Design Process and Methods (in Depth)
This will be in-depth documentation of my design process and methods. Amongst dozens of projects I did and I will do, every design process and method will be different; I am always introduced to new techniques to use as well as regrets and improvements on existing techniques.
However, we get stuck in the idea of double diamond and forget about little details that just make that make double diamond work. Through this documentation, we will try to document every techniques I could think of when it comes to the design process (Which, I will not go into details into how each works but focus on how that influences the process).
This will be ongoing documentation throughout my career.
I believe all decisions are driven by two things in this process; what data we need to achieve, what objectives we are working towards. In most design processes, the processes and methods are determined by the data and objective I am trying to get. If it is a small, UX writing issue on a button, I would simply deploy A/B test. Whereas a completely new feature will go through an entire design process from user need research, idea generation sessions, iterative prototyping etc.
I must guess most of us already know what this is. A process of double diamond, that goes through the discovery of problems, refinement of problems, the discovery of solutions, refinement of solutions, iterative development stages.
This is a way of thinking that allows us to really identify what the product value is and how we could deliver it. If you love a product, think about “why” that is so lovable, what benefits it gives you.
Please see Julie Zhuo’s wonderful essay on this: https://future.a16z.com/product-thinking/
I would describe this as a iterative human resources method. Setting up a sprint for a certain period of time (usually week, bi-week) to complete a set of projects. I think the key is communication between developers and designers. Can designers ensure that the project is ready to be developed by the start of the sprint? If so, how can we communicate this effectively and find anything a designer is missing?
Zero to One
This is based on Peter Thiel’s Zero to One. The guide on how we make a successful product and business. There are some really great tips for startup founders but also for designers. He provokes questions like is your product 10 times better than other products? What “secrets” do you know about the industry?
Now, there is a whole industry on this one too. I would say this has ties with agile methods as well in that it will create “experiments” to test out which method will create an increase in KPIs. For designers, growth hacking is a big good-to know because of AARRR(Acquisition Activation Retention Referral Revenue), for designers, helping users get to the core value of a product is extremely important, experimenting to optimize AARRR will generate a higher chance for users in doing so. As wells as creating a high impact on business outcomes.
Target customer definition
If you don’t have a solid customer base yet, or your product is constantly pivoting, it could be difficult to identify which customer base you need to reach out to. One way I do this is by just identifying which customers are using and talking to them, “what was the hardest part about your previous solution?”. This will give the most crystal clear and replictable view of what your customer is trying to get.
Other way to approach this is through Five whys and base your thinking on the product. Five whys will ultimately answer why people need to use your product. Perhaps if you are making re-usable package, ask why they want to use your product, why they want to reduce impact on the environment etc.
Also, I tend to think sometimes demographics like gender, age may have nothing to do with the target customers. For some companies, it might matter a lot, but it all depends on the user needs and behaviours.
Data Analysis is one of the most useful ways of identifying the business problem. Usually, this is handed out by a data analyst, marketer. But sometimes, especially in a smaller team, we might have to do one ourselves. For UX designers, churn rate is the most important like “how many people are dropping off at A screen?”, or a result from A/B Testing.
Before interviewing users, online ethnography is one of the easiest methods of understanding the user and start defining a question to ask. You can search on the various reviews on the internet what people are thinking about your product OR any product you wish to investigate.
It is a simple analysis of how competitors are doing in the market, what their market value and capital are, how good their product is etc.
However, as UX Designer, we need to think about user needs not the market. As saying goes, Nike’s biggest competitor is a TV, not Adidas. If they are in direct contact with user needs we are dealing with, they need to be dealt with more seriousness.
Also, thinking about 10x rule, at least one aspect of the product has to be ten times better than the original product to be successful if starting at a small scale. Think about ways a competitor’s product or your product has the potential to be ten times better.
Also, Think about the Kano model. How does your product compare to other products in terms of Must-haves, Performance boosts, Delighters?
User Needs Research
We can understand Jobs to be done by famous milkshake example. How users were looking for a milkshake that doesn’t melt quickly? We can identify the main JTBD by asking series of questions like this one from Eric Migicovsky.
1. What’s the hardest thing about [doing this thing]? 2. Tell me about the last time you encountered that problem… 3. Why was that hard? 4. What, if anything, have you done to try to solve the problem? 5. What don’t you love about the solutions you’ve tried
Plus, Five whys to ask ourselves.
This leads to understanding WHAT jobs our product is hired to do (this can be broken down into smaller steps) and WHY. This is a really strong way to innovate rather than just asking users what they want; most of the time they have no idea how the product to turn out.
User interviews are one of the most essential ways for a designer to understand user needs. One way I like to approach is: if it is about user need, take questions from Eric Migicovsky, if it is about understanding usability, take questions from Design Sprint where you categorize questions by screens.
Also, I love asking why. Why do they like certain screens? Why do they like doing action A? Most of the time it is a miss, but one hit can really make a difference.
A Focus group is a multi-participant workshop that asks users to go through activities like reciting their experiences and opinions about certain products. This is a great way to get deep into the problem. But just be aware that there is a potential of groupthink, over-representation of a certain participant. It is best to keep some activities silent to prevent this.
Co-design is an emerging design method usually implemented in academia. Co-design is basically a workshop where you come up with certain situations to design for and prepare props to help your target users to design hypothetical solutions. This is a method to discover user needs as well as filling in domain knowledge gaps (if you have discrepancy of knowledge with your user)
Co-design is perfect in situations where you are designing for situations you don’t have much domain knowledge about. For example, if you are designing something for optimizing medical training, working with clinicians to design a UX or UI will boost up the relevance of your work with the users.
User Needs Research Synthesis
One of the most popular and proven ways to synthesize information. It certainly has the benefits of finding qualitative information in the most logical way possible.
I think a storyboard is one of the least used but most effective ways to capture the main problem-> solution of your project or product. If done right, it will be a great way for your colleagues to commonly understand your product.
User Journey Map
A user journey map is really great if you are diagnosing the entire service. Especially with analysis of user’s emotions, it identifies where users have problems.
Hey, if you are talking to developers and business managers, they don’t have time to go through your comprehensive user journey map. Sometimes it is best to keep it in dot points. What the main problems you found, what are the future directions.
Problem definition / Creating Vision
Keep thinking back to product thinking. So you understand exactly what users need you for.
Jobs to Be Done
If you have done asking questions for JTBD, by this stage, you can define what the JTBD is.
Vision / North star
Paraphrasing to Dan Olsen, you need to take these into consideration when building your vision: What world do we want to create? People want to buy their future, not your product! So it is wrong to talk about product when talking about vision, think about user’s future.
North star is KPIs to end all KPIs. What do we ultimately need to achieve? To do this, we need to understand which game we are in. Do we need to consider how many time user spend on the app? OR number of transactions? But also remember it cannot always tie with the revenue, think about the user value and revenue should come later.
How Might We?
How might we is an activity to make a pain point into an actionable question that we can prioritise on. Once you have a picture of user’s experience on user journey map, there could be multiples if not hundreds of pain points. HMW is an opportunity to narrow down these pain points and your consideration of JTBD, 5Whys into fewer statements.
Solution Concept Building
Crazy 8 is a popular method for rapid concept development. This especially serves well in a team situation because we can compare large numbers of ideas (not that the numbers matter). I usually do this based on each of How Might We statements. Also, at the end of 8th, you can expect somewhat more matured idea. Similarly, you can use brainwriting.
Sketching would be a bit more elaborate details upon crazy 8. This gives you a chnace for your idea to be communicated by other people easily. I usually prefer this method especially when the problem is not sticky and require serious innovation.
Silent Voting (As per Design Sprint)
For each of the sketches, it is usually effective to vote for multiple ideas, in silence. The reason for silence is that someone with more influence might temper the results.
MVP by 80:20
This is somewhat a basic product management approach. Defining core features that will create most of the value and developing them will help the product to be deployed or tested ASAP.
Low-fidelity prototyping is a common method for early-stage developments. It has the advantage of more honest opinions, faster iteration speed in usability tests and product validation. But I don’t see the advantage of faster iteration since usually designers derive their UI from design systems with low effort.
High-fidelity prototyping on prototyping software is my go-to method. Any new ideas, I can whip up something in few hours to be tested. But only problem would be when we are dealing with AI, AR,VR where the features are not just represented on pre-set screens. While, there are few ways to get around in Figma.
I usually do this through Unity. If you build a base project on unity, VR prototyping is not a difficult task.
Experiments are one of the most used methods for testing. This is usually used to test the quantitative results of UX/UI changes. Even if it is not for growth, sometimes, you just have to deploy them and see how people use them.
Similar with growth experiments, if you are trying to introduce changes or new features, it is a standard procedure to A/B test them and compare how they perform. Just to keep in mind, the results might have conflicting interest. A version with higher engagement might decrease usability and vise versa. Just make sure what your aim is and discuss it among your teams.
The think-aloud test is a favorite and information-packed usability testing method. This is good to test something completely new, you will get good qualitative feedback as well as little bits of quantitative results.
This is a survey-based test. This basically asks users about their opinions about usability. Although it won’t give you “What” is missing, you will see the progress. I don’t use this often because this has a low return on investment on a small scale.
Applications (MAZE, UserTest.com)
Nowadays, usability testing applications are on the rise. You can just send out prototypes with the tasks to test users’ performance and ask for feedback whenever they want. This has a downside of transitioning a usability test into an interview.
User Testing Synthesis
As discussed above. A very standard procedure
User flow diagnosis
Based on the usability test, you can make quantitative, qualitative analysis per screen. This is a neat way of visualizing the use case funnel.
User journey map
Again, user journey map of an update version is another standard procedure. There is literally no limit to this so you can put a lot of data if you see fit
If you have done a growth experiment or is making a progress report, you might have to make a data visualisation or summary of the results.
Atomic Design/Design System
This is not necessarily a design process, but there could be a need to create a design system (especially if you are working at a startup.) I would only advise to use time wisely, think about the visual you want and make multiple ideas.
For any graphical work or VR/AR, you might have to do a 3d modeling. I suggest if this is just a small part of a job to use Adobe Dimension to make something quickly.
I found that in an agile, multidisciplinary environment, you don’t need to go too much into detail about your findings (Unless it’s a UX team meeting). If that is the case, make things concise in one document and just include the data as an appendix so you can always revisit them for any questions.
I am no expert at presentation actually, but I could a presentation with mainly graphics and little to no word most effective in most situations.
Slack Slack Slack
Slack is an obvious communication tool now. I use this to send over dot points, progress report, lunch hangouts.