Blogchat with John Marshall, CEO Clicktracks

I first met John Marshall at the Emetrics Summit in Santa Barbara in 2003. I was immediately struck by his vision for ClickTracks and web analytics in general. I have enjoyed meeting up with John at various events, as he always has something interesting to say in an interesting way.

John was on holiday here in the UK last month and and I managed to persuade him to meet up and share his thoughts on various topics to do with ClickTracks, Google Analytics and the state of the web analytics industry amongst other things.

Here’s what he had to say.

One of the things that I thought would be interesting to talk about is the merger of ClickTracks and J L Halsey. I was wondering what the main drivers for the merger were from your perspective?

I’m primarily motivated by delivering really great products to customers. ClickTracks was not VC backed, and that gave us considerable freedom to consider different opportunities. We can do what’s right for the customer, knowing that everything else follows.

The opportunity from Halsey was the right one for us, for several reasons. The first is that we retain our independence and that’s important for me. The second is the changing landscape for online marketing. Our product has been known for customer acquisition analysis and that is primarily search marketing. By contrast, customer retention marketing is the poor cousin. Search marketing has been so easy and effective that there hasn’t been any need to spend time on customer retention marketing.

Companies focus on customer acquisition and don’t really worry about customer retention and everything still works. However, there are some changes coming about. The rising cost of PPC is forcing companies to think about customer retention. You’re seeing the start of this trend.

The other company’s involved in J L Halsey are mainly about customer retention, email and RSS feeds, and finding ways of getting the customer to come back to the site. ClickTracks is not so well established in that particular field, and that gives an opportunity to build some really great products and focus around professional retention analysis, and that’s it. That’s what we want to do!

So, it’s email marketing integration that you are looking to do next?

Yes, and it’s difficult to do well. We’ve done it in the past, with partnerships, but I thought the solution just didn’t work easily enough for the customer. Now we can bring our customer acquisition analysis expertise to the problem of retention analysis. So we get to play on both sides of that equation. I think successful companies have been doing one or the other. There’s not that many companies that have been doing both, but I think they’re going to have to, because customer acquisition costs are just going to continue to go up.

What’s your perception of the way that Google Analytics has impacted the market generally and ClickTracks specifically over the past year?

We said, when Google Analytics first shipped free, that the interest in our product would go up. Some people thought that was gibberish, but it has proved to be true. In my opinion, Google Analytics is one of many products that I would classify as stats programs. I don’t actually view Google Analytics as an analytics program, because I don’t think you can easily or flexibly do what we define as “analysis�.

So, how do you differentiate between reporting and analysis? What actually is web analytics?

Reporting is “Today I know of certain things that I’m going to want to calculate” and I build a stack of tools which usually are wired directly to the website, so that the website is an expression of the data that you want to know. It could be something fairly simple like how many page views does this page get from pay per click visitors. So you need to know in advance that you’re going to want to get that number and you build your site in such a way that that number falls out of the process. Analysis gives you the ability, after you’ve done the data collection, to go back through historical data and say “OK, what was that all about?”.

“Reporting” requires that you know what you’re going to want to know, whereas “analysis” gives you the ability to learn “on the fly” what you don’t know, and then to know it. I think that’s fundamental and I think that Google Analytics shoots to the top of stats programs. ClickTracks was never designed to be a stats program, but it’s a really good analysis program - at least in my humble opinion! But that’s what it was built for; it was built for analysis.

The other thing that’s been interesting is the data privacy issue. We thought that customers would not be aware of the fact that in return for Google doing all of this stuff for free with your data, that Google is going to get your visitor behaviour data. We thought that customers initially wouldn’t be aware of that, but they are. There is some legitimate concern over the issue. So we did particularly well out of that. With our product, you can get the software, or you can get it as an ASP, and because we offer the software, we’ve got a really good answer to the data privacy issue. Not only do we promise that we’re not going to expose your data, we’re actually make it impossible for us to expose your data. You own the software, you own the servers, you collect the data, it’s yours, we don’t ever see it.

So overall, Google Analytics has substantially increased awareness of the need to get good data somehow, and then it’s a question of which route you choose.

One of the trends over the last couple of years with other web analytics systems has been the ability to either to get direct access into their data warehouse or the ability to export pretty granular levels of data into other systems. Is that a direction that are you are likely to go down?

One problem with that is that the customer usually wants your data to be exposed to SQL, probably with a star schema, and SQL is simply the wrong tool for the job. However, it’s the only thing that people know how to use. The trouble is that a web analytics system is dealing with so much data that if we export the raw data to people, it usually breaks. People don’t understand that, they don’t want to hear it and it’s very difficult for us to say to a customer or a prospect that they really just don’t understand the scope of the problem. We are planning to open up the database, but I but the first implementation won’t use SQL to do it because it’s just so inefficient.

So, we will do something about data integration, because people need it and want it, but I don’t think it will work as elegantly as people think it will.

How do you get over the challenge then of being able to create that single view of customer?

I’ll give you an example of a solution provided by ClickTracks that wouldn’t be appropriate and then a solution provided by ClickTracks that would be appropriate, just to give you two extremes. If the customer says, “I want a dump of all the raw data, just give me the raw data”, this is not going to work.

On the other hand, if the customer wants aggregated statistical data by segments where you might have a very large number of segments, then that technically would work. But in order for that to work, the customer has got to be comfortable with statistical sampling techniques and I just don’t see that people are comfortable with those sorts of statistical correlations. They want it to be “accurate�.

I’ve heard you say it many times, “good is good enough”. Do you think we’re obsessed in this industry on having totally accurate data?

Yes, I do. I want people to move beyond the accuracy thing. They are beginning to, through constant efforts by me and a few other lone voices in the wilderness!

One of the things I’m interested in is your perspective on where we are in the UK and Europe when it come to web analytics compared to the position in the US. What do you think are some of the main differences?

That is a very good question, and I think there are some fundamental differences that will never go away. One of the key differences is a cultural difference. There’s an old joke that American companies are run by salesmen, British companies are run by accountants, and German companies are run by engineers. I think there’s a lot of truth in that, so I have noticed that the British tend actually to be very good at the analysis process. People like yourself, who are very focused on the importance of real analysis.

Americans tends to shoot from the hip. They are more like the sales guy: “Just give me the numbers, give me a quick go through it, I can work it out, and everything will be fine”. In Germany, businesses that are hung up on the accuracy of the numbers will never make a decision. There’s just endless thrashing around in committee meetings. It’s great for building cars, but I’m not so sure that’s so good for such a fast paced environment as ours.

I think in some ways therefore that the UK is ahead of the US because of a greater willingness to grind through the numbers. On the other hand, Britain has always had rather more of a focus on branding. I’ve noticed that branding in marketing here is very important and the British are extremely brand conscious, more so than Americans. Americans have got more focus on direct marketing. As a result I think the direct marketing industry is stronger in the US, which helps web analytics.

I think that it’s superficial to say that Britain is a bit behind the US, I actually think that there’s more of just a fundamental difference than there is a time lag.

Where do you see the industry going? What will we be talking about in a couple of year’s time?

To answer your question in a round about way, when I think about web analytics, I ask myself sometimes “Am I selling a car or am selling a mode of transportation?” Is that the product I’m selling? For example, it might be that one web analytics company’s product is a car, and another company’s product is a motorbike, and another company’s product is a bus. What is a customer actually buying? Transportation is about getting from A to B.

Conversely, sometimes I think that we’re, figuratively, selling a car. In this case, the customer has already decided that they don’t want a motorbike, so they’ve decided to buy a car. And even then, I’m actually not sure that in web analytics we’re selling that car. In this market, perhaps we are just selling the engine. It’s an important piece, but the customer understands the need for an engine. The customer wants the car, and the successful product will be those that can move beyond being an engine to being a car.

Click Tracks has done OK at providing a solution and not just a box of parts–though I still have a long wish list that our product doesn’t yet do. We’re obviously going to do a lot better, because we have vastly more resources now.

I’ve started thinking that web analytics is a bit like CRM in 1990s. Everybody talks about it, but nobody actually knows what it is. Do you think that’s a good analogy?

I think there is a lot of truth in that. It might be a little unfair because web analytics is a lot less complicated than CRM. CRM had a bad rap because of the complexity of the implementation and the need for the company to change its business practices in order to fit in with the CRM system.

You say the CRM implementations were quite complex, but web analytics implementations can be pretty complex as well, and actually can involve change management processes as well…

Not if you buy Click Tracks!

So, other than buying Click Tracks, what are the secrets of successful of implementation?

I hope that shameless promotion gets expunged!

In all seriousness, we built the ClickTracks product deliberately so that it “does less�. You see, doing the more sophisticated and complex stuff is so complex that I started to become unconvinced of the benefit. I’m unconvinced that once you go through all the implementation complexities, which usually involve programming within the website, that people get a lot of value out of it. We want to make it really, really easy to do the implementation, so we just skip some of the more exotic functionality. And in the real world, a lot of people don’t do it anyway because it’s just too complicated or it just doesn’t work!

If all web analytics products had followed the ClickTracks approach, do you still think that there would be a need in the market for the WebTrends, Omniture, Coremetrics kinds of products with their feature rich functionality?

Yes I think so. An example of what Click Tracks is not built for is merchandising analysis. For example in the scenario when a consumer is looking at one product and you dynamically work out three other products to show along side that are related, we don’t do that type of analysis and we don’t want to. We don’t want to because it’s just so difficult to implement and we would rather not get involved. So there’s a need for products out there that can do that, although there’s only a small number of companies who want the data and are willing to go through the steps to get it.

We try to provide a product that’s like an iPod. An iPod does less than any other MP3 player, it has less features. But that doesn’t mean that once you have an iPod, that’s the only way you’re going to consume music. It’s just not that simple and a trend that we’ve seen is companies using more than one web analytics tool. That’s increasing. We regularly sell to people that are also using Google Analytics, for example.

Switching tack, what is it that you most like about what you do?

That’s a good question. I think it’s the fact that anybody that uses an analytics tool will improve their business. It’s as simple as that. The data is just simply lying on the floor, you just need to spend a few minutes sweeping away the junk to get at some real data and you can improve your business. In fact, you can double it. We started out with that mandate to make it really easy for companies to improve their business.

A lot of businesses that we help tend to be start-ups, bootstrap companies and small companies. We sell to large enterprises as well, but it is a little bit less rewarding. I am really proud of the fact that there are small companies out there that make money and that we played a part in improving what they do. That’s pretty cool actually and it’s fun.

I’ve always been in the software business, so I have a certain sense of pride in craftsmanship in building a product, which you can’t really express in a business sense. But, you know, it’s great fun!

Thanks John.

More from Applied Insights

See more: Applied Insights Blog
See more: Blogchat

Learn about our web analytics consulting Discover our innovative marketing applications Get to know Applied Insights