Great content that works is a real asset to any organisation. Every company these days is in the business of content creation, whether they like it or not. It may be limited to a single website or focused on an intranet; or it may be a global organisation, grappling with issues of governance and localisation around content creation. Whatever the scale is, testing this content is vitally important but often overlooked.
I spoke at a conference recently and cited the example of an international bank that felt it had met its responsibilities in communicating important information around money laundering to its staff. However, when we tested the content, it it turned out that over 80% of the staff didn’t understand what was written on the intranet site, and that the information was neither clear, nor easily accessible. The bank was exposed to massive fraud and financial risk. Content matters! And if you think it doesn’t, try not making it and then see if it matters.
Assuming that we all want to deliver a better customer experience using effective and engaging content, why wouldn’t we test it? It helps us to make better decisions about the content we are creating, it can inform information architecture and design and it helps to keep stakeholders aligned. But it does sometimes ruffle feathers - it doesn’t always produce the results that people want to hear.
Part of a cycle, not a process
Testing tells us what we’re actually achieving versus what we had in the original plan. It needs to be a constant, part of the creation cycle that will yield regular, actionable data. It needs to be a repeatable and scalable exercise.
However, testing is not the hard bit. Acting on the results is the hard bit. It will inevitably mean disrupting ways of working and changing people’s attitudes towards planning with data. Old production processes and a bit of testing won’t deliver the benefits that you’d like. You’ll most probably have to make some changes to your processes before you ever start a test.
Test for the things that matter
Content is there to do a job. Know what that job is and test for how well your content is doing that specific job. If your content strategy is aligned with your business strategy – and it should be – what are the business goals driving that content? Your test results should lead you back to knowing whether your content is serving that business goal.
Writing test cases
Now for the really practical bit. You’ll need to write up a series of test cases, each aiming to understand the accessibility and clarity of a specific piece of content. Set these out as tasks so that respondents need to find the answer to a very specific question. There must only be one right answer to each question – don’t leave any ambiguity. I have a set of basic rules that I always adhere to when writing tasks:
- No clues in the questions
- Use plain language and common speech
- Be concise
- Make tasks independent from each other
- Make sure tasks don't require confidential information
Here are some sample questions - as you can see, they are clear tasks that require a single, right answer.
- How many days holiday is a new, full-time employee entitled to in the UK office?
- Who should be the first person you contact if you have an accident while on site with a client?
- How many bottles of hair dye did the company sell in the UK in 2015?
Once you’re ready to test, you’ll want to choose your respondents carefully. Know the audience that you want to test against (who is the content for?) and ensure that they have no prior knowledge or bias.
If you’re testing a consumer site, you’ll want to recruit the right demographic as you would with any traditional focus group. Obviously, if you’re recruiting internally, make sure that you communicate properly with any volunteers, and ensure that they know that you are testing the site and its content, not them.
On the day, ensure that the environment is right for the tests. It sounds really obvious but do check your connectivity!
The best way to usefully record results is to use video. There’s no more powerful way of showing stakeholders where users are getting stuck or having trouble finding what they are looking for. You will be recording on-screen activity whilst, in the best-case scenario, an independent observer will sit to one side of the respondent, recording any significant comments or behaviours.
When presenting the results, make sure that you are not just presenting the data but making clear and actionable recommendations. If you are quoting participants, always respect their anonymity and quote them accurately.
Present the findings in a clear and efficient way ensuring that a management summary is provided for more senior stakeholders who may not have time to read all of the findings.
This might sound so obvious but it’s staggering how often, in the face of solid research, necessary change remains unimplemented due to development cycles, lack of resource or lack of stakeholder buy-in. If your content matters, test it and always act upon the findings. A live website is not an end in itself, it is just the beginning of the journey. Getting content right is part of an iterative cycle and testing should be as much a part of that cycle as writing the content itself.
If you’d like to learn more about content usability testing, or if you think your content could be working harder for you, please get in touch with me.
We're always looking for new faces to join the Cognifide family. Take a look at our available jobs.