ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

10 signs you don't understand web analytics

10 signs you don't understand web analytics Nicole Rawski

Web analytics has always been an important aspect of digital marketing, but only recently has it been considered critical to success. The data mined through savvy analytics tracking can maximize any marketing budget by driving conversions and results; however, this only comes to pass if you understand what the data are telling you.

Web analytics is critical to measuring the success of your website as you identify opportunities to improve your business and marketing initiatives. Most marketers are confused with business implications of analytics reports, so it ends up being simpler to track what you know.

Web analytics has made great strides in the past decade, from when hit counters were the most sophisticated metric trackers in use. But there is still a long way to go before analytics is used to its full marketing potential. In this article, I hope to articulate ways that marketers can maximize their analytics solutions. To do so, I will discuss the top 10 signs that the power of analytics is still eluding your company.

Sign 1: You bought the tool a year ago, and you're still not measuring your business objectives.

You were sold on the lure of data. You bought a web analytics tool, have dashboards sent to you every week, and have no idea what they're supposed to be telling you about your business. You know there is great information contained in these reports and dashboards -- but you cannot decipher how it helps achieve your goals. Don't worry, you're not alone. The overabundance of data makes it difficult to sync graphs with meeting agendas.

There are several avenues marketers can explore to align analytics with business objectives -- not the least of which is your analytics teams. Keep everyone in the loop as changes occur and reevaluate reporting on a regular basis to ensure marketers have the ammo they need to create educated strategies.

For example, one business objective for a B2B website is to increase the number of qualified leads that come through the site. Measuring the total number of leads is one relevant measurement of this goal, but it's more revealing if compared with comprehensive traffic numbers. Additionally, if campaigns are driving traffic, monitoring conversion rates will track whether website visitors reflect target audiences.

Sign 2: You still have to remind the executive team what your company's KPIs are (hint: key performance indicators).

It is understandable that analytics jargon is not mainstream yet. As more CMOs ask for reports and sound reasoning behind digital marketing decisions, terminology will become more commonplace, and this point will be inconsequential. Until that day, keep framing campaigns, objectives, and goals within key performance indicators. This will keep everyone on the same page. It will continue to drive home the importance and value of grounding decisions in quantitative data. You can start one person at a time -- and the sooner you begin, the sooner you can make a difference. Prove that web analytics and KPIs mean something by suggesting sound improvement opportunities based on data. Your executives will be impressed and memorize any acronym you throw at them.

Sign 3: You create these beautiful dashboards, and no one knows what they mean.

Ah, the infamous dashboards that have been defined to include only the most important key metrics, yet you manage to cram in every possible minute detail, making it impossible to read. Pause a campaign, write more blogs, or increase an SEM budget? You still feel like you are flying blind.

The answer: Throw it away and start over.

Refine and redefine your business objectives, and assign a KPI to each. Understand what influences your KPIs, and how fluctuations affect the success of your business objectives. Once this is established, dashboards start to look different and useable.

Sign 4: Your boss is always asking about the number of "hits" your company's website is getting.

This has not been a valuable success metric since the '90s, and it is like nails on a chalkboard to any analytics professional. Potential consumers are not "hits" -- they are people who expect valuable content and a quality user experience when they visit a website. This goes back to point No. 2. Education is needed if we are to evolve beyond basic metrics and start unearthing valuable user insights.

The next time your boss asks how many "hits" the website is getting, simply respond with a smile and say, "I don't know, but since we began optimizing the checkout process, our conversion rate has increased X percent, resulting in an increase in $X dollars last month. Do you want me to look into those 'hits' for you?"

That will likely be the last time you are asked about a hit count.

Sign 5: You're still scratching your head wondering what you should segment and how it will help.

A great place to begin is to differentiate between new and return visitors. How do they behave differently, and how do they convert differently? What are some insights that you could glean to help increase conversion rates? Analytics can also help segment out users who are not looking for your website. To do so, create a segment that does not include your target audience and learn what your true conversion rate is -- and optimize from there.

Continuing with the example from point No. 1, segmentation also provides value in measuring your business objectives. I mentioned that the conversion rate can help assess if the target audience is who is arriving on a website. Segmentation can help provide insight into identifying new and returning visitors. New visitors may not fill out a lead form; however, they may download a couple of white papers, and returning visitors may fill out the lead form on their second or third site visits. All of this information can help marketers target their audiences appropriately.

Sign 6: Even though your homepage has an 80 percent bounce rate, your boss doesn't want a change because he/she likes the way it looks.

This is a classic push-pull battle. The difference is that one group is comprised of potential consumers who are the key to a company's success. To convince decision makers to trust their data, run tests to prove that the reports reflect consumer preference. Remain persistent on the implications of what not changing it could do to your business goals. If bounce rates improve, there will be little left to debate.

When evaluating a homepage for performance and deciding what needs to change, there are a few things to keep in mind: What is the value proposition? Why should a visitor engage further with the site? What is the call-to-action? Where are you trying to drive visitors within the website? Does the homepage have a clear navigation path for various user types? If the answer is "I am not sure" to any of these questions, you've found a great place to begin testing.

Sign 7: You're running multiple online marketing campaigns, and you have no idea which performs better.

For the first time, marketing campaigns can be measured to show their direct correlation and impact. Start from the beginning and tag online media campaigns so that you can measure how well each drives traffic to your site and how it converts. Knowing where to move your marketing dollars will become obvious once campaigns are tracked individually.

Google Analytics provides a URL builder for those who are not familiar with tagging campaigns. However, analytics solutions vary, so it would be best to check your implementation guide to clearly understand the requirements. The following is a hypothetical example of a campaign parameter using Google Analytics:


The "?utm" portion of the URL lets Google Analytics know that a variable is being passed from a campaign. The source ID will let Google Analytics know where the visitor is coming from, the campaign medium, and other useful statistics. These variables can be established to reflect whatever data make sense for a particular campaign.

Sign 8: The one time you ran an A/B test, a winner was chosen -- but no improvements ensued.

Testing is only one part of optimization, the next -- and most difficult part -- is implementation. Optimization is a continual process to make sure you are putting your best foot, ad creative, and messaging forward. What improves conversion rates today may not be as effective three months from now. With analytics, marketers are equipped to do more than just keep up with their consumers, and doing so will help attain goals and drive business.

A/B and multivariate testing are widely used in search engine marketing to determine the most effective ad creative. If most of your sales are completed offline, run testing with ad copy including phone numbers/store locations versus creative without it. Track conversion rates to see if prominently displaying this information on search engine ads makes a difference.

Sign 9: You still can't figure out why total site visits don't add up in all the reports.

It's important to understand that web analytics is not perfect, and sometimes 1 plus 1 equals 3. This does not mean that the information is inaccurate, but with an infinite number of variables and moving parts, numbers do not always match up. Web analytics data are still extremely valuable and important in gauging how a website performs. Remember to look at trends and major changes in key metrics. Continue to investigate what may have caused drastic changes until you have an answer.

A metric that is often used to measure audience mix is a comparison of new versus returning visitors. This metric is calculated as the total amount of new visitors divided by the total amount of returning visitors. A small number (0.3) indicates that the website has a healthy retention of visitors, while a higher number (5.2) indicates that the website has an abundance of new visitors. Marketing strategies that are in place determine the optimal rate for this metric. If an acquisition campaign launches and the goal is to drive qualified traffic, you should expect to see the ratio increase -- hopefully not too much, as you want to see the new traffic return.

Sign 10: You still design with HiPPO (highest paid person's opinion) standards in mind.

Great design drives conversions, so it needs to be strategically crafted to contribute to marketing goals. What worked on a previous project or competitor's site does not necessarily translate to your company. I am not advocating that you discount people's opinions or blindly follow analytics -- it is important to take both into account. Data are a great unifier and can help keep people on the same page in board meetings. By combining objective and subjective points of discussion, it is easier to come to sound marketing decisions. Even if the topic on the table is outlandish, cutting edge, or uncharted territory, at least data can be a place to initiate the conversation.

In conclusion, combining marketing insight with data is an extremely powerful and successful strategy.

Nicole Rawski is a web analytics analyst at Geary Interactive.

On Twitter? Follow Rawski at @nicolerawski. Follow iMedia Connection at @iMediaTweet.

Nicole Rawski can teach us all a thing or two about performance. As an Analytics Manager at Digitaria in San Diego, Nicole uses data to drive clients’ online presence and customer engagement. When she’s not helping clients reach new...

View full biography


to leave comments.

Commenter: Brandt Dainow

2009, December 30

Great article! Some of the points are so on the button they made me laugh! How many times have I watched ineffective design stay because someone "likes it."!!!!!
To those who criticised this - remember the title! It's not trying to tell you what you should do - that's too complex for any article. But this is a fairly comprehensive list of symptoms that something ain't right. All anyone needs to take from this article is that if one or two of these points hit home - you need to do something about it. Most companies are paralyzed with ignorance or apathy regarding web analytics - they need a wake up call before anything else.
If none of this applies to you, congrads! you're in the top 2%.

Commenter: Amanda Miller

2009, December 29

Loved your post! So many of your points ring so true! I think the last one is my favorite though. Just came from a place like that (notice the past tense ;P).

Commenter: David Shor

2009, December 28

Came across your article now at the end of 2009 since your article was included in the top 10 for 2009

Over the past two weeks of the holiday season I've read the two leading "pop culture" books on web analytics (Kaushik, Clifton). While any digital marketer worth a dime already knows most of that content, there were some real gems and I recommend you read them.

Key gems (in my opinion):

1. KIA vs. KPI - KPIs are like grade school. KIAs are like university degrees. KIAs are Key Insights Analysis - focusing 100% of the effort on asking questions that can be partially or completely answered through web analytics--and NOT focusing on the KPIs themselves. Ask questions like "how much more could we earn if we changed the results of X by Y percent?" and less "what is my conversion rate?"

2. Importance of ongoing and repeat surveys and heuristic/expert analysis to shift the balance to incorporate the voice of the customer rather than focus on pure web analytics.

3. The fact that being an A player in analytics is actually far more about interpreting the data and voice-of-consumer properly to gain those key insights.

4. 90/10 model - spend 90% of your analytics budget on people and 10% on tools. If you buy into the goal of key insights rather than key indicators, then you need people rather than more technology.

Thanks again for the great contribution!

Commenter: Greg Padley

2009, June 17

Awesome Nicole. I laughed (knowingly) as I read #4 and it's referral back to #2.

Unfortuantely, "some people" need to be reminded again (and again) which numbers are important and why. This will likely continue for some time.

However, as you note, if you keep telling them and keep showing results, eventually they'll know what you're talking about and trust the numbers (and you).

You don't say it specifically but hit on it several times - a big part of gaining understanding and enabling changes based on analytics comes from the "presentation" of the numbers.

It's similar the old refrain - Tell them what you're going to do for them, do it, tell them you've done it, then tell them again. Reframe. Reframe. Reframe.

Leading others in your organization to understand analytics makes you and your imput more valuable.

Commenter: Aurelie Pols

2009, June 12

Hi Nicole,

Interesting post. Lots of good tips in there!
I've always found that the use of the Internet for finding information shows, at least partially, demand for a companies' products.
Putting unique views of product pages next to sales for example gives an indication of possible lost opportunities, on which then can be tweaked upon through multiple ways.
Just as setting-up a basic report where unique visitors to your website are put next to callers to your call-center & people coming to your point of sales allows you to add perspective to your marketing reflection.

What we're seeing today is that Web Analytics is certainly not an island. It's often used in conjunction with other data sources or even integrated more and more into BI.

However it's being used, the important thing to remember is that great tools such as Omniture, webtrends, Coremetrics, Google Analytics, Yahoo! Web Analytics, etc. are just that: tools. Just as Word will never right your prose or poetry, WA tools alone will never provide you insight without human resources.
You mentioned Avinash Kaushik's HiPPO rule but I missed the 10/90 rule: for every 10$ spent on a tool, spend 90$ on brains.
Of the different industry reports I read, we're not there yet but the shift is slowly but surely taking place in order to put back the Analytics in Web Analytics and move beyond (monkey) reporting.

Commenter: Heidi Strom Moon

2009, June 11


Great points on how to use analytics to inform marketing decisions, and I like the way you organized it by "signs."

Later this month, a colleague and I will be speaking on using Google Analytics to drive marketing and our presentation covers many of these same points. As we see it, the most important concept is to get beyond meaningless metrics (like hits) that aren't tied to business goals or KPIs and find information that can lead to Avinash Kaushik's holy grail of "actionable insights."

To put it another way, who's coming to your site and buying stuff and how can you get more of those people?

Commenter: Todd Wente

2009, June 11

Good information. I believe it is important to remember that ultimately, your web analytics strategy and implementation must accommodate the necessary fluidity of business. Objectives change, paradigms shift, and healthy business strategies are usually in an evolution of flux. The immediacy of web measurement can be either a blessing or a curse depending on how well you define the objectives and interpret the data. The key in my opinion - set well defined performance indicators that map to the higher business goals, but be open to the possibility that the results may very well challenge both your tactics, your knowledge of the customer, and even the fundamentals of your strategy.

Commenter: Andrew Ettinger

2009, June 11

Well written!

Commenter: Rick Fraser

2009, June 11

Those of you that need to chime in a negative way. Where are you coming from?? PLUTO!!

Commenter: Rick Fraser

2009, June 11

For all of you that obviously Nicole wasn't talking to… those folks that already know everything. She was not talking to YOU she was talking to me…Yeah the person that sometimes gets a little complacent and sometimes need to take a step back or those that just don't know yet.
These people that criticize need to realize what the big picture is. It is not about you!!
It is so funny… All those that speak they need more information that is "NEW” "more informative" are probably the same people, I am sure, that a decade ago used to ask me "how much money you got” in terms of a SEO budget. In retrospect we now know it was just a matter of a few days of keyword stuffing and that is how they justified their 30K, all about money.
What are the words I am trying to think of… Hmmm f**k y*u maybe works…. Naahhh… It is more like YOU ARE READING THE WRONG F**kin article. Go away. I hate people that shoot snipes to make them self fell good.

Commenter: Nicole Rawski

2009, June 10

Thank you for the valuable and insightful responses to my article today. I have learned a lot from those who have chosen to comment.

Jack, like Rick says, maybe my follow-up article will be able to provide more 'valuable information.'

Amman, You are right, every business does have different goals. I apologize if I was not able to convey more specifics in this article. I'll try to next time.

Michael, thanks for the support and great analogy :)

Rick, thanks for understanding my intentions with the article's layout

Barbara, I appreciate your kind words and reference

Commenter: barbara reiner

2009, June 10

This is a really good article: comprehensive, organized, relevant and succinct. The time spent in preparation is obvious. I intend so share w/clients and colleagues.


Commenter: Rick Miller

2009, June 10

Nicole, I thought you presented a lot of good information in this post -- and by pointing out what others don't know, you're really showing them what they need to learn. Explaining all the nuances to Web analytics and the proper techniques would probably take many more words than you were alotted for this post. Maybe you get into the "valuable information" in a follow up?

Commenter: Michael Daehn

2009, June 10

This is a great post Nicole. It seems like people don't really want to learn how this stuff works, but they still want to control how others use it.

Either learn how it works, or trust the advice of the folks looking at the numbers- isn't that what you pay them for?

I don't tell my mechanic what to inspect on my car because I am clueless in that area. As long as I trust him, I make the repairs he suggests.

Commenter: amman kaur

2009, June 10

Jack, I agree with you.

Nicole, It is easier to point out what others don't know. How about you providing solutions to what people don't understand about web analytics. Every business is different and has different goals and unless you know the true path that the company is on, you can't really say that if a company is not doing something it means the people working there don't know much about analytics.

Commenter: jack Marketer

2009, June 10

It might be useful if you would share some of the "valuable Information" that you refer to.
Analytics undoubtedly is here to stay and a wise decision maker knows what they mean to his or her company. Your tone is both annoying and uninformed. You sound like those 80's computer guys(my mother complained about) before everybody caught up with the jargon. That is a weak thread to hang your future on!