If there were one, surefire way to create a great website, UX design would be infinitely simpler and everyone would be happier (except maybe UX designers). But even though there is no magical solution, years of testing and user feedback have given designers a good idea of which approaches works, which don’t, and which lead to better websites.
So instead of one magic bullet, here are seven proven approaches that we have found to be critical to improving web results.
1. Humility: You can always improve a website
The first approach – and our starting point – is as humble as it is short: a website can always be improved. This is true whether you haven’t changed a thing since your site first went live or if you’ve just completed a comprehensive redesign yesterday.
If you’re wondering what might need improvement, or how to implement changes, then do check out our next six approaches.
2. Curiosity: It pays to know your users
Whether you’re designing a website for a third-party, the company you work for, or yourself, you’re ultimately creating it for your users. After all, they’re the ones you want to engage with it.
Because your users are central to your website’s success and your business, it’s crucial to get to know them as much as possible. To create effective designs you need to have an overview of who you are designing for and be able to answer the following:
- Who are your users? What groups of users exist?
- How do the different groups interact with your product? What do they want? What do they care about? What are their goals?
- In what contexts do they use your products, services and communication materials?
Answering these basic questions will help you create products, services, sales pitches, and websites that are tailored to your users rather than to yourself. It will help you combat your assumptions and allow you to move beyond best practices. Down the road, this knowledge will also help you refine your design and serve as a guideline for your decision-making process.
But don’t worry – if you can’t answer these questions, it’s never too late to do design research. And it is not necessarily hard to do – you just need to engage in conversations with your users: ask them questions, run feedback campaigns, and consider the ways in which they interact with your website. There are three methods we’ve found to be reliable and useful:
- Conducting interviews
- Ethnographic field studies
- Feedback campaigns
Keep in mind, as recommend in this article by the Nielsen Norman Group, that user research will ultimately depend on your goals and solutions available to you.
3. Think strategically: Understand your website’s role in your company’s sales and marketing strategy
No matter if your company is just you and your computer, or if you work at a multi-national company with a thousand employees, you need to make your website reflect the company goals.
While a site’s “funnels” and “conversions” certainly do matter, before you start tweaking to optimize such things you need to make sure you’ve got a clear idea of just how your site fits into your organization’s overall sales and marketing goals and strategies. This sounds like a no-brainer, and in a perfect world maybe it would be. In our world, a lot of companies haven’t clearly articulated their sales and marketing strategies – let alone how their web strategies ought to dovetail with these – and their web results suffer accordingly.
Consider your customers’ journey: just how do potential customers who have never heard of your company end up paying you real money for your products or services? What steps do they take on their way? What are the triggers that help move them from one step to another – and the barriers that prevent them from moving along? How does the website, along with other content types, people, and processes facilitate the journey towards purchase and satisfied customers?
Based on the answers you come up with, you’ll determine a set of “core” pages around which the rest of your website is built. For example, if you aim to sell a product, every step should lead users towards that product. When you have a clear idea of your customers’ journey in mind, web design is no longer only about individual pages but is focused on a flow of pages and interactions towards something.
Not all pages have to be part of that flow, of course: consider informational pages (e.g. “Who we are”, “FAQ”, “Contact”) which provide additional or contextual elements. These pages should nonetheless fit your site and company’s overall strategy. Rather than appearing as disjointed parts, they should be homogenous with the site’s overall look and feel. And they shouldn’t create a “break” in the flows you’ve created.
You should also rely on analytics to understand what users determine to be your core pages. Are the most popular pages the ones your strategy says they should be? Do these pages help with your general strategy? Are they also your core pages? Do they lead to conversions and if so, how?
Analytics will tell you, at a glance, just how well your site is supporting your company’s overall sales and marketing strategies. Highly popular pages that are outside of the flow are problematic, and so are core pages with low traffic. Consider what successful pages do differently from the problematic ones: is their design different? Are certain areas obscured or difficult to decipher? Do they break your flow’s continuity in any way? In short, think holistically: a well-implemented call-to-action (CTA) is important, but if your users never get to your (and their) end-goal, then the CTA should not be your main concern.
Beyond the hard numbers, what works and what doesn’t will vary from business to business, and from industry to industry. You’re the expert and the tools at your disposal should help you resolve this issue. If they don’t, then it’s time to turn to your users for some answers and engage in more research and user testing. Turn the questions around: what do they believe is your company strategy? How do they interact with your website? What are their journeys?
4. Quantify: Measure everything
Here at Mouseflow, we are big proponents of data-driven design. As one of my digital mentors, James Clear, says: “You can only improve what you measure.”
This is, of course, related to “section 2 – It pays to know your user” but analytics can lend an extra dimension to this. Broad analytics (e.g., page views) are a great starting point, but you want to gather much more data as you go along. When we think of having data at our disposal (and at yours, too!), we think about:
- Funnels and user journeys: A combination of Google Analytics and Session Replays will let you see what your users’ journeys look like. These should be curated from end-to-end. If they repeatedly take unexpected steps or leave the flow you had in mind, it’s time to fix it.
- Heatmaps: There are so many benefits to heatmaps that we dedicated a blog post to show how tremendously helpful they can be.
- “Big picture” statistics: It can be helpful to have an idea of your visitor numbers over several months or years. Such longitudinal views may suggest that your audience has changed, that how they browse has changed, that your website has continually improved or gotten worse in meeting goals – and more.
- User feedback: You may think you know your users, but don’t see it as a finite process. You need to be involved in an ongoing conversation with them. Just as you should have statistics and analytics that stretch out over months or years, you should know what your users were happy about or frustrated with six months or a year ago. How many tickets or complaints were submitted? How many forms were filled? What features were most requested?
- The more “technical” parts of the user experience: How long do your pages take to load? Are they throwing JavaScript errors your (or your users’) way? How long do your users stay engaged with a given page, or what do your bounce rates look like?
A data-driven design will help you continually improve your site by highlighting performance numbers, allowing you to measure effects of improvements and stepping beyond best practices by relying on numbers specific for your audience to tailor the user experience. Every industry, vertical, and business is unique. When you follow design rules and guidelines to the letter, or just follow the latest digital design trends, it leads to a cookie-cutter website that doesn’t resonate with your users.
5. Speed up: Slow loading times should be optimized, not ignored
We’ve all experienced it. We want to quickly check out a website, but the page takes ages to load. This may be due to poor optimization or a variety of other factors, but sometimes the impact is drastic enough that we’ll simply close the tab and either get the information we were looking for somewhere else or move on to the next thing on our list of things to do.
The correlation between bounce rate and loading times has been proven repeatedly: if a website takes more than four seconds to load, a noticeable number of users will simply leave. The reaction may not always be as extreme: users may follow your designed flow, but not without frustration—a negative feeling you don’t want to be associated with your company or what it strives to offer.
Thankfully, page optimization may sound technical, but it can be achieved with very little technical knowhow—and without knowing the first thing about coding.
To test the loading time of your pages we recommend that you either use Googles very convenient page-speed testing tool right here or have a look at the “render times” column in Mouseflows Heatmap list.
Once you have identified pages that load slowly, it’s time to dive a bit deeper. We usually start any optimization project by optimizing the Image files. Because they are often the largest elements on a page, they are likely to have the biggest potential for optimization. I will even go as far as to say that “the single most effective action you can take to optimize your pages is to make sure your images (both visually and in terms of file size) are optimized for your site”.
If you have not done this before and don’t know the first thing about image optimization or where to begin, we recommend this handy guide.
There are of course plenty of other ways to optimize loading time, but we will have to do a more in-depth description of these in a later blog post.
6. Follow the rules of good design: Optimize the usability of your site
Agreeing on what constitutes “good design” can be tricky, but most designers will agree: a good design, at its core, is one that gets out of the way and lets the users complete their own tasks quickly and efficiently.
Rather than coming up with our own design guidelines, we defer to Nielsen and Norman’s “10 Usability Heuristics for User Interface Design”. These fairly broad rules have been around for over two decades, and have proven to be reliable and applicable across the board.
Some parts may not work for your specific website or company, but they should be considered nonetheless.
7. Follow up: Don’t assume that a feature is “done” because it has been implemented
New features, like engagement with your users, are never really done. Designs that are great in theory may prove to be flawed once implemented. Or, they may indeed be technically great yet end up frustrating your users. This is why implementation shouldn’t be seen as the finish line but as a new conversational topic.
Once implemented, a redesign or new feature should always be tested in the real world. Internal testing will only get you so far. A/B and multivariable testing will give you the feedback you’d be unlikely to have obtained in-house.
At Mouseflow, we use the Lean UX framework quite a bit. This allows us to build Minimum Viable Products (MVPs) and implement them, then gather feedback and measure performance through an iterative process—continually improving our features as we go along.
We’ve actually covered quite a few of the core aspects of the Lean UX methodology earlier in this blog, because it simply makes sense. Some concrete starting steps:
- Collaborative research and assumptions: Gather information about your users, problems to solve, urgency, and what your solutions could offer.
- Come up with hypotheses: Once you have assumptions, it’s time to test them. This process is also collaborative and involves spelling out the issues at hand, why the feature offered in response matters, to whom, and what it will achieve.
- MVPs: Building a basic version of the concept, based on the “best” (up until that point) hypothesis.
- User testing and additional research: Now that you’ve implemented an MVP, it’s time to see if your hypothesis holds in the real world. Keep gathering data and feedback, then return to the drawing board to develop the next set of improvements/features.
Of course, this is a very basic version of what the Lean UX framework involves, and we highly encourage you to read the book Lean UX by Jeff Gothelf and Josh Seiden.
In our experience, each of the seven approaches we’ve outlined above works well on its own. None is a silver bullet, though, and results are far better when they are all applied together.
Similarly, it’s great that a web designer adheres to these basic approaches: this will lead to better results. But these approaches really prove their worth when they are adopted more broadly throughout the organization. It is when those responsible for marketing, sales, product management – and web design – all apply these approaches that real magic can happen.
