A Beginner’s Guide to the UX Audit

Get Design Feedback | Design Articles

Imagine running an eCommerce site. You know visitors will find you in search engines and will interact with your homepage. They even get your checkout process started. But they don’t convert, at some point. And you don’t even know why. It may be time to update the Hierarchy of Information. Or flows through the user. But how do you know what needs to be rejigged, and what not?

A User Experience Audit (UX Audit) is a way to identify areas of a digital product that are less than perfect, revealing which parts of a site or app cause headaches for users and stymie conversions. As with financial audits, a UX audit uses empirical methods to expand an existing situation and offer user-centric enhancements, in this case, heuristics-based recommendations for improvement. Ultimately, a UX audit should let you know how to boost conversions by making it easier for users on your site or software to attain their goals.

The purpose of this beginners’ UX audit guide is to equip teams with the basics for conducting their audit, or to better understand the benefits and limitations of an external audit.

What’s going to happen in a UX audit?

The big questions first of all. What exactly happens during a UX audit and how does the usability testing fit in? An auditor will use a variety of methods, tools and metrics during a UX audit to analyze where a product is wrong (or wrong):

Review of company and user targets
Conversion criteria
Data on Customer service
Sales statistics
Traffic / Dedication
Compliance with UX Norms
Heuristics in Usability
Mindsetting
Wireframed & Prototyped
Best UX Practices

The difference between usability testing and a UX audit is one of the direction of information flow: an audit creates issues from a set of pre-established standards or goals, while testing creates problems from user actions. If an auditor does not have access to the fundamental metrics, an auditor can use usability testing during an audit, but must combine the findings with data collected over the longer term and measure them against industry standards and product objectives.

What can you tell a UX audit, and what are its limitations?

It is important to remember that a UX audit is not a panacea for all UX woes on a platform. This is unsuccessful if guidelines are unworkable or unfolded. It also requires a substantial expenditure of time and energy, to the detriment (or at least delay) of certain activities as the audit is carried out by the internal team.

However, while a UX audit can not solve all of an ailing site or application ‘s problems, it can be used to answer some deep questions:

What works, and what doesn’t?
What metrics are being collected, and what should be collected?
What is the information telling you about user needs?
What has been done before, and what effect has it had on metrics?

An efficiently conducted UX audit gives a product many benefits. This includes actionable, objective evidence-based follow-up practices and not hunches. It supports strategic planning. It produces metrics which can be used in tweaks to come. And it helps to shape theories about why users are behaving in a certain way, and how they will be behaving in the future. Most notable of all, it helps boost conversions and ROI once follow-up action is taken.

Who will do an audit of the UX, and when?

Tim Broadwater, writing on LibUX, sets a strong thumb rule when you may want to perform a UX audit: ‘(an audit) should be performed in the very early stages of a website, web application, dedicated app, or similar redesign project.’ The word ‘redesign’ is important here; audits are typically performed on a product or service that has been live for some time with data backlog to review. With usability testing, new apps and products are more likely to be put through their paces instead of a holistic audit.

Generally speaking, companies without a dedicated UX team are most likely to benefit from a UX audit; those with an in-house team are most likely to evaluate the product and continually adjust the experience.

If cash flow allows it is best to have the audit carried out by external parties: it is difficult for internal teams to separate themselves from the product, and latent biases can hamper the process. Nate Sonnenberg gives a helpful overview of how much it costs to call in the auditors: upwards of $1000 for a few days with a one-person team; full amount of a UX team coming in for four weeks and providing in-depth, goal-oriented insights might cost up to $ 10,000. But, according to Nate, 80 per cent of issues will be noticed in 2-3 weeks, which is enough to get going.

All is not lost, however, if the budget does not apply to an external audit: you can audit the product internally by following an unbiased process, using the broad range of available resources and (if not already) being au fait with UX best practices and standards.

Let us take a look at what you need to get your UX audit underway.

Which is it that you need to get a UX audit done?

You’ll want a cross-section of the team – designers, developers, product strategists, and business managers to be involved. It also helps to appoint an audit chief, who can take procedure and time frame decisions.

As with any other project from the get-go, the following must also be agreed upon:

Audit objectives (conversion, ROI, etc.)

A time limit which is critical because you could audit forever in theory

How many resources do you wish to devote to the audit: time, workforce, money

An overview of the process

It’s time to adumbrate the method, once you’ve set the fundamentals. Starting from a birds-eye perspective, the UX audit consists of six main stages: compilation of metrics and materials; confirmation of results; data organization; analysis of patterns and trends; reporting of findings; development of recommendations backed by evidence.

Methods and Materials Inventory

Perhaps the hardest part of a UX audit is the first step, the collection of the relevant materials. If goals were properly defined before you embark on the audit, you would know what kind of information you need; now you just have to think about which metrics will give you that information.

Get team members on board in sharing information and tracking useful metrics you currently don’t have, and this step will be easier.

Here are some metrics and content sources that are helpful in an audit:

A heuristic product evaluation: Perform a cognitive product walkthrough to see items from the viewpoint of a consumer. Take notes when trying to reach user goals, and focus on identifying potential obstacles. Be mindful that your product awareness will make this job challenging – it will help you remain focused by basing the method on proven criteria such as Nielsen’s heuristics.

Website and mobile analytics: If you get qualitative data from a heuristic evaluation, analytics tools will provide you with the necessary quantitative information. Most people should be familiar with Google Analytics’ core functions, such as traffic source, traffic flows and trends over time; more advanced functions can elucidate website user flows, conversion (and abandonment) hotspots, and what users do before and after visiting your site. Tools such as Kissmetrics and Crazy Egg can add features such as heat maps and churn rates to basic analytics; app analytics can be collected either through Google Mobile Analytics, or through a dedicated tool such as Mixpanel. Make sure that in analytics you go far enough back to understand patterns, rather than focus the analysis on individual data points.

Conversion rates or sales figures: If eCommerce is the premise of your website or app, UX audit sales or download figures may be useful. For example, we calculate how many blog readers access our prototyping tool here at Justinmind and from which specific posts: this gives us an insight into how our content fits in with Justinmind ‘s wider user experience and how we are meeting user pain points.

Interviews with stakeholders or consumer surveys: You have to get out there and talk to real people as in every UX project. Start by interviewing internal product stakeholders such as product owners and developers, asking them for insights into the strategy, requirements and ongoing challenges of production of the product. You may also ask what they want to see from the UX audit, which will create goodwill and confidence in the audit process. Also, find out if user surveys have ever been conducted by the marketing or sales department: there is likely to be a wealth of comments and feedback within these surveys you can use in a UX audit. You can organize this feedback into categories – findings, for example, per screen or task – and by severity.

Previous product requirements: Acquiring access to the original requirements of an application will save you time and help you understand why design decisions have occurred the way they did; this information will be useful when writing viable recommendations.

At this stage, the qualitative data collected via usability tests can be paused and validated. For instance, if past user surveys showed that the process of customer check-out was complicated, conduct usability tests to see whether you can back up this unfounded argument.

How to arrange the collected materials

Spreadsheets, in a phrase. All of the information collected in phase one should be tracked and aggregated on a worksheet. Import the spreadsheet into the cloud and make it a live, shared document where questions and suggestions alongside relevant metrics are documented.

If you’re unsure what to bring into the stack, try these helpful templates:

UserFocus offers an Excel booklet, in which you can calculate metrics alongside nearly 250 best usability practices
Usability.gov provides a Word template for measurable goals.

Look for trends and trends

The moment you have to turn data into insights is often nervous: turning metrics into meaningful change is a profound issue beyond this article’s remit. Suffice it to say that there are methods to help you make sense of the information before you, such as data mining, card sorting (not only for UX architects but also perfect for aggregating any mound of information) and insight incubation. Check out Steve Baty’s UXmatters post to find patterns in UX research.

Findings

After mining data for insights, it’s time to develop hypotheses about the user experience status of the application: why do users act as they do instead of in the way stakeholders want them. You can compare your insights against four keystones of successful products:

Relevance: is the site or app addressing user pain? When users find your product, is there a disconnection between expectation and reality?

Value proposition: is the user ‘s value clear and convincing?

Usability: are the user design points of confusion or misunderstanding, or do consumers intuitively understand what to do?

Action: Are calls for action noticeable and appropriate, motivating users to take action?

Evidence-supported recommendations

Finally, data-driven UX enhancement recommendations can be written. The key is making recommendations as applicable as possible. We’re fans of Joseph Dumas, Rolf Molich, and Robin Jeffries’ recommendations in ‘Describing Usability Problems: are we sending the right message? ‘:

Emphasizes the positive
Express annoyance tactfully
Avoid useful jargon
Be as specific as possible

On a more substantive level, don’t forget to add examples to recommendations, rather than just identifying areas for general change. For example, in this sample UX audit report from Intechnic, recommendations include “forms and resources, rearrange by number of clicks” and “drop-down navigation, remove images.” Suggesting solutions for the design-development team will always be a more positive, effective tactic than merely criticizing where user experience has failed.

A user experience audit requires significant investment in time and human resources if done internally, and money if contracted to professional UX auditors; it should not be undertaken slightly. However, the benefits to an existing site or app are clear, particularly if sales are stagnant or slow-growing and users’ voice is not reflected in the process of product improvement. You can make significant, data-driven changes to an application by undertaking a UX audit and see upticks in user satisfaction and ROI.

Get Expert Design Feedback: Improve UX!