Transcript of Podcast with Brett Crosby and Avinash Kaushik
Podcast Date: June 13, 2007
The following is a written transcript of the June 13, 2007 discussion between Eric Enge, Brett Crosby, and Avinash Kaushik:
Eric Enge: Hi, I am Eric Enge, the president of Stone Temple Consulting. You can see our website at www.stonetemple.com, I am here today with Brett Crosby and Avinash Kaushik from the Google Analytics team. You can see the Google Analytics website at www.google.com/analytics, and you can see Avinash's blog at www.kaushik.net/avinash. So how are you guys doing today?
Avinash Kaushik: We are going great, thank you.
Brett Crosby: Great. Thanks, Eric.
Eric Enge: Okay, super. So one of the big things that has happened recently is you did this release of Google Analytics on May 8th that had a really dramatic change in the interface and many new features. How is that been received by the industry?
Brett Crosby: It has been received very well. We have been very happy with the response so far and we are really proud of what the team has put together. They have done a phenomenal job. I don't know if we can give our engineers on the front and backend enough credit or how well it turned out. But we are pretty happy with the results, and so far the public's reactions have been very, very positive for us. One other thing, I just wanted to mention was, I talked about it briefly at Emetrics, when we launched, was a little bit about the impetus for why we get the redesign.
There were several reasons, but the primary thing that we saw was a bit of a paradigm shift in the industry, and that is that the previous model was around how people who were used to web analytics were interacting with their web analytics tool. And the previous interface was pretty much designed around that model, just a relatively small web analytics community. But what happened over the past couple of years, is that we really increased the size of the industry and the amount of people outside of the industry who wanted access to you web analytics data increased dramatically. There is this whole new kind of audience has cropped up for us. So we did some pretty dramatic things. We have really redesigned the entire interface so that it is more contextual, and data is very, relevant within other pieces of data in the reports. And then, the second thing is that you can now email these reports out so that expert web analytics people do not have to sit there and explain what everything means and then email it out and kind of present it to the rest of the company. They get to focus on doing their deeper analysis and we added a lot of tools for them to do that in this interface. And they can have automated emails sent out with really relevant and contextually informative data right in the interface so that people can see those reports and take action right away, so that whole kind of combination we think is a pretty big step forward.
Eric Enge: Right. I think there is an increasing appreciation of how web analytics can help a web-based business, or web-based portion of the business make money, or make more money than they are making without it has really grown. And I think that is absolutely the case, more and more people understand what web analytics can do for them and therefore more and more people want to get access to, it including senior managers at companies that don't want to take the trouble to log in, they just want the data presented it to them in the most convenient way possible for them. I think that was a good step forward, in terms of adding that email capability that came, you've put in the product.
So you released an additional new update yesterday. I think this actually included a few tweaks based on user feedback. And one other thing that was interesting is that there was a lot of demand for seeing the data on an hourly basis as the old interface supported and initially you had removed from the May 8th version. Can you talk about that a little bit?
Brett Crosby: Sure, happy to. This latest update that we launched yesterday, is something that we are happy to get out there. Once we launched the new interface, we heard a lot of people asking for a few specific features. We heard pretty loud and clear that these were things that people wanted. As a result, we added, hourly reporting back in the product as you mentioned. Our friends at Yelp and guys like Andy Beal really wanted us to get those things back in there, just to name a couple. But, many others wanted those things. Clickable URLs was added into the product, and this is actually done through an icon. You can click into URLs that link to your site, or your own website URLs. And, sometimes if they are dynamic, you can click them and they won't work, that is fine, that is to be expected. But at least, it is better than copying and pasting it out of the product and finding out it as that you are getting…
Eric Enge: Right. Getting the same result.
Brett Crosby: Exactly. Danny Sullivan has been harping on that forever for not having that in there. We have been wanting to put it in there for a long time. In the new interface it was actually kind of tricky to figure out, because you can click into the URLs and keep drilling down to find out more information about the pages, so we could not actually make it a clickable URL, so we added an icon instead that then links to the site that brought you the traffic.
Eric Enge: Right. So one of the things that I like about hourly reporting is that if you are marketing to, or just being found and put on, social media sites, it gives you different level of visibility as to when you achieve success. You can see when you achieve front page status on Digg, and when you left the front page. It provides you a good visualization of the time frames for that particular campaign, that I think you can't get quite as well if you just reverse engineer it from the daily numbers.
Brett Crosby: Right, exactly. That is definitely one of the key things, to see when you have been posted on Digg or something like that. Or the other thing is that a lot of people like hourly reporting so that they can do hourly bidding within AdWords, and we understand that as well.
Eric Enge: Right. So I guess the perception there is, for certain hours of day there are hours that offer the best conversion, is that the idea?
Brett Crosby: Right. Exactly, and what some people like to do, or what some people find is that there are certain key words that can mean multiple things, and their audience is buying for those things at certain times of the day, maybe around the lunch hour or maybe before work or after work, or and it could vary by time zone or geography as well. But, you find people that really get into certain keywords and spend a lot with them to manage these little eccentricities that really work for them as well.
Avinash Kaushik: The other thing is that in some of the blogs and websites there was feedback that people wanted to do more than look at days in silos, but lots of people like to compare the patterns of shopping on their websites this week versus last week, or this month versus last month or last year. Then at the aggregate level they can understand the behavior of their customers down to an hourly level and then maybe even sub-segment it by geography or city so that they can get more complicated about how they should be running campaigns, whether with AdWords or otherwise. In those scenarios where you do have an ability to react fast enough, hourly reporting can certainly be something that is quite helpful.
Eric Enge: Right. That makes sense. Another big area was the addition of cross segmentation by network location?
Avinash Kaushik: Oh definitely. The reality is with all the complexity that is going on with DHCP and IP addresses, that people still used this feature a lot. It sort of goes to show that the response of the team in reacting to what our customer base would like to use and see as an optimal feature set for them. Network location is great, and we have learnt that a lot of people will use it to simply size their core key data in Google Analytics or create segments of data in Google Analytics that allows them to understand the IP addresses of where people are coming to their websites from, which seems to help them quite a bit in terms of creating the experience they want on their websites and understanding the usage on their websites. It is something that a lot of people seem to be using and something that was very quickly added to the latest release. Only a month after the recent major release went out.
Eric Enge: Right. So we are going to see a new release in another month?
Avinash Kaushik: No comment!
Eric Enge: What a surprise! Yeah. I think that is great that you have come up so quickly with an update release because that is what happens whenever you put a product out there, you are always going to get some very fast feedback of a few things and it is good to address it.
Avinash Kaushik: Exactly and I think Google has focused on ensuring that it is very responsive to customer feedback as well as providing the data and feature-set that are relevant to its customers. For example, you'll also notice that the number of rows we can export at an instance is now increased to five hundred rows of data on a single report. And the beauty of this is that you can actually export segmented narrow data that you actually need rather than getting a span of aggregated data. Give me all ten thousand rows or all five hundred rows of search key words on my website. What you can do is, you can say no, I want all search keywords from people who came from Google during these hours in a day, you can actually segment, you can go down to the data you actually need and export that out so they can create another presentation layer for your management, for example if you want, or merging it with other offline sources.
I think it is great, that the number of rows you can take out on a single report has increased. And also, interesting metrics are now highlighted very prominently, such as Bounce Rate, one of my favorite metrics. I am in the process of writing an article for marketing professors and the working title is Sexiest Metric Ever- Bounce Rate. There is so much value in that metric in understanding the quality of the acquisition strategy that you have for your company, and the traffic that is coming to your website, again that has also been enhanced with the latest release. So there are a good bunch of things that have been released with the latest upgrade for Google Analytics and I am not sure if you noticed, but it is already out of beta, it is probably the shortest beta ever in the history of Google.
Eric Enge: Google does have products that are still in beta that were announced years ago.
Avinash Kaushik: Yes.
Brett Crosby: Absolutely.
Eric Enge: Yeah. So, another thing I really like about the new release overall is the path analysis functionality. I think there are some really good things, I know you mentioned that to me, Avinash when I was out at GooglePlex during my last visit. But I would really like to talk about some of those things, such as the navigation summary, for example and entrance pages, etc. Can you talk about these?
Avinash Kaushik: Actually, the core essence of the issue, Eric, is that for the longest time people who are consuming web analytics' data, or trying to analyze the data, have judged the effectiveness of a page simply by the number of people who might be exiting from it. And it is usually a very primitive barometer of the page, and if it is performing the way it should be. I find a lot of people who are getting deep into path analysis where they are try and recreate and restructure every single path, that every single person has taken on a website. And then try to understand what the most common path is on the website. The challenge with that, is it the most common path on the website might be actually only taken by one percent of the site traffic. Say one percent of site traffic is following the same path, what would you do? Would you take any action on it, and so what I am noticing is there is this gradual paradigm shift in analyzing the value of pages. People are actually looking at much more sophisticated amounts of data all in one place. For example, with Google Analytics you could go down to content, you can pick any page you want, and very quickly you will see the default detailed data, which gives you a very quick understanding of how many times a page was seen, how many people bounced off it, how many people are, and how long did they spend on the page.
Then with a very quick drill down on the analyze button, you can truly begin to understand if the page is performing correctly by looking at entrance paths, which probably is a very unique feature to Google Analytics. What it tells you is very very interesting. It takes a lot of the mathematics and complex computations behind the scenes, and you can see everywhere that the people who saw this page came from, but then the beautiful thing is you get a box where you can say, what was the next step that people took from this page. By clicking on each of those next steps you could see how much that particular next step is contributing to where the people ended up. So let us say you are trying to drive a lot of orders on your website, you could say from the product page, what is the next best step that people can take, is it the next best step that they go to a comparison chart, do they go to a product overview page, or do they go to one of the user generated content. Also, what is the next step that has the most amount of influence in people ending up placing an order. That is easy for you to understand, all of the math and computation is done for you already, and you can take quick action.
If turns out that one particular next step is causing people to buy more, well you could figure out how you could merchandize better that particular step, or how could you learn what is not happening in the other paths, and this is a very unique report to Google Analytics. The nice thing is that it actually helps you drive action. The same thing goes to the other two things you mentioned, Eric, which were entrance sources and keywords. I am always encouraging people to try and understand customer intent because if you understand customer intent, you will be much better able to analyze if the content on the page is actually delivering value to them. So by analyzing data, by entrance sources and entrance keywords, you get a peak into the customer's mind, what were people thinking of when they enter the page, and what were the top keywords that built traffic to the page.
If you find that there are specific kinds of keywords that drive traffic to a page, you can match up to say, hey is the page doing its job, and do I have content related to that those keywords on that page? Same thing with entrance sources. What are the websites that are driving traffic to this page, and am I communicating on this particular page? You can see that within three mouse clicks now, you have a wealth of information that you can use to understand customer intent, why do people come to your website, how they are coming, but also then optimize their content in the page and the navigation structure and merchandizing to ensure that people are able to reach the goals that you want them to accomplish on the websites. It is a very powerful set of tools that are now available to all users.
Eric Enge: Right. So in terms of entrance sources, what you are really looking at is, are people who are initially landing on the page that you are examining from some other person's site, is that right?
Avinash Kaushik: That is absolutely correct. In fact with the massive proliferation of search engine optimization strategy, and all of the robots from search engines indexing your websites, one of the myths out there is that, the home page is still most important page in the world. Well, it is a important page, but it is not the most important page because roughly eighty percent of the world now starts at the search engine when they are looking for something and they are going directly to a deep page in your website. So they are not looking at the holy message that you might have on the home page, rather they are going directly to the relevant content to them, and it is important that you treat every page on your website, or at least most pages on your website, as the home page, with the same love and attention you give to your home page.
What the entrance sources helps you understand is, if people are landing directly on this page on my website, where are they coming from and what is the value proposition I can create on the landing page to ensure that customers get what they are looking for.
Eric Enge: Right. So the Entrance Sources report allows you to evaluate pages as initial landing pages. Whereas in the case of the entrance page report you could be looking at a step that is already part way down the process, and it is not a necessarily the initial landing page on the site. It could be the second, third, or a later page someone visits, but you are really trying to see how they behave on the page and compare their behavior on that page to their behavior on other pages.
Avinash Kaushik: Exactly. A simple way to think about it is, you start your analysis by understanding what are the sources that send traffic to my website, that is the entrance sources. Then you want to understand if people came from a search engine, what keywords they are using to find this page. Once you have a good understanding of that, the next step for you would be analyze the entrance path, what happens from this page on and is this page effective, and what next step is effective in driving people to reach a goal on your website. So that is how I would order it, sources, keywords, paths.
Eric Enge: Right. Excellent. So, Avinash, not too long ago, you were an employee of Intuit, back on March 6th, I believe you made the announcement that you were leaving. What was it that motivated you to leave Intuit and accept a role with Google?
Avinash Kaushik: It is the great food. I had a great omelet with some basil, cilantro and pepper, and tomato sauce. It is like the best omelet I had ever had.
Brett Crosby: Avinash, always shares his Google diet.
Avinash Kaushik: Right. Seriously, At Intuit I had the opportunity to work with some tremendous people in a very good company that really does focus on customer driven innovation and is very focused on analytics. And during that time, I had the opportunity to build out a team and a framework, and that is working, that was received well, and after building that team and putting all of the structure in place I was thinking about what should be the next step that I potentially could consider. And Brett and I actually met at a Frost and Sullivan event almost a year and a half ago and we had been having conversations and chatting about Google.
There was just a great opportunity that Google and Brett crafted for me, where I could come and play the kind of role that I really felt very passionate about, and try and take some of my modest accomplishments in Intuit and see how I could actually help evangelize some of those to other companies and to a broader ecosystem. That is definitely a passion of mine, education, teaching, and speaking. I felt like I would have the opportunity to help build something. The Google offer to me was a great value proposition that, along with the great food, that helped me make that decision to take this next step in my career.
Eric Enge: You began that outward focus when you launched the Occam's Razor blog. You could see it in your blog posts, you are constantly talking about interesting ways to use analytics to help your business, and have shown that passion for quite some time through the blog.
Avinash Kaushik: Thank you, Eric. Yeah. I am actually very gratified with the growth of blog, it has just been around for a year now, and it gets lots of traffic and some amazing conversations. In comments, readers of the blog have written just as many words as I have written in the blog, which is fantastic. So it is great to be able to have this sort of a conversation with people from all over the world, and help move the ball forward when it comes to web analytics and the mind set that should be applied and the kinds of approaches that can yield very powerful results for a company of any size.
Eric Enge: What are your responsibilities in your role at Google?
Avinash Kaushik: I have two primary goals at Google. One of my roles is actually to focus inward at Google. Google as you can imagine does tremendous amounts of great marketing of itself for its products and services. It also uses analytics to actually measure the effectiveness of the website that Google has. Almost all Google properties have Google Analytics on it. So not only do we use Google Analytics, but analytics scales to the power and size that a company the size of Google needs in order to meet its analytics needs. I work with these internal teams to help lay out an approach, metrics, reports, and analysis that can be done to better understand behavior on our sites as well as effectively measure ROI on marketing campaigns that Google itself runs. A part of my role is inward focused.
Then a part of the role is actually externally focused, and which is something I tremendously enjoy, which is to help through speaking engagements, writing, and various other things that we are going to be doing, to help improve the knowledge and expertise in the ecosystem in the broad external web analytics communities with big and small companies, and to help them simply use data better. Because Google Analytics is pretty much on every site in the world already, we are not trying to get more people to use Google Analytics. The focus is what are the specific things that we can do to ensure that people get maximum value out of the analytics tools they use and how can we actually provide tips, tricks, best practices, lessons, and case studies so that people can begin to actually extract value from the data that is already been collecting from the website.
A lot of these will be covered using Google Analytics, but essentially a lot of the curriculum that is going to go out is something anybody using any tool could use and benefit from. There is nothing that is so top secret that only Google Analytic users will benefit from it, the goal is to improve the knowledge of the ecosystem so everybody can benefit and use their tools and data much more effectively to improve their website, improve their customer experience on the website, and of course improve conversion rate. Does that make sense?
Eric Enge: Absolutely.
Brett Crosby: In traditional Avinash style is selling himself a little bit short and being very humble. He is also doing a lot of other things, not just those few things, but he is also helping us develop, new features and ideas for new releases and things like that. So, Avinash, we are extremely lucky to have you here.
Avinash Kaushik: Oh, thanks. The thing is Eric , one of the things that we want to have is an extreme amount of customer centricity to a tool and you saw that definitely with the version-2 launch, which happened before I joined Google. I was thrilled to see the amount of thought that had gone into creating the new version of Google Analytics which is focused on understanding who are the customers, what do they need, and how can we make their lives better and easier by improving the interaction layer, by allowing custom dashboards and all these things.
All this happened before I got here, I was so excited to see the next version of Google Analytics. One of the other things that I hope to bring is through my experience and background, having worked in big and small companies and worked with real people and real users who actually use the data day-to-day, is to bring that customer centricity to the team here so we can continue our evolution, so that Google Analytics will remain a tool that is squarely focused of it's customers and meeting the needs of the customers in a very quick and prompt fashion.
Eric Enge: Right. One last quick question for you guys. I saw on your blog yesterday, Avinash, the only announcement I have seen publicly that John Marshall is leaving ClickTracks.
Avinash Kaushik: Yes. John has decided to leave ClickTracks, I have known John for a long time. He is a great guy, and we've collaborated very often. I have used ClickTracks over the years quite a bit. And John has added a lot of value to the overall web analytics ecosystem, and I think that he has exciting plans for what he is going to do next, and I am very excited to have worked with him because he has been very good participant of the ecosystem. I think ClickTracks will probably miss him, but John is doing great. I know Brett, you have met John several times as well.
Brett Crosby: Oh yeah. I think very highly of John. Everyone has shared their stories on their blog, so I'll share mine. The first time I met John Marshall was, I believe at the 2003 SES San Jose show. It was the first time I had spoken at that show and I think John's first time too. It was interesting because, it was pretty early on in SES history, I think there were maybe a few hundred people at the show or something, and pretty much all the vendors were essentially doing vendor pitches. John did something completely different, and it was great. He talked about the poster that shows Napoleon's march into Russia and all the casualties along the way. I forget the French guy's name who put the poster together, but it was a great presentation. Basically his point was, look this is a great way to show data and someone who is very passionate about illustrating a lot of information in one presentable format. He said as web analytics vendors we have a long way to go. Obviously he was right at that time and he is still right, we do have a long way to go. I think it gives a long bright fuure to the web analytics industry. Hopefully this latest version of Google analytics is a step forward on that, and I do not think we are fully there yet. We have more to go, I know there are more features people want out there, plenty of things, and we have got a very healthy road map ahead of us.
Eric Enge: Yes. Indeed. Well guys, thanks very much for taking the time to speak with me today. Very much enjoyed it.
Avinash Kaushik: Thanks, it was pleasure, Eric.
Brett Crosby: Thanks, Eric.
About the Author
Eric Enge is the Founder and President of Stone Temple Consulting (STC). STC offers Internet marketing optimization services, including SEO, Social Media and PPC optimization, and its web site can be found at: http://www.stonetemple.com.