If a customer gives up just before completing an online transaction, do you know about it? Do you know why? What if many customers are reaching the same roadblock? Software from Tealeaf Technology, a spin-off from SAP, can give you granular insight into what your online users are doing. CustomerThink founder Bob Thompson talks to Tealeaf Vice President Geoff Galat about how it all works.
The transcript of this interview, which was recorded March 12, 2007, was edited for clarity.
Bob Thompson
I’d like to welcome today to our Inside Scoop program, Geoff Galat, who is V.P. of Marketing and Product Strategy at a very interesting company called Tealeaf Technology. Tealeaf is based in in San Francisco. Geoff, welcome to our program.
Geoff Galat
Thanks, Bob.
Bob Thompson
We’re going to be talking about how you analyze and ultimately improve the online experience for customers. Of course, customer experience management is a very hot topic these days. But one of the areas that is most difficult to get your arms around is the web experience. We’d like to talk about what Tealeaf does and how it addresses some of these experience issues and what businesses get out of that.
To get us going, though, could you please first just talk about your background, in terms of how you came to Tealeaf and what you are doing there today.
Geoff Galat
Sure. I’ve been at Tealeaf for going on six years now. I’ve been with the company since its fairly early stages. My background prior to Tealeaf was actually split across three different companies that were in all various aspects of the customer experience base, although coming at it with a slightly different kick at the can. I spent a number of years—almost seven—at Mercury Interactive. I was very instrumental in bringing to market both the testing tools at Mercury as well as the performance management tools later on. And I was instrumental in bringing those technologies both to the web space, in particular. I think that that background, alone, helped me collar the view and the perspective that I’ve had about what was necessary to ensure a positive customer experience right from the earliest time that I joined Tealeaf.
Today at Tealeaf, I’m responsible for marketing, so I’m the outbound voice and evangelist, as well as driving demand for the company, as well as being very involved with strategy. Being in the space for quite a while now, I have a fairly broad experience level, in terms of what customers are asking for, what analysts are looking for, etc. I also spend a significant amount of time talking to prospects and customers and function really as the evangelist for not only our products but for the space, itself.
Bob Thompson
When was Tealeaf founded?
SAP spinout
Geoff Galat
Tealeaf has been in business since 1999. We’re actually a spinout from SAP. In fact, we’re the only spinout from SAP.
Bob Thompson
Spinout in what sense?
Geoff Galat
The original value proposition of Tealeaf was created within SAP. The founder of Tealeaf, who is still our CTO and a member of our board, Robert Wenig, head of advanced development at SAP, which was the group that was bringing SAP to the web originally. He built those first applications for SAP’s customers and was immediately struck by the lack of visibility that he had and the inability of his team to adequately support those customers, compared to the way he’d been able to support client-server customers in the past. It was largely because web customers do strange things, and the web is not a structured environment, much like a client-server application would be. So Robert took a portion of his team and asked them to go off and build a solution that enabled him to have some visibility into what the web customer was doing so that they could better support them. That technology that was created—and the group of people—ended up being the underpinnings of what Tealeaf is today.
Bob Thompson
Does the name have some connotation like looking in the tea leaves to see what’s going to happen in the future?
Geoff Galat
That’s exactly where the name comes from, reading the “Tealeaves.” In fact, the earliest tagline of the company was, “We see customers in your future.”
Bob Thompson
Let’s talk about the solutions that Tealeaf provides. I want to quickly get into how these are being deployed to help analyze and improve the online customer experience. So, what is it that you’re bringing to market?
Geoff Galat
Tealeaf has a family of products that is called the Tealeaf CX Suite. CX is an acronym, if you will, for customer experience. Our entire business is based upon improving the customer experience for companies. And what CX represents is, as I mentioned, a family of products. It’s a series of different applications that provide visibility into what the customer sees and does for various parts of an enterprise. They’re all based upon one consistent component, which is called the Tealeaf CX data store. The Tealeaf CX data store enables you to capture every customer, every one of their interactions and every interaction that your site does give back to the customers for every transaction every time.
Bob Thompson
Well, that sounds like just a massive amount of data if you’re talking about a bank and they’re capturing every transaction or have millions of transactions per year.
Fifty million sessions
Geoff Galat
In fact, Wells Fargo Bank, one of our larger customers, the last statistics on it I saw were about 50 million user sessions every month through just their retail banking site.
Bob Thompson
Take us through an example of how Wells Fargo or other companies like that might use your solution to figure out what’s going on with their online customers.
Geoff Galat
There’s a couple of different things that are done. The core solution—or the flagship product, if you will—that sits on top of Tealeaf CX’s data store is an application called Tealeaf cxImpact. And what cxImpact is for is identifying and rapidly resolving customer-facing web issues. So what Wells Fargo or any other company would be able to do is actually multithread it. There’s really four ways that a company can become aware of the fact that a customer is experiencing some sort of a problem with a transaction going through.
And those four ways are:
- They can observe some change in the trend in their business, so they can see that their conversion rates or their adoption rates or some measurement step to step has degraded from what it’s expected to be.
- There is the way of finding out from some other system, so if you know that you had a catastrophic failure with half of your web servers and you know how many customers are typically on your site at any given time, you can make an assumption that you have significant impact. Not always true, but you can at least base it on, “There’s a customer impact” or something like that.
- The third way of awareness, which is probably the one that most companies rely on, is to wait for the customer to call you and tell you they’re having a problem, which is the most reactive and the most difficult position for companies to be in.
Bob Thompson
And customers don’t tend to call. They just leave.
Geoff Galat
They don’t. They don’t. But when they do call—when a customer calls when they experience a problem—it is not to help you diagnose the problem. Their objective is just to solve the business problem they’re trying to solve. If they’re trying to move money from one account to another account, when they call you and they get the 800 number and call up the bank, they’re not trying to help you diagnose what went wrong with the web site. They’re just trying to move money from one account to another account. So they’re not a great mechanism for helping you even resolve an issue if they do call.
And the fourth way is, through our product, you can actually set up interesting ways of getting awareness, either by proactively monitoring critical business processes—so if you see a certain customer, a certain class of customers, a certain value transaction that fails, you can be alerted to that failure and immediately jump in and say, “Well, wow! if he had an issue, let’s go look at that”—or you may not do that at a granular level if your bank has millions of customers.
But you might look and say, “We have 100 people in the last hour that tried to do something and weren’t able to do it. We should go look into why that is occurring.” So that’s one mechanism. The other mechanism is to just be able to more tightly monitor those trends that I mentioned in the beginning of the discussion. There’s a series of trends that might be really early indicators if you see that people moving from Step 6 to Step 7 in your business process are failing at 100 or at a percentage rate of your business, wouldn’t you want to know why that is happening so you could drill into that and try and understand it?
Bob Thompson
Well, these all sound great in theory, but can you give us an example of how a real customer is using your technology and what they get out of it?
Geoff Galat
Sure. Let me give you one example: a customer of ours, another San Francisco-based company, actually Esurance. They’re an online insurance company. Most people have probably seen their advertising. They’ve been all over the place recently. So Esurance has some very critical steps in their business process that they monitor very tightly. They spend a lot of money driving people to their site. They compete with all the other online and offline insurance companies. So some of the things that they monitor very closely are policy starts, what they call “quotes-to-buy”—the quote-to-buy process and policy completions. They track very closely what their conversion rate is on all the steps of that policy process. And one day, they noticed a degradation of a trend in the quoting process. They were seeing just a slight degradation. We’re not talking about a massive percentage here. We’re talking 1 to 2 percent degradation from what their baseline was. They said, “We should look into that because we don’t want to have any degradation.” They started drilling into that and immediately found out through our product—because we have everything that every customer saw and did—a customer example of a user who had made it to the step in the business process where you put in your vehicle ID number (which, for those of you not familiar with it, is a 17-digit character set, both letters and numbers, and it’s unique to every vehicle. And you have to have it to get a policy.)
‘Six people every day over the course of the year is 2,400 policies.’
They saw that somebody had failed in putting it in, and they looked at it and the person had put in 17 digits. And it looked like it should have gone through, but it hadn’t. And the result of it, when the customer hit the submit button, they were kicked into an endless loop. It literally fired them back to the very beginning of the policy quoting process, which is a seven- or eight-step process. That customer abandoned. They immediately said, “Well, wait a minute. This shouldn’t have happened. Why did this happen?” They started looking at it fairly closely and realized they’d created a fairly free-form field for the logic in the application. But that free-form field enabled you to put spaces in. There are not spaces in vehicle ID numbers. When you go look at the plate on your car inside your windshield where your ID number is, there are no spaces in it. However, certain states, when they print their vehicle registration number sheets, print it with spaces in it.
And this customer obviously did what I would do: ran out to their car when they needed to fill in that part of the application, grabbed their registration and typed it in exactly as it appeared on the registration. And it failed because their logic just couldn’t handle the space. It was a 15-second fix once they understood that we needed to be able to handle spaces here because this could happen.
But the next step for them was, “Well, wait a minute. If this happened to one person, who else has tried to do this exact same thing?” Because there’s a lot of states that might be doing the same printing process. They went back and pretty quickly found out that there was about six people every day, which represented this small degradation in their conversion rate or part of it. Six people every day. They went back and looked through the Tealeaf CX datastore and found out that the issue had existed since their new version of the application had gone live a year earlier. So six people every day. That’s not a huge number. They’re selling thousands of policies every day. But six people every day over the course of the year is 2,400 policies. Their average policy is somewhere around $1,200. So, you know, $2.3 [million], $2.4 million worth of opportunity costs.
Bob Thompson
Right.
Geoff Galat
And these are people very deep in the business process. They’re people who’ve made it all the way through to the point where you’re going to put in that vehicle ID number, which is signifying to them that you’re pretty serious about buying this policy. You’re not just dropping off at the first step when they give you a quote.
A thousand cuts
Bob Thompson
Right. So it’s like being nicked to death by a thousand cuts.
Geoff Galat
Yeah, we use the term all the time: You’re dying a death by a thousand cuts.
Bob Thompson
Do you find that your customers are using your solution more to diagnose problems in the transaction in the actual marketing-and-selling process?
Geoff Galat
I think it’s an evolutionary process. I think with most companies, the earliest starts that they get, the earliest big-bang return they’ll get, is when someone calls in. If I were to call in to a company that had Tealeaf and say I was having an issue with something, you could very quickly find my sessions and be able to resolve the issue that was facing me and then do the quantification to see if the thing that was affecting me affected anybody else. And how many were those and how big a problem is this, really?
Bob Thompson
What kind of industries do you find are interested in these types of problems?
Geoff Galat
It’s anybody who’s doing mission-critical web-based transactions. But what I would say is that the earliest adopters of our solution in the customer experience and general management solution, in general, have been financial services—being banking, brokerage and insurance.
Bob Thompson
And why do you think that is?
Geoff Galat
Well, I think that the value of the transaction is higher. They’re interested in the long-term value of the customer and the customer relationship. And the applications are fairly complex. They’re not typical applications. There’s a lot of moving parts. You could have 10 people all having accounts at the same bank, all living at the same city, but everything about them is different, which means that everything that they get in their individual experience is unique every time they log in.
Bob Thompson
Yeah.
Geoff Galat
It’s a challenging thing to manage for the banks.
Bob Thompson
What other industries are interested in this?
Geoff Galat
Retail and travel. Retail is a very obvious one. Travel is very similar to the banking example with very complex transactions. Every user, every time they’re doing something, is putting in something different. They’re all looking to go on a different flight at a different time to a different location, different airlines, different frequent-flyer numbers, etc. And then, we’ve seen a significant uptake recently in less “obvious” categories. I would say some of the things like transportation and logistics with trucking companies and railroads that have bought the product, companies that are doing manufacturing. We’re seeing an uptake in high technology, as well, with companies who are doing things like monitoring and managing the renewal process on their key software. They’re monitoring trends to say they have an expected renewal rate on their software and that they’re not seeing renewals come in at that rate. Are there reasons why they’re not seeing those renewals happening? And how much of that degradation is happening because of the customers’ inability to renew, based upon some issue with the site?
Bob Thompson
And how many customers does Tealeaf have so far?
Geoff Galat
We are just north of 225 customers, and that accounts for more than 300 deployments of the software and number of customers. Like a Wells Fargo, for example, has employed numerous applications.
Bob Thompson
And can you give us a sense for how your solution is priced and deployed?
Geoff Galat
Yes. I call it a very fungible pricing model. The pricing model is really based more on the size of the environment and the size of the site or application that we’re monitoring and then the number of users who are going to access our applications. As I mentioned, we have really five different applications. So there’s different layers there. But at a baseline, what most companies do, the key component of the price is the cost of the CX datastore, which is the baseline application that captures every user, every interaction every time. And the CX datastore pricing is based solely on the size of the application that we’re measuring. So, if you have a two-web server, four-CPU, relatively small retail site, we can be down in the $50,000-to-$75,000 range completely deployed and installed for you.
Bob Thompson
Now when you say installed, is it literally installed or is it implemented as a hosted service?
Geoff Galat
No, it’s installed. It’s on-premise software deployed at the site of the customer who purchases it.
Bob Thompson
Including all of the data that’s being captured?
Geoff Galat
Correct. Absolutely. Everything stays behind the customers’ individual firewalls as part of their environment.
Bob Thompson
So you can start at around $50,000 and then go up from there, depending on the size?
Geoff Galat
Yes and, obviously, if you’ve got a 400-web-server farm, massive travel portal or a very large banking site, it can be more than that. So our range is really 50 [thousand] to 75K on the bottom end and up well into seven figures on the higher end.
Bob Thompson
All right. Let’s take a look down the road. What do you see in the future for Tealeaf? For example, are you going to get acquired by SAP?
Geoff Galat
Re-acquired?
Bob Thompson
Re-acquired. Yes.
Delivering value
Geoff Galat
Our goal, and as a member of the management team—and I have been the whole time that I’ve been here—those are really not discussions I can honestly say … We don’t have those discussions about what is the end game. We’re really focused on building a solid company. We’re profitable, for example. We’re acquiring customers, as I mentioned. We’re growing very rapidly. Our goal is to build a solid company that has all of the measurements or the metrics in place that would tell you we’re a solid, successful company.
And one of those metrics—and the one that we’re the most focused on—is delivering value to customers. We believe that the longer-term thing—whether it were to be to go public, to be acquired—those things will come into play or fall into place as a result of that. As long as we have all the right things in place to be a solid company, then those other things are things that just will take care of themselves.
Bob Thompson
Well what about the expansion into other facets of customer experience? You basically look primarily at the web experience and you’re providing analysis tools for that. But customer experience management is a very broad field already, even though it’s relatively new. Could you see the products expanding in some different areas over the next couple years?
Geoff Galat
Yes. I think we definitely have some ideas of some places where we could go with it. I think that there’s a couple of very normally adjacent markets that we already have some early partnering activity in, which are both with CRM solutions—the ability to enable a CRM solution to be able to give a customer service agent visibility into the customer sessions in very real time—as well as business intelligence solutions—the ability to take the rich data set that we have and combine it with other data sets, as you described, so that you can have a broader sense of, “What is the cross-channel or multi-channel aspect?” How many people research a mortgage online but actually go into a branch to buy the mortgage. And wouldn’t that be interesting to understand? And as we look at expanding what we do, either through partnership or potentially through some other products of our own, you can see there’s some very adjacent areas that make a lot of sense.
Bob Thompson
Last question for you, Geoff, is: What do you think are going to be some of the bigger challenges that your company will face to realize the vision that you’ve laid out?
Geoff Galat
I think the biggest challenge is the challenge that it’s not much different than the challenge that we’ve already had to experience, which is the problem that we solve for people—as you described it—is the death by a thousand cuts. And sometimes, if you don’t know that you’re bleeding, you don’t know that you’re dying. And there’s a tremendous amount of inaction that we see out there from companies who don’t necessarily recognize the impact of these issues on their business. So, if you’re a business and you’re growing from $350 million online last year to $500 million online next year, you might look and say, “We don’t have any real issues. Our growth is astronomical. We’re growing ridiculously fast. The boss doesn’t have a gun to my head, telling me things are broken.” We have a lot of uphill fighting to do at times to explain, to really get down to the point of: Could they be growing faster?
Bob Thompson
Of overcoming the apathy, perhaps.
Geoff Galat
Exactly. And part of it is you don’t know what you don’t know. There has never been a way to measure or understand the impact of customer-facing issues on a business before. A lot of these companies, they believe that they’re doing the right thing. They don’t know what they don’t know.
Bob Thompson
Geoff, thank you very much for spending time with me on Inside Scoop. I appreciate your insights.
Geoff Galat
Thank you, Bob.
