With a growing number of ambiguous ‘solutions’ made available by web consultants, it can be difficult for a website owner – who can see that her website isn’t achieving it’s potential, but doesn’t know why – to identify the core issues and find relevant help.
At Contentini we’re fans of data, evidence and an analytical approach. Although problem solving often requires breadth of knowledge and depth of experience, we’ve just started to investigate how different types of website issues could be automatically identified be examining standard web analytics data.
These are our first published thoughts on the topic, which we hope you’ll add your comments to and help progress to the next iteration.
Information Architecture Problem
Information Architecture (or IA) defines the structure and relationship of your website content and pages. You have an IA problem if your visitors can’t easily find what they’re looking for.
- High ‘pages per visitor’ combined with low ‘average time per page’ or low ‘time on site’
- High bounce rate, especially on the home page
- Re-label navigation items
- Re-design navigation devices (menus, breadcrumbs, search)
- Visually prioritise relevant content on each page (re-structure the pages)
User Experience Problem
User Experience (or UX) describes how your visitors feel about using your website, based on their expectations and subsequent interactions. You have a UX problem if visitors aren’t completing tasks on your website (e.g. checkout, register) or aren’t returning to use your website as frequently as expected.
- Low number of goals/tasks completed
- Low customer/visitor loyalty (repeat visitors)
- High number of abandoned (started but incomplete) tasks
- Re-assess and research user needs, priorities and abilities
- Re-design functional elements and workflows
Content Strategy Problem
Content Strategy defines how your website content is planned, delivered and maintained. You have a Content Strategy problem if your content isn’t actively supporting and positively contributing to your business objectives by increasing sales, attracting visitors, decreasing support costs, etc.
- High ratio of content page views to goals completed (e.g. purchase, registration, enquiry email)
- High bounce rate (especially off content pages)
- Low social media sharing of content (potentially, depending on type of content)
- High number of irrelevant or similar support emails, telephone calls, etc.
- Content gap analysis (what content do your visitors need that you don’t provide?)
- Content strategy development (how can you use and manage content in the long-term to support your goals?)
- Editorial strategy and guidelines (what language to use, enforce consistency, etc.)
Search Engine Optimization Problem
Search Engine Optimization (or SEO) describes the process of shaping your website code and content to increase it’s ranking in search engine results pages for potential customers who search for relevant terms. You have an SEO problem if your website attracts a relatively small percentage of it’s traffic from search engines.
- Low percentage of relevant search engine referrers
- Keyword research (find keywords that are relevant, high-intent, low competition and high-volume)
- On-page optimization (changes to HTML and content)
- Link-building strategy (link-bait, guest blog posts, etc.)
There are of course many other potential problems – technology platform (security, performance, efficiency), marketing, workflow, business model – but for the purpose of this post we’re sticking to the ‘front end’ issues.
Content Strategy, UX, IA and SEO should ideally be considered prior to launch, rather than reactively; even so, as a business and customer-base evolves, these issues are likely to need constant analysis and revision.
If we can use quantitative data to automatically diagnose problems as they occur, this brings us closer to the possibility of a self-healing system. For example, an IA issue might be detected in a particular website menu; the system could automatically A/B test a number of revisions to the menu to automatically resolve the problem and optimize the website. It seems far-fetched, but if the biotech community is researching nanobots that can perform a similar task for our bodies, then why not a self-optimizing website?