April 9, 2019
How user tests fit into your project management process
There is a strong link between projects and user tests. A company that carries out projects - and not simply in the IT development field - must consider integrating user tests into its entire work process. On the other hand, a company that would carry out user tests to validate a concept, without managing the project thoroughly and methodically, would simply be getting ahead of itself.
Benefits of a Project Management Process
This raises the question of the relevance of a methodology to manage a company's projects. However, there are few cases in which the application of strict formalities proves to be more negative than positive. Usually, these are contexts in which reactivity brings a much greater benefit than planning and quality validation.
A common example is young startups that test both their business and their methods; or even quick prototyping projects that aim to develop a product in just a few days or in the time of a hackathon.
In any other case, when a team needs to produce foreseeable results, it improves its results by following an established process. By "foreseeable", we mean projects whose results must meet an expectation, whether in terms of quality, timeframe and/or cost (see The Quality, Cost, Timeframe Triangle).
Monitoring a process ensures that the work to be carried out has been thought through beforehand; that we are in an action and not in a reaction; that we do not forget any important step, and that the required validations are established.
Cut out and Conquer
This is the basics of project management. When we have to do work that is too important to be done by one person in a short period of time, we have to cut out, again and again.
Defining a process is to identify the different steps through which the projects go. A few key questions need to be asked at each of these steps:
Who should perform the step?
How should this work be done? What tools and best practice should be used?
What are the expected results, and how are they measured?
What time budget is allocated to it? After what period of time is profitability not guaranteed?
Is this step mandatory? Can it be done at the same time as another step?
Allocate or not, but Control Quality
Some people oppose agile methodologies and workflow implementation. This is a common mistake.
Agile methods are built on both repetitive and incremental developments. Simply put, it consists in making short and repeated advancements, where the customer checks at each repetition that we are consistent with his needs. This is to be contrasted with an approach that would lead to huge specifications, and then launching into a long development process, which in the end would not correspond - or would no longer correspond - to the client's needs.
But each repetition of an agile project should not be handled as a simple task involving only development. It is necessary to (re)examine the client's needs, possibly by taking up again the operational study steps, however short they may be.
In many cases, the difference is that an agile approach means not assigning projects to team members but letting them choose which projects they will work on as they complete their previous tasks. This way of working is in no way in conflict with the establishment of a process made up of steps. The assignment is then not made by a single person, but by a group of people.
On the other hand, whatever the methodology used, some steps of the process need to be monitored. Quality is a key factor in the success of projects, and it is important to set the expectations for it. This means clearly specifying what needs to be produced, what steps need to be monitored, who needs to monitor the result, and what indicators are monitored.
Project Process and User Tests
Why user tests are an essential part of your process
Sometimes user tests are done without any project management logic. This can happen when these tests are carried out by someone who is not part of the day-to-day project management process. But is this a good idea?
This occurs most of the time when a team needs to identify subjects that will later become development projects. We associate this step with prospecting, rather than tests that would be integrated into a project process. The concept is to first collect user feedback, which will then be analyzed and broken down into several projects - which in turn will be properly managed.
But even if we want to conduct such a forward-thinking campaign, we should keep in mind the following factors:
User tests cannot be improvised. They must be prepared upstream, to ensure that they are not just about anything and everything; and they must be analyzed downstream, so that they can pay off afterwards. At the least, a test campaign must be managed as a full-scale project.
A strictly prospective approach often indicates that we are moving flying in the dark, that we don't know where to begin, and that we feel tempted to leave the responsibility for this choice to the users. But remember what Henry Ford said, "If I had asked people what they wanted, they would have said faster horses." Instead, cut it out into smaller workshops, each of which is directly linked to an improvement project that will be managed as a full-scale project.
Ideally, a large, randomized study should be avoided. Tests will always be more effective if they are done with a specific focus, even if it means conducting several tests over the course of the project. The best being to integrate them into a project, in order to help define and/or validate it, while at the same time benefiting from project management in terms of schedule and resource monitoring.
Where do user tests fit into your process?
To put it simply, user tests can be integrated at two key moments in the course of a project: during the design phase and before implementation.
The design of a project (product, service or other) is a crucial part of the process. It is usually entrusted to one or more people who use their experience and instincts to meet the needs of future users and customers.
When we conceive something, we have our own vision of what we need to create, what needs to be improved, what doesn't work perfectly and what we want to improve. This vision will always be more specific and more effective if it is enhanced by user feedback.
Getting rid of such information means taking the risk of missing an important design detail, or designing a product for oneself but which will not reach its target (see "Listen to users to an order to prevent a 43% chance of failure in your projects' redesigning").
Once this work is done, the specifications are defined. And then, as the saying goes, "the devil is in the details". Depending on the case, you will create a written specification, or graphic mock-ups, or even a non-operational prototype. Again, user tests will be useful to compare this specification with the opinions of actual people. Listen to their feedback, study their reactions, make sure they properly understand your design. This way, you can improve your product before you even develop it, because it's much cheaper to change a specification than to modify a finished product.
At the end of the process, projects often undergo quality tests before they are considered ready to be launched (putting a website or a mobile app online, launching a product, launching a service, etc). This is necessary to guarantee that the finished product complies with what has been laid down.
Yet, specifications may have been carefully followed without achieving the result that customers expect. Again, it will always be more cost-effective to perform user tests before releasing the product on the market. At best, we will get a go-ahead for the launch; at worst, we will be able to review the product before investing in a launch that will not bring a good return on investment.
How to manage user tests within your process?
The issue is not about managing user tests; it's about managing projects, based on a process that includes user tests.
The first thing to do is to clearly define the steps that make up your process, as seen above ("Cut out to rule"). User tests should represent one (or several) step(s), ideally with a clear definition of responsibilities, expected results and allocated time.
You then need to integrate this process into your project management tool. You'll need to be able to rely on it, whether it's for meeting deadlines or for assignments. Almost all project management software is based on tasks, which must be assigned as projects progress, with no real management of deadlines. This actually doesn't represent a workflow, which is the only way to guarantee the efficient management of the established process.
Therefore, it is essential to choose a tool that will effectively assist you; not by relying on simple tasks that are difficult to master, but by managing all the steps of your process. You will gain in efficiency and productivity.
Remember to give user tests all the resources they need for their full exploitation. This may require people to take care of them, time to properly run them, and time to analyze them in detail. It's up to you to define your resources according to your objectives. Ferpection is at your service for all these aspects, do not hesitate to call on them.
About the author: Amaury Bouchard is a project management specialist and creator of the Skriv solution. We would like to kindly thank him for writing this article for the Ferpection blog.