Web analytics for small e-retailers

Google Analytics is the most popular provider of web analytics to smaller e-retailers, but a quarter of Second 500 users find Google doesn’t meet all their needs.

Allison Enright , Editor

Google Inc.'s Google Analytics tool is by far the No. 1 web analytics tool used by the largest web retailers in North America, with 534 of the 1,000 retailers ranked in Internet Retailer's 2013 Top 500 and Second 500 Guides using it to track performance statistics like time on site, clicks and referral traffic sources.

Retailers say the key selling points for adding Google Analytics tracking codes to their sites is that its basic edition is free, and that it helps make buying and managing ads on Google properties easier. That's because e-retailers can make Google Analytics data, such as keywords and top-clicked products, easily feed into systems like Google AdWords for managing paid search ads on Google.com and display ads on Google's DoubleClick ad network.

But while e-retailers use Google Analytics more than any other analytics service overall among the Top 1000, it's with smaller retailers—those ranked in the 2013 Second 500 Guide, where 2012 revenue ranged from $75,000 to $18.7 million—that Google Analytics dominates. 340, or 68% of retailers in this range use it, whereas 39% of retailers in the 2013 Top 500 Guide say they use Google Analytics. (The No. 1 analytics system used by North America's Top 500 web retailers is Adobe Systems Inc.'s Omniture, with 41% saying they use it, although some in this group use more than one service, according to Internet Retailer's Top500Guide.com.)

But a closer look shows that 25% of Second 500 e-retailers that use Google Analytics also employ at least one other service to deliver site analytics data. That suggests Google Analytics doesn't provide all the data smaller e-retailers need.

Some of the analytics services Second 500 e-retailers say they use are tied to the vendors that they use to run their web sites. For example, many Second 500 retailers that run their sites on the Yahoo Stores platform use Yahoo analytics along with Google Analytics. That makes sense, as Yahoo Inc.'s analytics come built in with their service. But a lot of Second 500 retailers say they use services, most of which they pay for, that measure the same activities Google Analytics can track, or perform the same or similar functions that Google Analytics is capable of doing, too, for free.

Why pay for these supplemental services? Retailers cite a multitude of reasons. The most common one is that these other tools, which include Mixpanel, Optimizely, CrazyEgg, KissMetrics and others, are easier to use than Google Analytics and point retailers more directly to the actions they should take next. Smaller retailers say that's important because few of them have hard-core data analysts on staff able to tease out salient data from Google Analytics and interpret it in a way that drives results.

Google says it is hearing that message "loud and clear" from small business and web site owners. In response it is developing training programs aimed at non-technical people to teach them how to get the most out of Google Analytics. That's in Google's interest because as retailers learn Google Analytics better they will also learn how to use that data to feed into Google AdWords campaigns, and that could lead them to spend more advertising with Google. It's a message Google is particularly hearing from larger Second 500 merchants who are particularly likely to be supplementing Google Analytics with other software to get the information they need to grow.

In addition to Google Analytics, office supplies e-retailer Honest Office Inc., which with $1.6 million in 2012 sales ranks No. 973 in the 2013 Second 500 Guide, uses a handful of other analytics services to measure web activity. Those services include Roxr Software Ltd.'s Clicky, which Honest Office pays $10 a month to show on a dashboard key metrics including daily visitors, where visitors came from, time on site and bounce rate.

The e-retailer also uses Mixpanel, whose pricing varies based on the number of so-called "events" Honest Office asks it to track. An event is an action taken on the site, for example a consumer adding a product to the shopping cart. Mixpanel's analytics then let Honest Office see subsequent actions, say how many people then went to the checkout page, and to segment that data in other ways, such as by what country they came from. It provides visibility down to the individual user level, and retailers can track up to 500,000 events monthly for $150, according to Mixpanel, although Honest Office president Jeremy Biron says he pays less.

Biron says he knows he can get the same data he gets from Clicky through Google Analytics, and can track events to some extent through a tool added to Google Analytics last year as part of its Universal Analytics update Google began rolling out last year. But he likes the ease of the other systems and hasn't yet updated HonestOffice.com to Universal Analytics. Biron says he checks his Clicky dashboard daily, Mixpanel a little less frequently and Google Analytics only when he really wants to poke around in the data. "In terms of the number of data points, no one can say they offer more than Google Analytics, it offers anything you can want," he says. "But it can be overwhelming. It's just not that easy to break down."

Honest Office, with fewer than 10 employees, doesn't have an analytics expert on its team that can dig through data all day long, Biron says. "These other tools make it easier to see what really matters," he says. "I can make quick insights and take actions based on those." He adds that the Clicky and Mixpanel interfaces were simple to set up and their reporting dashboards are easy to interpret. Honest Office uses an analytics integration tool called Segment.io to channel its site data to Google Analytics, Mixpanel and other services, such as web site testing tool Optimizely. He does that through a single API, which Biron says saves him time and money.

At jewelry e-retailer Brian Gavin Diamonds, which with more than $7 million in web sales comes in at No. 700 in the Second 500 Guide, executives choose analytics products in large part based on how easy they are to use and how little stress they put on staff developers. Danny Gavin, director of marketing, says he checks Google Analytics reports at least every other day and makes adjustments to paid search ads with the data fed from Google Analytics to Google Adwords every day.

But when it comes to understanding how the 40,000 to 50,000 customers who come to BrianGavinDiamonds.com each month behave on the site, he prefers site analytics service CrazyEgg. CrazyEgg uses heat maps to display where consumers click on the site, and Gavin says CrazyEgg provides more usable details about click paths than Google Analytics, which does not offer heat maps.

"Let's say I have a link at the top of the page to my diamond page and a link at the bottom to the exact same page. Google doesn't look at the positions of those links, it only looks at the total," he says. "Looking at a Google Analytics report, I'd just know that 50 people clicked, but I wouldn't know how many clicked on which. Having that level of clarity is important when you are moving things around on the site." CrazyEgg shows which link got the most attention.

CrazyEgg fees are based on traffic levels and they start at $9 a month for fewer than 10,000 visitors. BrianGavinDiamonds.com started using CrazyEgg several years ago when the retailer was attracting less traffic. For its current traffic levels, it pays $49 a month for the heat-mapping analytics program. "The fee is really reasonable," Gavin says. "I got so much data and insight at such a low price point that it sucked me in [several years ago]."

While not a web analytics tool in the strictest sense, BrianGavinDiamonds.com also uses Optimizely Inc. for A/B site testing. Google added a site testing tool to Google Analytics in 2012 called Content Experiments, but it requires a merchant to create a new page to run tests. At BrianGavinDiamonds.com that would require roping in the development team, whereas Optimizely requires no coding or additional work for the development team, Gavin says.

Plus, Gavin and his marketing cohorts can create and turn on and off tests as they please. That's worth the price it pays for Optimizely's mid-priced "silver" package, he says. "I try to keep I.T. out of it. In our company we have a list of [development] priorities, and having a way for marketing to do this ourselves is appealing," Gavin says. "I'd rather spend $71 a month and do as much as possible."

Google is aware of these complaints and says it is looking at ways to make Google Analytics easier to use, especially for marketing personnel. "A lot more people are getting into analytics and they are not programmers, they are marketing people," says Justin Cutroni, Google Analytics evangelist. "Analytics may not be their primary function, but they need to understand the data to make decisions." Cutroni heads up Google's Analytics Academy, which provides free online courses to teach people how to glean insights from Google Analytics.

The first course launched last fall and a new course was added in March. Each takes about three hours to complete and sub-lessons are comprised of a video tutorial, an activity and an exam. Students, through the Google Plus social network, can connect with Google employees who act as virtual teaching assistants to ask questions one-on-one, Cutroni says.

More than 30,000 people completed the first course, and about 50,000 signed up for the new course in the first two days following its launch. "These courses will help people that don't have a tech background understand the actionability of the data so they can make better decisions," he says.

Naturally, that includes spending money with Google. "We want people to understand [Google Analytics] is one piece of a pretty big system, including AdWords and DoubleClick, and not to think of [Google] Analytics as a stand-alone tool. We want them to be able to immediately take action on that data," he says.

Action is what Second 500 retailers that are close to breaking into the Top 500 Guide want from their web analytics data. These retailers are more likely than smaller retailers to have in-house analysts devoting time to exploring the far reaches of Google Analytics and other analytics systems.

Web-only men's apparel retailer Trunk Club, which ranked No. 558 in the Second 500 Guide based on 2012 sales of $15 million, for example, has four to five people deployed on monitoring analytics and the performance of TrunkClub.com and its applications. Another 25 employees tap into analytics on an ongoing basis to get data they need to do their jobs. "Analytics is a core part of our culture," says Tim Grace, head of product for the retailer, which requires its customers to fill out a profile about their lifestyle and clothing style and then pairs them with a stylist who chats with them before shipping them outfits to try.

"It's the marketing team that uses Google Analytics the most, largely because they use it to understand the performance of their marketing efforts and to understand referring traffic," he says. "I'd say it's about 90% the marketing team and 10% the rest of us." That 10% often spend their time working with Mixpanel and on KissMetrics, which Trunk Club uses mostly to track and record the actions that lead to a conversion, including interactions that occur offline. KissMetrics tracks activity down to the level of each shopper so Trunk Club can see the path John Smith takes to become a Trunk Club customer. That's not something Google Analytics can do, because it only provides anonymous data and only monitors online activity.

Despite the multiple vendor-supplied analytics tools it uses, Grace says Trunk Club has had to develop some analytics processes on its own. A particular hurdle was how to track and record campaign attribution for marketing on mobile devices. "We struggled with that, and we found analytics [services] were doing a poor job of tracking what was going on there," he says. "We had to figure out what the right way was for us to do that." Trunk Club's in-house development team ended up creating its own way to track attribution on mobile.

Grace says he's largely satisfied with the capabilities provided by Mixpanel and KissMetrics, but would like to be able to generate more custom reports with the data those systems collect. "We want to use that data to answer specific questions, and that's not something you can do very easily," he says. Dan McGaw, director of marketing at KissMetrics, says the vendor is working to create multiple new reports to answer more questions its KissMetrics users have. Mixpanel did not respond to an inquiry.

Children's apparel e-retailer FrenchToast.com president Michael Arking says he wants answers too, and he enlisted earlier this year for the first time the services of a business intelligence consultant to extract them from his analytics data. The e-retailer, with Internet Retailer-estimated 2012 web sales of $9.3 million, ranks No. 692 in the Second 500 Guide.

Arking says he's a big fan of Google Analytics and he looks at it daily—sometimes hourly—to monitor top-line metrics like visits, revenue and transactions to keep track of how the day or week is going for FrenchToast.com, and he's fine with doing that. He also checks analytics reports provided by other vendors the site uses, such as SocialAnnex, for data about Facebook social network activities that Google Analytics can't get. Google Analytics users can see what traffic arrives from Facebook, but Facebook doesn't allow it to track other activities on the social network. By providing tools like social login and social sharing, SocialAnnex can collect a greater amount of social network activity data. But Arking says he wants to get more out of the data he knows is stored in Google Analytics, and that's where he ran into problems.

"There is a ton of data available from Google Analytics, and it is getting way more complex," he says. "As we dig deeper, it gets harder to be actionable. I am not a business analyst, so I went hunting for a third party that could dig in deep and answer the questions that are gnawing us." One such question: Why is the conversion rate on FrenchToast.com lower for consumers using the Google Chrome browser than consumers using Safari? One answer it found is a security pop-up appears on some Chrome browser versions.

With answers to these types of questions now flowing in, Arking says he's feeling more confident in the actions he's taking based on his analytics. For example, FrenchToast.com is now testing alternate designs on pages where analytics indicate the bounce rate is higher.

The e-retailer also is wading further into analytics. It will begin using Tealium's tag management system next month to better track FrenchToast.com's digital efforts. The tags will allow for more specific reporting on things not being tracked now, such as how many consumers start filling in text boxes, like those on a shipping or payment page, but don't complete them. Arking says with custom tags he'll be able to pinpoint where consumers abandon a form. Right now, the site only tracks how many submit the form.

As they pursue growth, the smaller retailers that comprise the Second 500 Guide want to use the data they collect through various analytics platforms just as effectively as e-commerce's biggest players. For now, it is a matter of vendors being able to deliver the data in a comprehensible way that personnel without computer science degrees can act on quickly.




April 2014 magazine, BrianGavinDiamonds.com, Clicky, CrazyEgg, Danny Gavin, FrenchToast.com, Google Analytics, Google Inc., Honest Office, Internet Retailer Second 500 Guide, Jeremy Biron, Justin Cutroni, KissMetrics, Michael Arking, MixPanel, Optimizely, Roxr Software, SocialAnnex, Tealium, Tim Grace, Trunk Club, web analytics, web analytics program