Ten Steps You Can Take to Improve Your Web Analytics Program – Part 1

Web analytics can provide significant insight into how your website is doing as far as contributing revenue and leads to your company. Often times, a company will install a web analytics tool and simply use the “out-of-the-box” reports, including visits, visitors, unique visitors, top pages, average time on site, and so on. When used properly, a web analytics tool can provide you with the data you need to make decisions regarding web strategy, content, layout, usability, applications, paid media, referral partners, and personalization. If the goal of your website is to generate sales or leads, the bottom line is the conversion of visitors to customers.

To get more out of your investment in your web analytics tool and the personnel who are using it, and to increase your conversions, here is a list of ten ways to get the most out of your web analytics program. These questions will be presented in two parts. Part 1 will address  visitor behavior, and Part 2 will address converting visitors to customers. Here are the topics for Part 1.

Measuring visitor behavior

1. Identify where your visitors are dropping off

If the goal of your website is to generate revenue, either directly from sales of goods or services on the site, or by generating leads that are followed up off-line, it is important to know how effective your site is in getting to that goal. One of the more useful tools that your analytics tool provides is a conversion funnel.  A “conversion” is loosely defined as the desired outcome of a visit. This can be a purchase, reservation, subscription, registration or a form completion to request more information or be contacted for an offline sale. Depending on your tool, a conversion funnel can be called a “fallout report” or “scenario analysis”. It is set up by defining a list of URLs that represent the desired steps that a visitor should take to get to the goal. In some cases, you need to tag each page ahead of time. In others, you set up the pages in your analytics tool.

Once you have your funnel set up, you can see the percentage of visitors who either make it to the next step or who “drop off” or “fall out” of the conversion process. At first glance, you can see which pages become “bottlenecks”, where you lose a large percentage of visitors. You can then study the pages to better understand why visitors do not go any further. With a proper A/B or multivariate testing program, you can make changes in these pages and then test their effectiveness in increasing conversions, and thus, revenue from your site.

The second item to analyze is where visitors go if they leave the conversion funnel. Here some of the questions you need to answer:

  • What percent of visitors just exit the site?
  • What percent go to the home page?
  • What percent look for product reviews, privacy information, refund policies, FAQs or other content?
  • Do they get distracted by having too many non-conversion links on a key conversion page?
  • Do they have the opportunity to go back and change their mind instead of completing the purchase?
  • Do they read some content then come back to the funnel?
  • What percent use onsite search?

By analyzing the paths that visitors take from your conversion funnel, you can better understand your visitors’ behavior and thus make improvements to your site with the goal of increasing conversions.

2. Use segmentation to help you understand visitor behavior

One of the most common terms in web analytics is “average”. We have average pages per visit, average time on site, average revenue per order, and so on. While averages are very useful for trending purposes and comparing metrics from month-to-month or year-over-year, you must keep in mind that there is no such thing as an “average” visitor. Each visitor to your site is unique, and to better understand visitor behavior and how to optimize your site, you need to segment visitors based on something in common.

Some of the most common segments are as follows:

  • New vs. return visitor
  • Paid search vs. organic search
  • Direct visit vs. referrer link
  • Google vs. Bing vs. Yahoo!
  • Geographic location
  • Returning customer vs. returning visitor
  • Paid search vs. banner vs. email campaigns
  • Online vs. offline marketing
  • Weekday vs. weekend visits

One you have the appropriate forms of segmentation, you can then look at metrics such as average time on site, average pages per visit, average searches per visit, average revenue or leads per visit and so on for each segment. By analyzing specific group behavior, you can create more targeted action plans to better talk to each type of visitor. By using cookies to store segment categories on each visitor’s browser, you can later read these cookies to enable more personalized or targeted content, with the goal of increasing conversions.

3. Measure interactions with Flash objects on your site

Developing Flash objects can often require significant resources in terms of man hours and budget. With proper planning and tagging, you can track the usage of these assets and determine if they bring value to your site and business.

At the most basic level, you can insert tags in the Flash source to count the number of times an object is “touched”. You can also go further by tagging interactive buttons on the object and then links to other site pages. If the Flash object is playing a video or an animation, you can insert tags at key stages in the progress. These can be at percentage points, such as 25%, 50%, 75% and 100% viewed, or when specific topics are reached. Depending on your analytics tool, you can create “events” at these measurement points, pseudo pages, or conversion variables.  You may be able to correlate visits to particular points in the Flash to orders or leads. You can also see how much of your message was viewed by those who engaged with the application.

If your Flash application has a “Call to action”, you can measure what percent of those who engage with the app take the next step to reach the action page. By tagging this as the start of a conversion funnel, you can determine the conversion rate of those who take the desired action, compared to your overall site traffic. By developing metrics around your Flash objects, you will be able to provide data on the effectiveness of the application, and whether to refine it or ditch it.

4. Use tagging to tell you how visitors are using your site

Many sites have tools such as onsite search, and some have third-party tools such as click-to-talk and click-to-chat. If these links are not tagged, your reporting can provide inaccurate or incomplete results. For example, if you have a reservation-based site, and on the payment page there is a click-to-talk or click-to-chat button, it is important to tag these links as pages. If not, your reporting may show that the “abandonment rate” is higher than it really is. If the visitor decides to click to speak with a representative and completes an order over the phone, or through a chat interface, your reporting will show that they simply exited the site. It would be more accurate to report that the visitor “visited” the chat or talk page instead.

It is also important to track where visitors are either clicking on click-to-chat or click-to-talk buttons along with using onsite search to better understand their behavior.  If there is a high use of these services on particular pages, it may indicate a usability issue on your site, or missing information that is needed to continue. In the case of onsite search, if you can capture the search terms associated with each step in the conversion funnel process, you can gain some insight to where visitors may be lost or missing information they need to continue with their purchase, reservation, registration or other desired action. By looking at patterns, you can make adjustments to your site and reduce these distractions, increasing conversions.

By properly tagging these actions and other events, you may also be able to look at pathing from these stages to see where visitors go after they do an onsite search or click to talk or chat. By understanding their behavior, you can make the needed improvements. You may also be able to measure what percent of these actions still lead to a conversion, and which ones do not.

5. Determine if visitors are seeing your key messages

Content managers for websites often spend significant resources to create their brand message or “Why Us” message to differentiate your company from its competition. The question is, are visitors seeing it and how can you improve its visibility? Key message pages can be important in creating your company’s unique selling position and convincing visitors to do business with your company. In a typical analytics environment, a “Top Pages” report will be generated that shows that the key message pages received “x” visits and “y” page views, and that they were the number “z” page in ranking. While this top-level information is nice to have, you can go deeper and provide more insight into how these pages are used.

If your site has multiple key message pages or pages that focus on different topics, you can group similar pages into defined content groups. For example, a branded drug website may want to know how well their site does in the following areas:

  • Information about the disease and complications
  • How the featured drug provides benefits
  • Clinical studies and other physician-related information
  • Call to action (talk to your doctor, request more information)
  • Engagement (user tools, calculators, worksheets)

Instead of providing reports on all these page visits and views, you can simplify this by tagging these pages to place them in different content groups. You can then provide high-level metrics or KPIs that show what percent of visitors take part in each of these content areas. This will be more meaningful to those who manage the site’s content. There are several ways to tag pages, including hard-coding each content group onto each page, using JavaScript to identify the group based on the URL, using a content management system to place tags automatically, or even by filtering the content with your analytics tool.

Along with tagging content groups, your team can also identify “quality pages”, or pages that they deem important to either the site, brand or company, then tag these pages as a “quality page”. You can then create a metric that shows how many quality page views per visit your visitors see, and what percent of total page views are quality pages. If these pages are not resonating with your visitors, you can then look to identify reasons why. Some of these reasons can include:

  • Links are positioned “below the fold”, especially on laptops (link is below the initial visible screen area).
  • Links are hard to find
  • Links are embedded in Flash navigation that is not user-friendly

By tagging and testing link placement, you can improve the visibility of the links to the key message pages.

This completes Part 1 of this two-part look at ten ways to improve your web analytics program. In Part 2, I will take a look at five ways to use your analytic tool to measure the process of converting site visitors to customers or leads for your business.

2 thoughts on “Ten Steps You Can Take to Improve Your Web Analytics Program – Part 1

  1. To optimise conversions you should employ a web analytics provider such as clicktale to show you how your customers enagage eith your site. From the results fo the aggregate behaviours, real time videos and heatmaps you can mend any deficiencies and make it easier for visitors to your site to become buyers.

  2. Pingback: Ten Steps You Can Take to Improve Your Web Analytics Program – Part 2 « Edgewater Technology Weblog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s