Using Analytics to Your Advantage: Yahoo! Web Analytics Tips From Michael Whitaker

Today’s Y!Store blog is a guest column by Michael Whitaker, CEO of Monitus, which is located in Mill Valley, CA. Michael develops tools for Yahoo! Stores, blogs about Yahoo! Store and analytics, occasionally has seminars about Yahoo! Store, and is the author of two Yahoo! Store books.

I had the great pleasure to speak on the Data-driven Decision-Making panel at the recent Yahoo! Merchant Summit in Boston. As always seems to be the case, there wasn’t enough time to cover everything, so I foolishly offered to write up a list of analytics tips. This is by no means an exhaustive list, but I hope that it will provide at least some food for thought, and help you think about ways you can start using web analytics to your advantage.

Here are some basic tips I recommend for online retailers getting started with web analytics:

1. Add revenue metrics to reports wherever possible.

Revenue basically tells a more compelling story than conversion rate alone. Increasing your conversion rate may be a good proxy for success, but ultimately it’s about increasing revenue. Revenue also provides the basis and context for what you might want to improve on your site.

Let’s take keywords as an example. Suppose you have two different keywords, one with a conversion rate of 10% and the other 2%. If I give you the option of increasing both conversion rates by 20%, which keyword would you choose to optimize? The first one looks like a sure bet because your conversion rate would go up to 12%, whereas the second one would only go up to 2.4%. But what if you bring revenue into the equation? Let’s now assume that the first keyword generates $1,000 in revenue, but the second one contributes $10,000. The picture looks different now and you’d clearly pick the second keyword as it would make you an additional $2,000 instead of $200. Perhaps this is just stating the obvious, but your optimization efforts should take into account revenue and not just conversion rate.

This is of course not just about your best keywords. You also want to know your top landing and content pages, top traffic sources, top products, etc., all in terms of revenue. For more on this subject, check out Dennis R. Mortensen’s VisualRevenue blog article, "Using a Page Revenue Participation metric for Conversion Optimization."

One other note about this example. Notice that I have given you a choice to optimize one thing or the other. Sorry, but you can’t optimize both as you just don’t have the time! Seriously though, you cannot do everything, you cannot look at 80 reports, you cannot write great unique content for thousands of pages. You have to focus your efforts where you will get the best return. I will return to this theme again.

2. How do I add revenue metrics?

Here is a simple process you can follow: Pick a standard report, e.g., Navigation > Entries > By URL. This shows your entry/landing pages with visits and page views. Good traffic info, but we can do better. Next click on "Customize Report." Choose a revenue metric and drag into the main section. Hit "Save." By all means play around with this function and add other metrics that you might be interested in. Once you have a report you like, click on "Bookmark Report" and give it a meaningful title. This report will now be listed in the Bookmarks section in the left navigation.

Repeat the same process to build other reports. Standard Report > Customize Report > Add interesting metric > Bookmark report. Doing this also has another huge benefit. Once you’ve created a few powerful bookmarked reports you can almost ignore all the other standard reports! Instead of having dozens of reports, you can now focus on a select number of reports. Every once in a while you will want to do a deep dive into other reports, but having just a few key bookmarked reports is far more manageable.

Bonus tip: Send those bookmarked reports to yourself by email on a regular schedule. I am a big believer in prioritizing tasks by using reminder and to-do lists. Let’s face it; you’re busy, and life gets in the way, and you forget to check your analytics. So have the analytics come to you, perhaps once a week on a Friday. You can get extra fancy by using alerts, which will only send you email if a certain condition has been, such as an abnormal increase in traffic. To learn more about setting up alerts, you can consult this online help page.

3. Too much data. What to do?

See tip 2 and use a few bookmarked reports as opposed to looking at 80 different reports. It’s important that you set yourself goals that are actually attainable, and for that you have to ask yourself specific questions. If your goal is to increase your global conversion rate, where do you start? This question is far too vague and broad to elicit action. How about: "Let’s look at the top keywords driving traffic to my top landing page. It would appear that the third keyword contributes almost no revenue and has a high bounce rate. Let’s check the text on the page to see if the page is actually relevant for this term. Let’s also check where the traffic came from to see if we spot something." This is a specific problem that can you attempt to solve.

The name of the game is segmentation. Although we tend to look at overall conversion rate, it is not very actionable. The fact is that you have many, many different conversion rates, for landing pages, keywords, products, campaigns, etc. Don’t use site-wide averages – segment your traffic into meaningful buckets. Ask specific goal-oriented questions that get you to take action. YWA is chock-full of segmentation options and you can even segment by gender, age and some behavioral demographics.

Bonus tip: use the merchandising capabilities (located in the Settings menu) to bucket products into product categories. Simply upload a CSV file to YWA to associate products with categories (you can use a relatively simple breadcrumb RTML template to help you generate this file). The reason why you might want to do this is because it is easier to optimize, say, 20 product categories, than it is to optimize 2,000 products. Or at least the task will not seem as daunting.

4. Go for easy wins

Site Search

I love looking at internal site search data. These are the keywords that visitors enter in your site search box to look for information. Are visitors finding the content they’re looking for? Are they using different words than you do to describe a particular product? Try to incorporate the words your visitors use into your copy so that they will get a more relevant result. I also love the Zero Results Internal Searches report; this shows you the site search queries that resulted in no results being shown. One actual example I can share is the case of visitors looking for a brand that the retailer was not carrying. This definitely qualifies as actionable data because you can either start carrying that brand or suggest viable alternatives.

Bonus tip: Segment your site search keywords against referring keywords, i.e., those keywords that brought traffic to your site from paid or organic campaigns. For PPC keywords in particular, are visitors using site search on the landing page? If they do, this could suggest that you’re not sending traffic to the best spot.

Bonus tip two: If you add a revenue metric to the site search page, you may find that it has relatively low traffic, but accounts for a significant amount of revenue. Don’t treat site search as an afterthought.

Track 404 pages

Hopefully you get lots of traffic from search engines and other sources, but you also delete and create new content all the time. And of course it does happen that other sites link to pages on your site that no longer exist. In this case visitors will see a 404 page aka "sorry page not found." (You do have a 404 error page set up, right?) With a pretty simple technique, you can track in your analytics reports when those 404 pages are generated, along with the referrer info. I send myself this report every day.

5. What are your favorite analytics reports?

In tip 2 I suggested to start off with a standard report: Navigation > Entries > By URL and to add a revenue metric. But we can do one better than that. Customize this report again and drag the referring keywords dimension to the left column. This will now give you the top landing pages and the top keywords in one report! Not only is that very cool, but also very useful. These are the questions you can now ask: Are my SEO or PPC efforts working effectively? Are you trying to optimize for certain keywords that just don’t work on the landing page (by having a high bounce rate or low revenue participation rate)? If you think of matching or exceeding your visitors’ expectations in terms of content, good things will happen! Resist the temptation to optimize for too many keywords per landing page.

I also very much like to get a sense of where visitors go AFTER they have landed on your site. If the landing page is a section page, where do they click next? I’ve often seen that if you have a typical section page with a couple of dozen of product thumbnails in three columns, then the three thumbnails in the first row will get the clicks! Now, how did you decide how to order these products in the first place? Random or in alphabetic order? Why not arrange them in order of sales? Another way to look at this is that your visitors will click on the first three links, so you’d better show them the links you want them to go to.

Extend this idea to the main product navigation. There is no rule that says that you have to list your main product categories in alphabetical order. Consider putting your most important categories first and gently guide your visitors to where you would like them to go!

6. Be goal-oriented

Setting goals for yourself or your company is pretty much common sense, but you can extend this approach to web analytics. Having a bounce rate of 45% or 55% is not necessarily good or bad – what’s important is that you improve metrics over time, to go from a 55% bounce rate to 45%, for example. Trends are more important than absolute values. Set yourself an attainable target to improve ______________ [fill in the blank] over time. Goals in YWA are called ACTIONS and you have a number of prebuilt actions at your disposal, with ACTION = 01 SALE being the one you ultimately care about. But the way to increase sales is not always direct and you should also track "mini-goals," such as newsletter subscription rates. Not every visitor is in the market when they visit your site and could be in research mode instead, so perhaps you can try to sign them up for your newsletter. Or perhaps a customer comes back to your site to look for shipment tracking, so you could measure how effectively she can find the information. You should help your visitors find the information they need even if it’s not related to making a sale conversion. Another way to think of this is that visitors will interact with your visit across multiple visits and many different channels. Improving these mini-goals will help your business over time and hopefully turn prospects into customers, and then repeat customers.

Feel free to contact me with your analytics questions. — Cheers, Michael.

Michael Whitaker
CEO, Monitus
Guest blogger for Yahoo! Small Business

Dennis R. Mortensen talks about his new Web Analytics book and recent Yahoo! Web Analytics release

Today’s post is an interview with Dennis R. Mortensen, who sat down with us to talk about his brand new Web Analytics book, published by Wiley, and the recent release of Yahoo! Web Analytics.

Dennis is a pioneer and expert in the Analytics industry. He is an accredited Associate Web Analytics Instructor at the University of British Columbia, the author of Yahoo! Web Analytics: Tracking, Reporting, and Analyzing for Data-Driven Insights, and a frequent speaker on the subject of analytics and online marketing. An entrepreneur, Dennis was the COO of IndexTools until it was acquired by Yahoo! in May 2008. Today he is the Director of Data Insights at Yahoo!, sits on the Board of Directors at the Web Analytics Association, and maintains the highly popular analytics blog, VisualRevenue.

New features for Yahoo! Web Analytics were recently released. For Merchant Solutions customers, these included new demographics and interest groups reports, along with segmentation and cross-reference filter capabilities. What kinds of insights can merchants gain from these new features?

I honestly believe that there is an unlimited amount of insight that one can gain from this. The biggest barrier to a successful analytics practice within your company is not the analytics technology anymore – it’s likely to be the amount of time you, as a merchant, dedicate to the process. I, of course, believe that this is one of those processes that you should force upon yourself or your team, and that it should be no less than a weekly activity.

What I really like about the new release is the opportunity for merchants of all sizes to gain detailed insight and information on the demographics of their leads and customers, and that this information can be segmented down to individual products or groups. This is typically something that is exclusive to the enterprise company. Imagine the ability to get understanding and insight such as:

Target Segment for Product X = Female New Yorkers between the age of 18-24 with high interest in sports!

If that isn’t sexy, I don’t know what is. AND remember, this is something you can go do today. It’s not the only thing you can do, of course, but I believe it shows the value of the new dimensions. The most beautiful part is that we don’t just provide this as pretty reports, but as actual dimensions in the system, so you can use this to slice and dice your data any way you want. Creating a simple report on gender and average order value, as pictured below, is easy and super powerful, showing that the average order value from males are +$100 more than from females.

Report on gender and average order value.

Even with all this data at their fingertips, a hindrance to many busy merchants is lack of time in their day. If you had to pick key metrics that merchants should keep close tabs on in the limited time they have, which would these be, and why?

If you have very limited human resources, I suggest that you rely on Alerts for metrics such as:

• Visit to Sale conversion rate
• Average Order Value
• Traffic Distribution

Make sure that you are keeping an acceptable visit-to-conversion rate while maintaining an acceptable average order value – and should you not, be sure that you are alerted instantly. Finally, knowing that there are no major changes in traffic influx, makes sure that you don’t lose out on a potential opportunity. Remember, alerts aren’t just for negative scenarios.

Create automated alerts to stay on top of your key site metrics.

You recently published the book Yahoo! Web Analytics: Tracking, Reporting, and Analyzing for Data-Driven Insights, a resource that promises to give readers "Yahoo! Web Analytics secrets and tricks not found anywhere else." Which topics covered in your book are a definite "must-read" for online store owners?

My philosophy in regards to Web Analytics as a whole today, is that you should focus on three different but equally important tasks. I have divided the book into those three parts to reflect these broad tasks: A) Collecting Data, B) Reporting on Data and C) Deriving insight from Data. Depending on one’s vantage point, one or more of the chapters will be in focus.

The book of course holds tremendous value for all existing users of the system, and will be a hefty resource for all those new accounts we are setting up these days. BUT I feel it is important to mention that it actually holds great value beyond users of Yahoo! Web Analytics, or this is at least something I tried to author.

Yahoo! Store owners have the luxury of not having to tag their sites manually, so as a first step I would actually recommend Part II about reporting on your data and Part III on deriving insight on your data. Once you master this, to become a true web analyst Ninja (as my good friend Avinash Kaushik would call it), you can go back to part I and look into more sophisticated data collection methodologies. These are the two MUST-read chapters, even though I believe the whole book is a must-read of course:

• Merchandising Tracking (Chapter 4)
• Merchandising Reports (Chapter 10)

That said, I believe that dependant on one’s vantage point, certain chapters will be in focus, such as:

• Campaign Manager: Paid Search Analysis and Optimization with YWA (Chapter 11)
• Web Analyst: Using Segments in YWA Reporting (Chapter 7)
• Manager: Using Dashboards in YWA (Chapter 9)

You can find the official book page on my blog, and you can go get a sneak peek of the book at Google Book Search or over at Amazon.

Front cover of Dennis R. Mortensen’s new Yahoo! Web Analytics book

Yahoo! Web Analytics is available to and used by merchants of all store sizes and experience levels. How helpful is your book to merchants on each part of this spectrum?

I don’t necessarily believe that the book caters to a specific store size or experience level. I am more eager to believe that it depends on the person and what they want out of it. I left nothing out from a feature set point of view and as such it is indeed sophisticated in its attitude, but I tried to include examples and use a language that makes it pleasant to read. I believe the bellowing two testimonials might confirm that:

"As the individual who drove the initial development of Yahoo!’s Web Analytics tool, Dennis managed to conquer mind-numbingly complex issues by presenting them in a simple and useful way. It’s no wonder that he did it again with his Yahoo! Web Analytics book."

Bryan Eisenberg,
New York Times bestselling author of Call to Action and Always Be Testing, and cofounder of FutureNow Inc.

"In Yahoo! Web Analytics Dennis Mortensen manages to do the impossible by adding real value to our knowledge of web analytics in an already crowded market. His clear language and excellent examples make this book required reading for any web analytics practitioner interested in extending their use of freely-available tools. Dennis is one of the best and brightest in the web analytics industry and Yahoo! Web Analytics reinforces that with every page."

Eric T. Peterson,
Author, Web Analytics Demystified

Thanks, Dennis, for taking the time to talk with us about your new Web Analytics book, and the recent Yahoo! Web Analytics release.

For merchants attending Internet Retailer Conference & Exhibition in Boston next week, you can talk to Dennis about Yahoo! Web Analytics in-person – be sure to stop by the Yahoo! booth (#449). Dennis will also be presenting at the inaugural Yahoo! Merchant Summit, being held in conjunction with IRCE on June 18.

Jennifer Farwell
Yahoo! Small Business

Yahoo! Web Analytics: New Look, More Reports, Enhanced Features

Have you checked your Yahoo! Web Analytics reports this week? If so, you may have noticed some improvements and enhancements to the Yahoo! Web Analytics reporting interface, along with new demographics and interest groups reports, segmentation, and filters; a demographics and interest groups dashboard; the ability to create more scheduled reports; and the addition of custom fields along with a greater number of actions (more on that in a bit).

New Yahoo! Web Analytics reports, segmentation, and filters

How well do you know your site visitors? Probably not as well as you’re about to. Now you can get even closer to your current and potential customers, with more detailed demographics and interest group reporting. Next time you’re using Yahoo! Web Analytics, remember to check out these reports, most of which are also included in the new demographics dashboard:

• Age report – breaks down the age of visitors to your site in the following categories: under 17, 18-34, 35-54, and 55+.
• Gender report – breaks down the gender of visitors to your site as male, female, or unknown.
• Age & Gender report – contains cross-tabbed age and gender information (e.g., 18-34 female).
• Interest Categories report – shows the top interest categories of your site visitors.
• Y! Properties of Interest report – shows the top 20 Yahoo! properties also visited by your site visitors.

Yahoo! Web Analytics demographics dashboard

You can also apply age, gender, interest categories, and Y! properties of interest segmentation and cross-reference filtering to your reports. This information can help you with demographic targeting of Yahoo! Search Marketing campaigns on search or content networks, and of AdWords campaigns on content networks. Learn about report segmentation and cross-reference filters.

More scheduled reports

Previously, you could schedule up to three reports to be delivered via email or FTP on a daily, weekly, or monthly basis. As of this week, you can create up to 10 scheduled reports to better help you keep tabs on your site activity, and to help you with planning promotions or fine-tuning marketing campaigns. For example, if you evaluate your search campaigns and campaign spend once a week – say, every Tuesday – you may want a report delivered every Monday of top performing campaigns for the previous week.

Get into the action(s), and introducing custom fields

When Yahoo! Web Analytics was first released for Merchant Solutions Standard, Professional, and Yahoo! Store accounts, you could define up to nine custom actions to manually include in your site tracking. With this latest release, you may now define up to 50 actions. Learn how to define and track actions.

We also offer limited support for creating and tracking up to 38 custom fields. You can use custom fields to add specific reporting categories to the reporting structures already available in the system. Merchants should note that custom fields require manual setup and page tagging. We’ll soon be providing a Merchant Solutions-specific help page where you can learn more about custom fields and how to integrate them.

Digging into your data

The information available in your analytics reports can play a major role in helping you increase your store’s sales and revenue, along with the effectiveness of your marketing campaigns. If you’re not sure where to begin, or if you’ve enabled Yahoo! Web Analytics and are looking for help and training materials, there are several resources available to you:

• Our online help center offers a number of help articles that take you through enabling Yahoo! Web Analytics for your store, customizing tracking code, using key reports and features, and common troubleshooting questions.

• MarketMotive offers a free 30-minute video showing you “4 Quick Ways to Improve Your Store,” using information from your analytics reports. This video tutorial is part of a full Yahoo! Web Analytics course, which covers important topics such as actionable reports, key performance indicators (KPIs), segmentation ideas, campaign configuration, custom reports, and more.

  
  MarketMotive’s free Yahoo! Web Analytics training video

  The complete course is available for $199, and MarketMotive is offering a 10% discount to the first 2,100 merchants who register using coupon code YMRK3YWTEN. Find out more about MarketMotive’s Yahoo! Web Analytics training.

• For merchants interested in Yahoo! Web Analytics consultations, several Yahoo! Merchant Solutions developer partners offer consulting services. For a list of these developers, please see our Yahoo! Web Analytics training and consulting page.

If you’re a Yahoo! Merchant Solutions Standard, Professional, or Yahoo! Store merchant and haven’t yet enabled Yahoo! Web Analytics for your store, we strongly encourage you to do so. The information in your reports can provide you with important knowledge and insights that are key to your store’s success.

Jennifer Farwell
Yahoo! Small Business

Yahoo! Web Analytics tips from Rob Snell and the Yahoo! Store Power Hour

If you missed 1 Choice 4 Your Store’s recent Yahoo! Store Power Hour, be sure to have a listen over at their show archives. Rob Snell — long-time Yahoo! Store owner, occasional Yahoo! Store guest blogger, and author of the Yahoo! Store book: Starting a Yahoo! Business for Dummies — spent the hour talking with Shawna Fennell about Yahoo! Web Analytics, and why merchants should be using this feature to collect and analyze their site data.

During the show, Rob discusses:

• how to learn which channels are best for driving sales
• segmenting traffic
• using Yahoo! Web Analytics for short-term troubleshooting, long-term site improvement, data collection, and identifying trends
• what retailers need to watch for
• favorite reports and filters, and where to find them

Go to the Yahoo! Store Power Hour archives

Jennifer Farwell
Yahoo! Small Business