analytics will be fine in this lousy economy because, unsurprisingly, we are all making money off of web analytics and we would very much like to continue doing so. The WAA, IQ Workforce, my friend Jim, E-consultancy, Omniture, me … our collective businesses are all more or less explicitly tied to continued investment in the sector. So why wouldn’t we look for data that suggests that the picture continues to be rosy and the future bright?

Why indeed.

In terms of the data presented above, as a former researcher I would offer this assessment: many of these surveys appear to suffer from sample bias. Asking the Yahoo! group, members of the Web Analytics Association, or the audience attending Emetrics about their interest, investment, or organizational focus on web analytics is kind of like asking your average Democrat in Portland, Oregon how they feel about Barack Obama. The problem is not the audience, the problem is the interpretation: I think it is misleading to extrapolate the responses from a non-random sample of businesses and business people to the larger audience.

This kind of sampling leads to claims like “52% of online marketing managers are currently engaged in A/B or multivariate testing …” Fifty-two percent implies that tens of thousands of online marketing managers are testing. Which sounds great, except that when Offermatica and Optimost were acquired by Omniture and Interwoven they had a few hundred customers between them, and Stephane Hamel’s WASP tool reports that 0.4% (zero point four percent) of the Top 500 online retailers are using easily detected A/B or multivariate testing tools.

Don’t get me wrong, I too have been guilty of sampling biased audiences, although in the past year I have stopped conducting primary research due to both the sampling issue and the plethora of free research that suddenly appeared in the marketplace.

Ultimately I’m suspicious of this optimistic data that we’re seeing, especially in the context of statements like this one made by Mr. James made on the earnings call referenced above about the effect the economy is having on Omniture’s ability to forecast Q4 and 2009:

“Towards the end of September however, it became apparent that the challenging macroeconomic and financial environment may have some impact on our business going forward although it remains difficult to quantify the uncertainties specifically.”

Mr. James and his CFO specifically don’t want to talk about 2009 on the call. Which makes sense to me, since here are some other data points:

* The economy sucks, and without belaboring the obvious, it appears that this suckiness will stay with us for quite some time;
* While I don’t question Mr. James assertion that his best customers make excellent use of web analytics, in my personal experience this is not universally true;
* Some of the largest consumers of web analytics products are starting to struggle;
* Despite the conventional wisdom that dictates that brilliant analysts are safe when times are tough, I am getting more and more calls from brilliant analysts who are being laid off or being offered severance packages to walk away.

It is this last point coupled with something I learned at Emetrics that has me the most concerned. In D.C. at Emetrics I heard Liz Miller from the CMO Council say that most CMO’s are a few years away from fully understanding the value of web analytics. If Liz is right, and her credentials are impeccible when it comes to the CMO’s office, then given the anecdotal evidence that continues to come in I wonder if web analytics is slightly more discretionary than we’d like to believe.

Don’t get me wrong, I sincerely hope to be wrong in this assessment. As an author, public speaker, evangelist, consultant, and conference co-producer focusing on web analytics I honestly hope to be able to write a follow-up post in six month saying, “Wow, I was really super-wrong about where the web analytics industry was going …”

But what can you do if I’m more right than not? What if you work in an affected sector or work for a company known for their web analytics acumen that is suddenly faced with bankruptcy or worse? What if the folks you work for who profess a great love for data-driven decision making are really HIPPOs in their heart and when the real bloodletting begins are just as likely to look for savings in areas that can be easily cut (human resources, for example) as opposed to those that would require breaking contracts?

What indeed.

If you’re in any way concerned about the current economy and your personal employment situation, here are five tips that I would offer to help you best prepare for the worst.

Tip #1: Focus on Increasing Profits, Not Minimizing Spend

My friend W. David Rhee just published a great response about the relationship between web analytics, sales, and marketing in a down economy. To paraphrase Dave, if the bosses begin to panic, you don’t want to be in a situation where you appear to be an expendable marketing cost that can be cut. It is far better to be focusing your analytical efforts on how the organization can be increasing profits, even if you have to fight to spend more time conducting analysis and less time generating reports.

Essentially you want to take Mr. James statement above to heart and work your butt off to optimize the lower-levels in your conversion funnel, working with what you already have, not what you might be able to attract. The good news is that the technology supports this analysis; the bad news is that more often than not, the deeper you get in the funnel, the more difficult optimization becomes for a variety of reasons, not limited to the business, IT, and “the way we’ve always done it!”

Be a profit center, be big picture, become truly invaluable.

Tip #2: Don’t Be a Report Monkey

The unfortunate reality about web analytics work is that far too many smart people spend far too much time generating far too many reports that far too people actually read and even fewer actually derive real value from. Sound familiar? When I started the conversation about process in web analytics in 2006 at Emetrics, over 80% of the audience said they spent too much time on “reports” and not nearly enough on “analysis” … sadly I’m not confident that things have changed much in the past two years, especially on a percent-of-practitioners basis.

There are any number of great posts about why reporting is over-rated and how the real value in web analytics comes from careful, business-focused analysis of the data, there are still too few companies that have put the hub-and-spoke model into practice and are able to effectively leverage web analytical resources.

My advice to to step-up and find the real value in your data, even if you have to conduct the analysis on your own in the wee hours. It’s not as if you can just stop generating reports (tempting as that may sound) but if you’re a good analyst, taking the time to figure out where the real opportunities to increase revenue are is the work you want to be doing anyway. Taking the initiative to make data-powered recommendations and presenting them is a good way to demonstrate your skills and commitment to the business (but don’t stop doing the job you’re being paid to do!)

Analysts conduct analysis and make recommendations. Be an analyst.

Tip #3: Start Watching the Job Boards

Even if you feel pretty good about the situation you’re in you have to admit that the most accurate term to describe the current economy is “dynamic.” In situations like this the worst thing you can do is be caught off guard and so I would offer that spending a little time surfing the Web Analytics Demystified Web Analytics Job Board (also see the WAA’s version) would be time well spent.

According to the nice folks at SimplyHired the number of job postings looking for “web analytics” experience of some kind continues to increase.