In just a few hours you can improve your website usability, increase conversion rates, evaluate your information architecture, or enhance click-through rates, by following these six simple steps.
Before enhancing the positive areas of your page (where you want people to click), lets first eliminate "artificial links" - elements on your page which visitors are frustratingly clicking because they look clickable, but aren't.
View the Click Map report in clickdensity, which displays "link" clicks in green (i.e. clicks on hyperlinks or other clickable targets), and "non-link" clicks in red. Look for concentrations of red clicks on individual elements - this is usually a sign that a number of people are mistakenly assuming it is clickable. Update the style/design of any such elements accordingly, to reduce your visitors' frustration.
For the next step, we want to maximise and prioritise the visibility of "positive" links on your page (those you want the visitor to click).
View the Heat Map report in clickdensity; this will show exactly where visitors are (and aren’t) clicking, with the most popular regions in "hotter" colours.
Examine your main navigation devices (top menus, side menus, footers), and check the usage patterns. If there are items that are rarely clicked, these could be good candidates for re-wording (do visitors know what they mean?), re-prioritizing (moving down the menu), or possibly re-locating altogether (into a sub-section or elsewhere). Conversely, popular links in sub- navigation devices (e.g. footers) are good candidates for prioritization (moving into more obvious locations or menus), or this could be an indicator that the content from these popular sub-sections should be made available on the current page.
Other (non-menu) 'lists' on your page should undergo similar analysis. For example, if you're displaying the latest five blog/news items, but only the most recent two or three are ever clicked (possibly because you have a high percentage of returning visitors), then this is a candidate for contraction (to fewer items), freeing-up the space for more valuable content.
The Heat Map report will also highlight other patterns of user behavior, specific to your website. For example, the majority of visitors may click the thumbnails of story links rather than the text links, or there may be particular areas of the page that are busier (and potentially more 'visible') than other areas.
We can now attempt to identify and reduce confusion and hesitation, using click-time data.
View the Hover Map report in clickdensity; this will show usage data for individual elements. Hover your mouse pointer over the most important/popular elements on the page. A 'typical' click-time distribution is seen in the image below; peaking at around 1 or 2 seconds, then quickly falling off, with a long-tail.
If your important/popular elements are displaying a flatter/more varied shape, such as that shown below (which was for the most important element on the page), this could suggest that visitors are not easily discovering the link/image.
These important elements could be made more visible, by prioritizing their location on the page, increasing their contrast (in a generic sense, with respect to their surrounding content), or otherwise re-styling or re-labelling.
A complementary method for identifying usage change over time is to switch back to the Heat Map report, and use the Click Time filter to produce two different reports; one for 'faster' clicks (e.g. < 10 seconds), and one for 'slower' clicks (e.g. 31-60 seconds). These will enable you to spot more generic trends in usage. For example, if your Search facility and Site Map link are used more heavily in the slower (31-60) report, this could imply that a significant number of your visitors are unable to find the information they require on the page, and are resorting to the Site Map/Search after first scanning the page for the content or navigational sign-posting.
In the last step, we introduced the idea of identifying problems by comparing patterns of user behavior from two different groups (those who clicked quickly, and those who clicked slowly).
We can gain further valuable insight by continuing with this approach - using the clickdensity report filters to compare segments of visitors (those that share similar attributes).
In the Heat Map reports, click the Show Advanced checkbox, to view all the filtering options. Try first filtering by Screen Size, to check if any of your important elements further down the page are missed by visitors with smaller resolutions.
Next, enter one of your Goal Page URLs into the Went To box, and update the report (a 'goal page' URL will be a page that you want users to visit, e.g. a purchase confirmation page, or a sign-up form). This will only show clicks from users who - at some point in their session - visit this goal page, and will therefore help you analyze exactly which "routes" (links, menu items, adverts) are and aren't working towards this goal.
You may also want to try segmenting by date; we often find different modes of use between weekday visitors and weekend visitors (for example, as identified in this research on museum websites), which may persuade you to re-distribute Pay-Per-Click advertising campaigns (e.g. away from weekends, towards mid-week, or vice-versa).
6. Split Test
Some of the previous steps will have raised questions and hypotheses regarding your current page, with no clear solution or answer. Luckily, there is a simple way to test possible solutions.
Conceive and create a potential fix for one of the problem areas on your page (e.g. a new graphic for an advert that isn't performing, or a re-worded menu item). Then, visit the main administration page in clickdensity, click the a/b tests button, and add a test.
In clickdensity, A/B Tests work by replacing an existing element on your page (Version A) with a different version (Version B), for 50% of your visitors. Ensure that the element you want to replace (whether it is an <img>, <a>, <div> or something else) is assigned an id attribute, then use the form to specify the HTML for the new version (version B), and the number of times you want the test to run. You can preview the two versions (original A and new B), to check that the code is correct.
Once the tests have run for a number of your visitors, you can use the A/B Test Reports to check if click-through rates, goal page visits, or some other pre-defined metric has changed positively. In the example below, we can see that the new version under test (Version B) produces almost twice as many click-throughs, the majority of which then go on to visit the Goal Page.
We've presented a series of six simple steps that anyone can perform to evaluate and improve their website. These steps can be used on all your key pages, and can be iterated, to continually refine your site.