Data strategy for web analysis - part 2

Choosing the best tools for your data strategy

Analytics

Analytics is probably the first tool you'll think of when you think of website analysis. It will give you a lot of information about your website, but it's still a tool with limitations. Let's look at what analytics data is, it's pros and cons, and how you can use analytics data.

What data the tool collects

Analytics data is basically user interaction data. The tool tracks your visitors and gather information about how they interact with the website. The first page they viewed, how long they stayed, how many interactions, if they used the internal search function, and much more. There is a lot of information available automatically when using an analytics tool to collect data, but you can use it to collect even more data. For example, you can track campaigns, set up events to track specific actions (e.g., clicking on a specific button), and you can set up goals to track conversion on the website.

What the data does and doesn't tell you

Analytics data will tell you how your visitors use your website. You can see if they viewed a page or not. If they clicked on a button or not. What keywords they searched for. It will also give you some information about how the visitors found your website and if they have visited the website before (which is limited by cookies and when you started collecting data).

However, analytics data will not give you any information about intentions, experiences, and emotions. You will not know if the page they viewed was the page they wanted to view. You will not know if they liked the content. You will not know if they left feeling happy or frustrated. You will not know why they didn't click on a button or convert. 

It's important to take the limitations into consideration as you make your analysis, or you may draw incorrect conclusions from your data.

How you can use this data

Analytics can give you a lot of good information, such as what content the visitors interact with the most, how the visitors navigate, where they tend to lose interest and leave the website, which campaigns are most successful, how many visitors return to your website, and what keywords the visitors search for. Analytics will help you track and evaluate KPIs, conversions, engagement, marketing strategies, the internal search engine, and much more. 

Heatmaps

Heatmaps is a tool that visually shows how the visitors interact with the website. In that way, it shares similarities with analytics, but has it's own limitations too. Let's look at what data you get from heatmaps, it's pros and cons, and how you can use the data.

What data the tools collect

Like analytics, heatmaps track website interactions, but the data is presented very differently. Instead of giving number of clicks and views, heatmaps show that data as hot areas on a screenshot. Heatmaps can show you where the visitors clicked on the page, how the cursor moved, and how much of the page the visitors views. 

What the data does and doesn't tell you

Heatmaps will also tell you how the visitors use the website. You can see if they clicked on different buttons and if they tried to clicked where there aren't any buttons. You can see how much of the page the visitors tend to read. You can also get an understanding of how the visitors engage with the page. A lot of this information can be collected with an analytics tool too, but the visual aid of the interaction clusters on top of a screenshot can make it easier to understand what the data is trying to tell you. 

As for analytics, heatmaps will not give you any information about why the visitors interacted with the page the way they did. You will not know why they didn't click on a button, or mistook another element for a button. You will also not know why the didn't scroll through the entire page. However, as you can easily combine clicks, scrolls, and placement of different elements on the website, it's easier to make fairly accurate assumptions with heatmaps.

How you can use this data

Heatmaps can give you a lot of good information about individual pages, such as which content is most engaging, which design works better, and where you should place important information and elements. Heatmaps will help you to evaluate content, engagement, design, layout, and navigation. 

Session recordings

Session recordings are another visual tool, but instead of summarizing data from many visits, it records individual visits. Let's look at what data you get from session recordings, it's pros and cons, and how you can use the data.

What data the tools collect

Session recordings are just what they sound like, recordings of individual sessions. However, you can tweak when and where you want the recordings to start. From the recordings, you can see the cursor movements and follow along during the entire visit. You can also get some addition information about each recording, such as the length of the visitor and number of pages viewed.

What the data does and doesn't tell you

Session recordings also tell you how the visitors use the website, such as which elements draw their attention, where they click, how they prefer to navigate, how far they scrolled on each page, and if they seemed to read the content or if they scrolled fast, seemingly looking for something else. Again, a lot of the data collected from session recordings can also be collected with analytics, but session recordings give a better understanding of the behavior of individual visitors.

Session recordings will also give context to the data. For example, the analytics can tell you that they used the internal search function, but the screen recording of that specific visit can tell you that they didn't scroll down far enough to find the link to that information. And they therefore chose to use the search function. Or you find that the visitors don't stop and read the landing pages, even though many views looks good in your analytics. 

However, as before, session recordings can not tell you why the visitors behaved the way they did. You will know that they didn't scroll down far enough to notice the link, but you can only guess as to why they didn't keep scrolling. You will know that they preferred the search function over the main menu, but you will not know if that is always their preferred method or if there is a problem with your menu. But as for heatmaps, the recordings provide some context to the interactions and help you make some educated guesses about the reasons behind the behavior. 

How you can use this data

Screen recordings can give you a lot of information about how the visitors interact with the website. Where they click, how they move, how they prefer to navigate, if they follow your intended structure, how far they scroll, and more. Like heatmaps, screen recordings can be used to evaluate content, engagement, design, layout, and navigation. However, screen recordings give even more context to the behavior of the visitor.

Web surveys

Another method for collecting information about your website is to use web surveys. These are surveys that either pop up on your website, are triggered by a button somewhere on your website, or are embedded on some or all of your website pages. A survey is what you make it, but let's look at some of the data you can collect from a survey tool, it's pros and cons, and how you can use the data.

What data the tools collect

Unlike the previous tools, with surveys, you decide what data you collect. You can ask about anything in a survey, meaning that you can get information about the interactions from surveys too. However, if you want to know about interactions, surveys is not the recommended tool. Instead, surveys are better for asking about experiences and opinions, such as their general impression of the website, how easy it was to navigate, how they perceive the brand, and more.

Besides decided what data to collect, you also decide if you want quantitative data (closed-ended questions with options) or qualitative data (open-ended questions where the respondents type their answers), or a mix of both. Surveys is the only qualitative website analysis tool.

What the data does and doesn't tell you

Web surveys are great for asking about the experience on the website. You can learn how they found your website (e.g., recommendation, knew about you despite first visit, ad in newspaper, and so on), how often they return, if they found what they were looking for, if they encountered any issues, their opinions about specific content, how accurate they found the search results, and much more. In many ways, surveys can be used to fill in the gaps from other tools. While analytics can tell you if the visitor viewed a specific page, surveys can tell you if that was the page the visitor was looking for. While analytics can tell you where the visitors left the website, surveys can tell you if they left happy or frustrated. While heatmaps can tell you how far down they visitors generally scroll, surveys can tell you if they needed any of the information further down on the page.

Surveys also allow you to target specific visitors. If you're only interested in the visitors on specific pages, you only add the survey to those pages. If you're only interested in visitors with a specific purpose, you can use the survey to identify those visitors and only let them answer the remaining questions. 

Although surveys are flexible, they too have limitations. For example, you will never get a 100% response rate. For a pop-up survey, you can expect a response rate of about 1-12% - but that is only from the visitors still on the website when the survey is triggered. However, you don't need every visitor to answer a survey for it to be representative of the visitors at large. Sometimes, twice the response rate doesn't change the result, because the smaller response rate was still representative.

You also need to consider that the data you get is not completely accurate. Some respondents are annoyed at having to answer a survey (even when it is optional) and may just answer anything to be done. Some misunderstand a question, and therefore answer erroneously. Some remember incorrectly, and also answer erroneously. And some do not feel comfortable with the question and answer with a lie. This doesn't mean that survey data is useless, but you need to be aware of this. You can also reduce these possible errors with a good survey. 

How you can use this data

As you can use surveys in a variety of ways, they can be used for a general analysis of the website, but also to analyze specific pages or specific target groups. Surveys can help you evaluate why visitors come to your website and how that matches your content, how successful they are navigating your website, identify problem areas on the website, how the visitors feel about your company and/or products, how likely the visitors are to return, and much more.

You can also use surveys to track and evaluate KPIs, but unlike analytics, surveys can also give you more information about why the KPI changed or didn't change.

Additional tools

Beside the tools that collect data about your visitors, there is more data you can collect about your website. For example, you may want an SEO tool to improve the website's ability to be found in search engines. Or you may want a tool for web performance, such as loading time, crashes, and broken links. 

Combining tools

As you will have noticed, there is some overlap in the information you can gain from the different types of website analysis tools, but they all have their own strengths and weaknesses. Each can be used on their own, but combined, you will be able to make a more complete website analysis. If you aim to combine tools, you should ideally use a service that supports all the tools you want to use, and Extellio offers the four main ones. By connecting the different tools, you are able to use information from one tool in your analysis of the data from another. For example, if you found that visitors struggle to find content that is available on your website, you can use survey data to create a segment for visitors looking for that content, and then use that segment as you look at the analytics data and see which pages they viewed instead, or as you look at a screen recording to see how they navigated. 

Selecting the best tool

Although the are insights to be gained from all the different tools, and even more if you combine them, you shouldn't select tools based on what data you can collect but rather what data you need to collect. Anchor your decision in your data strategy and then see which tool or tools match the needs you have identified. As your data strategy evolves, so may the tools you use. You can learn more about data strategy here.

Another thing to consider when selecting web analysis tools is the rules and regulations you need to follow. You can read more about compliance here.

Related articles

Data Strategy for Web Analysis - Part 1: Crafting a strategy 

Data Strategy for Web Analysis - Part 3: Keeping it compliant