Home
Consulting
Community
Contact

Monday, April 20, 2009

Web Intelligent Part 2

So, we're at #4, you have a voice of customer tool.

Where you need to go: At this point, we're talking about the difference between understanding your online customers and REALLY understanding your online customers. The difference here is like reading someone's birth certificate, and reading their tax return. If you were a car salesman, wouldn't you just love to see your potential clients tax returns, before you chose which one to focus on? (If cars actually sold in this economy anyway!)

If you have a voice of customer / comment card tool, you can already read a lot, but we can go further, much further. Many people have asked me, how far is too far? How big brothery can we get? The answer is that consumers pretty much don't mind you collecting information as long as they give it to you, or your collecting it anonymously. The big caveat here is that the information has to be used to improve the visitor's experience, not for your own black book of secrets.

Next is Customer Experience Management. Basically, these tools give you some serious visibility into how users are interacting with your website. They record actual user sessions so you can review them.

As websites become more and more complex with dynamic functionality, web analytics alone gets harder and harder to interpret. Imagine a sudden drop off in your store cart. Thanks to dynamic content, it's very difficult to see what's going on. You know from your web analytics data that things are going downhill, but you have no idea why and no one has told you through your comment cards.

Enter Customer Experience Management (CEM). CEM systems like TeaLeaf actually record user sessions and let you review them. Think of this like a call center manager listening in on sample calls. "Your call may be monitored for quality assurance...". Yep, so might your web experience.

So, you choose a couple of user sessions that were abandoned, and watch as they go through your cart... hey wait... why is that form returning that error? You've never seen that before... must be a glitch... someone call the developers STAT!

That's just one example of what you can learn with CEM. You don't have to watch every user session every day to see what's happening. All you need is a few typical examples and even a small correction can make a HUGE difference.

5) You have a Customer Experience Management tool

Where you are:Seriously, you ARE web 2.0. You're watching, you're listening, your changing, you're adapting, you're making more conversions. And you're doing it all and not being annoying or creepy about it!

Next HUGE step....
Where you need to go:Business Intelligence Data. You have no idea how difficult this will be. Eric T. Peterson thinks Web Analytics is hard, and I agree. This is... monumental... and the rewards for doing it are equally incredible.

Take all the previously mentioned technologies and information, then link all that with your CRM data, your ERP if you have it, your sales data and any other customer data collection system you might have. Then, lump it all into one well organized data warehouse that you can query at will with business intelligence tools.

Here's a quick needle in this very large haystack: Imagine you're an electronics retailer. I visit your site. I've bought something there before, so I'm logged in, and you know my name, where I live, what I've bought before etc. If you only had this CRM style data, you could send me an email to tempt me to buy something else related to what I've bought before and it may work. I bought a TV, you send me an email about a Blu-ray player. But imagine this.

The systems are linked with the web analytics data. The systems monitor my surfing patterns and quickly "realize" (through a scoring type system) that I'm actually looking at Blu-Ray DVDs, not the players.

Probably, I already have a player. Zoinks! You can now lump me into a segment of people that already have a player and market DVD's to me, instead of a player. That may seem like a downgrade, but if I already have a player, why would I buy another? Send me an email or show me an ad on your site about a buy 5 get one free deal for movies, and I'm WAY more likely to jump in.

The power of web analytics comes from segmentation. This is segmentation right down to the person. Imagine knowing who I am and what I'm interested in. Then imagine showing me an offer based on that information.

Here's another slightly less enthralling, but important example. You manage several web properties with various levels of technology behind them. They have different reporting, analytics, shopping cart and CRM capabilities. A data warehouse can take all that disparate information and combine it into one "single version of the truth" (thanks to Bernie Jeltema, my instructor at the University of California at Irvine Web Intelligence program, for that quote.

That's web intelligence. That combined with targeted content delivery can deliver monumental results.

Ok, that's great, how do I do that data warehousing thing? That's the topic of this website, this blog, and the communities we're trying to get moving. The technology is complex, but not unachievable. The brain power in deciding what data to look at, and how to use it, is the real trick.

Join in the discussion by clicking on the communities tab.

Labels: , , , , , ,

Bookmark and Share
posted by Dan Linton at 0 Comments

Friday, April 3, 2009

How Web Intelligent are you?

Part 1
Wondering how far along your enterprise is with real, actionable web intelligence? Use this guide to pinpoint where you are now, and where you need to go to achieve your targets and greater online success!

1) You have a website with log stats

Where you are: You have a web site, but it hasn’t really made it out of the 90s. It's only there because your company expects to have some kind of site, but it rarely gets updated, and when it does you need an agency to do it for you. Likely your provider gives you some basic site traffic stats in an email once a year, which you promptly delete.

Where you need to go: If you actually expect to recruit new customers or keep new ones, you need to step it up. Get a web marketing coordinator or at least a webmaster and welcome yourself to the year 1997. If you're this far back, you'll need some specialized help, a complete redesign of your site, and a company commitment to the web. If you can't or won't get it, prepare to be left behind.

2) You have basic web analytics reports

Where you are: Ok, you've got a website, you have a department responsible for it, and you even have a web analytics tool installed on it. By basic I don't necessarily mean free or low cost (Google Analytics for example, which isn't basic at all), I mean basic analysis. Maybe you spent $100G on an enterprise level implantation of WebTrends or SiteCatalyst. A bunch of your employees then went for training on the tool, then went back to their normal day-to-day duties with the expectation that the tool will tell them something useful.

Shockingly, nothing changed since then, except you're down $100G in budget. Marketing campaigns are plodding along, and your executive team gets a report once a month exclaiming "visitors, visits and page view numbers are flat." Well that's deep. CEO then calls the VP marketing into the office and asks why the "visitors, visits and page view numbers are flat."

Umm.. we've reached full market penetration?

Two things are wrong here: 1 - the company as a whole and the executive team in particular don't understand the web as it affects the business, and 2 - the people responsible for the website are usually so engrossed in their everyday activities that analytics becomes an unimportant to-do task once a month.

Believe it or not, a lot of very big companies with very big websites that do a lot of business on the web are in this stage.

Where you need to go: You need a web analyst, or a web analytics department to start you on the road to real web intelligence. A dedicated resource who a) understands the software and technology being used, b) understands your business, and c) can relate not only basic stats but meaningful analytics information to a variety of key stakeholders at various levels. Where does this person come from? A lot of enterprises snagged someone out of IT for this role. They're great with software, technical implementations and spreadsheets. Unfortunately a lot of these folks, great as they are, have never been required to understand marketing in any sense. As a result, they may not (or they may, don't get me wrong here, I'm not saying that no IT person could ever do this, I'm just generalizing) have the critical understanding of eMarketing and the actual business goals to truly interpret the data.

In my experience, try for an experienced web analyst with a marketing background. Alternatively, find an extremely geeky marketing person and get them training on the technology. The key is that the person been naturally analytical and have a curious mind. There is no more critical question in the world of web intelligence than why?

3) You have a Web Analyst(s)

Where you are: Great, you have a web analyst(s) who understands your business and is producing reports with actionable recommendations. You're ahead of 90% of other web businesses. Whether anyone in your company listens to them is another question, but at least you're there and your company is making the shift from HiPPO (highest paid persons opinion) decision making, to data driven decision making when it comes to the web.

You will start to see wins now. More customers will come; marketing campaigns will be more effective. If your analyst is good, you'll even start doing A/B and multi-variant testing to tweak your site.

A web analyst can tell you WHAT is happening on your website. They can also make a lot of inferences about why things happen the way they do. The answer to the question of WHY though requires a simple technique.

Where you need to go: Ask the customer.

Better yet, give them the opportunity to tell you without being annoying about it.
Welcome to the world of Voice of Customer. Feedback forms, surveys, and usability testing all fall into this category. While I'm not here to plug any one technology, I will point out one that I am familiar with that seems to stand above the crowd. OpinionLab offers electronic comment cards on every page of your website. The trick is though that the comment card is linked to, and can be specific for, that particular page.

Comment cards let users rate a particular page, section or site as well as provide feedback. Customers can actually tell you that information is missing, forms don't work, or data is out of date. The reporting is also very sophisticated so you can look at your website as a whole and pick out specific sections or pages that need improvements. If you have an enterprise level web analytics tool like Omniture or WebTrends, you can even display all the Voice of Customer data in the same interface alongside your web analytics data.

Suddenly, you start to notice that the parts of your web site that you thought were just fine are ranking really poorly. Several comments tell you what's wrong with it, and you can go about fixing it.

Bingo, now you know what is happening (web analytics) and why it's happening (voice of customer).

4) You have a Voice of Customer Tool

Where you are: Sweet, you're now ahead of 95% of enterprise web sites. You have a web analytics department and you're letting customers tell you what needs fixing on your site.
Yet we're only halfway up our scale. Is 50% good enough for you? If so, then pop the champagne! Wondering if there is more you can do? You bet there is.

Stay tuned for Part 2

Labels: , , , ,

Bookmark and Share
posted by Dan Linton at 0 Comments