Web analytics is a topic that gets mixed reactions, much like SEO really. Some people think it’s a myth, others are intimidated by the data, some know it is important but are limited to the basics and others can’t get enough (that’s me!). As an analytics specialist I find that my role includes the bits you would expect – implementing tracking code solutions and analysing data – but it also requires offering assistance in making the most of the data, from changing clients’ attitudes to encourage them to use their analytics more, to showing how the data can be used much more effectively.
The big thing behind this is that analytics isn’t about numbers.
Yes, you heard me right, it’s not actually about the data.
It’s about what the data MEANS.
It’s what you can DO with the data.
It’s about finding a context to the numbers and using this to understand the ‘what’s’ and the ‘why’s’, then using this understanding to make a plan.
I’ve written about finding the context behind numbers being an important and often overlooked issue in the past (see Beef, Arsenic and Web Analytics for details!), but I want to emphasise an important point today and help you find ways to make the most of your analytics through one simple method:
Using your web analytics data to tell a story.
Now, the word story may make this sound childish, but if that puts you off then try to think of yourself as being the Shakespeare for your website. It’s not child’s play; it’s about an epic novel or a thoroughly captivating play. It will have a beginning, a middle and an end. This is where you come in: by analysing the beginnings of the data you can test things out and optimise your website and marketing (the middle) before creating the best end to the story possible – business success!
There’s a reason we’re web analysts and not accountants (no offence!) – we’re the creative types looking to bring in colourful information about the users, the activity happening on site and the psychology behind conversions to make the data real and then help everyone achieve the best situation to make the numbers increase and the story reach a happy ending.
By bringing the numbers to life and making it real you can make much better decisions about what needs to be done than if you look at the numbers as numbers.
“Last month we had 8,435 visits” means nothing.
“Last month saw a 70% increase on last year” is better.
“Last month saw a massive 70% increase compared to the same time last year due to the marketing campaigns which drew new and more qualified traffic to the site, so let’s keep attracting more of this traffic while making sure the website suits their needs and we could see an even bigger increase next year!” is a much more real and usable analysis of a website’s analytics.
Obviously just a sentence can’t solve everything, but by adding some reasons, experiences and context to the data you can start to identify what parts of the data are successful and what the best routes might be for the future.
Being able to segment the data will help you bring it to life.
User Segmentation
You can segment your users to identify characters; this will then show you how differently they act and which ones are successful. Use this data to identify which target audiences your marketing campaigns need to reach.
On Page Activity
What happens on site is the plot, it’s what people do. By understanding what happens for different characters on site you will be able to understand which characters have a successful journey, which ones need help on their journey, which shouldn’t be in the story at all and how you ultimately make every target character’s journey through the story a successful one.
Website Review
By reviewing the design of your site you will understand the landscapes and scenery that the users are navigating through. An analyst should always be familiar with the website itself, know which areas are for which purposes and the pages that are deemed most important.
All of this can still be done in a professional manner, your reports don’t have to start using the words ‘characters’, ‘plot’ and ‘landscape’! This is just to help you see how useful your numbers can be when put into a story or a context.
Let’s take a look at how we can use this way of thinking to make the most of some SEO data.
Keyword reports – a common item reported on by SEOs and fundamentally important because search functions through the use of keywords!
It’s time to think about how these could be broken down differently. There are two main types of keywords; ‘money makers’ and ‘research terms’. If you can split out the data in to these two groups you could then profile the types of people behind each, find the pages most relevant to each, see whether users turn from one to the other (and why) and ultimately use all of this to plan how to attract the money makers more.
If you were to then break down one of these sets of keywords, let’s say the money making terms, you could learn even more about them. How about breaking them down in to high, medium and low groups for conversion rate, ROI or importance to the site and then comparing the groups. This can help you identify whether your site is focussing more on the keywords that don’t bring the best ROI, or identify that one particular product type has a lower conversion rate than others.
This way of keyword analysis suddenly gives you answers to business questions such as ‘what should we stock more of’, ‘what is our most successful product group’ and ‘what do we need to be working on in order to make more money’.
The analysis can lead to marketing plans, product ordering decisions and website promotion ideas. You can identify which products to feature on the Home page, which to push more through SEO, which aren’t worth the SEO effort and which ones you’re not competitive enough on.
As an SEO, you may think that some of this is the remit of someone else, but if the company can’t make a good ROI on certain products, how long do you think it can keep paying your SEO bill? Knowing the products that are most successful within the business can be very beneficial to planning your SEO campaigns, it gives you the most valuable targets and a higher chance of being able to show successful ROI for your work.
Let’s have a look at a different report now – Content Reports, the landscape of your user’s journey. An important area for web designers, conversion optimisers and people promoting the pages through marketing campaigns.
Find out which pages are considered to be of the most important. Also find out which tasks on the site are seen as the most important. For starters, combine both of these with the data to see if anything matches.
If the important site tasks don’t tie in with the important pages or the data shows that these do not see the expected level of pageviews then you have work to do to mould the landscape in to one that suits the users and the business plans.
Take each important task of the site and identify who the users should be, who the users actually are, whether the user journey is the expected one, what problems might be found within the journey and try completing the task yourself from different user points of view. Once you’ve done this you will be able to plan how to get the right users and how to make the journey as successful as possible. If you have the right users coming into a smooth journey you’re well on the way to a successful website!
A twist in the plot!
Analysts will need to always be on the lookout for unexpected changes – be prepared to have to explain why data has changed, not gone to plan or seen an unexpected success. Again, this can be done through segmentation – which pages, which users, which keywords etc, then apply those that changed to other data to build the full story and identify the actions to take.
Combining the data throughout is a very important aspect of data analysis, 8,345 visits will not tell you anything about the actual success. How many of these were valuable? How much did it cost to get them? How many completed the right tasks? What types of users were attracted? You don’t know any of this information unless you use more than one number. Try to encourage others within your business not to focus on single numbers alone, but aim for a real life conclusion.
Conclusion (or will it be a cliff-hanger?)
Data analysis is a never ending story, every day will see different activity, different results and frequent twists and cliff hangers. You may reach your desired objective but would you want to conclude the project there? Try updating your aims, goals and objectives regularly, always looking for new challenges and ways in which the success can be improved through different methods. Making your data real will really help with this.
It also makes it easier for those people who don’t like numbers to feel more comfortable in planning routes to success and understanding results. Additionally, it can help you get an understanding of the full picture, the business goals, where you fit in and what additional aspects of the story you can influence.
And then all the business owners, analysts, search marketers, web designers, web optimisers, inbound marketers, gurus, and ninjas all lived happily ever after…
The end.
Image credit: Follow the Yellow Brick Road by http://www.flickr.com/photos/pasukaru76/