The Lean UX Method

Based on my interpretation of
Jeff Gothelf’s Book: Lean UX

Overview

Jeff Gothelf’s LeanUX , it is a method to improve designs of UI/UX on digital interfaces. The method has been of great help during my last few years as a Software Designer. Although my process is a combination of multiple disciplines, I use it as a tool to frame problems or design requests in order to achieve more accurate outcomes.

My current implementation of the method still under the testing phase. So far, it is working very well and I would like to share some highlights of what I had learned. I recommend to read the original book as I have cut and modified the method to adjust it to my own need.

Framing the problem

Many designers are used to receive design briefs from a manager who usually deal directly with the customers and synthesise the requests into a single page. As designers progress in their careers, they get to deal directly with the clients who quite often require some guidance on how to communicate their needs.
Without a method to guide the clients, the design process can be extended far from comfortable and profitable and results could end up completely far from what the client wanted.

One of my best example I remember is from a client from the fibre optic network design industry who requested a simple icon for a software they were working on. This is what I received. I recommend to read it all:

“The module takes the output from the FSAM Solver (the cables and joints) and rewrites the IDs of cables and joints and the START_ NODE_ID and END_NODE_ID of the cables. It’s used by planners when they make manual changes after running the FSAM Solver.
For more detail (and a very simplistic example): It starts at an LJL (that’s a joint) and renumbers the LSS cables (there IDs) that come from the LJL (usually up to 4 LSS come off an LJL). It updates the START_NODE_ID of these cables with the ID of the LJL. It then updates the IDs of the AJLs which connect to the LSS cables, and updates the END_NODE_IDs of the LSS cables. It then updates the IDs of the MSS cables, and the START_NODE_IDs of the MSS cables using the IDs of the AJLs. and so on and so forth.
(note sure how much of this you already know…) It would usually have LJL – LSS – AJL – MSS – MPT – SDS – PCD. It would be like a tree with one LJL, upto 4 LSS coming from the LJL and an AJL at the end of each, up to 10 MSS coming from each AJL and a MPT at the end of each and then up 8 SDS coming from each MPT with a PCD at the end of each. Obviously, any logo doesn’t need to include all that detail.. but hopefully it helps for thinking about ideas.
let me know if you want any more/different info.

Cheers”

Scientist are usually not the best communicators. Although this example is quite an “edge case”, it is not uncommon in my profession to receive this kind of requests.

After digging a bit and expending few hours in back and forward communication, the translation of the request to a normal human language was:
“We have a software that helps to join 2 separated fibre optic cables. Can you create an icon for this?”

Fill the Grid and start the design loop

The ancient philosopher Plotinus created a well-known analogy of a sculptor cutting away that which does not enhance the desired image. This process is absolutely necessary for us.

What makes the Lean UX method very effective is its collaborative nature. The process begins in a meeting with the design team and client to identify users and, roles and desired outcomes. The grid is filled up with assumptions that will require further testing. This process is made in collaboration to align vision and expectations.

See example below: an exercise for a Cafer Roster app.

Example

Template

First Draft

After filling the grid the fun part begin: the whole team is invited to a drafting exercise and no special drawing/design skills are required. All ideas are valid. The different points of views enrich the experience and participation can be over multiple locations simultaneously.

It is very important to limit the time to 5 minutes and use pen a small paper. Any digital tool would be an obstacle at this stage.

Assumptions

After presenting and comparing the 5 minutes drafts, all assumptions are put into test. The collaborative nature of the process brings to light new interpretations and solutions to the problem. The experience is very rich and immediately feels like a big time and issues saver.

However, the test it’s only internal. We need to create metrics and benchmarks to test our hypothesis. In this particular case, the metrics are binary: the application is being used or not. The entire functionality depends on employees and employers using the app, meaning feeding data into the system. Without data, the result will be schedules in apparent available spots for rostering.

This could happen for few reasons:

The app is not user friendly and employees resist to use it.
Employers allows manual process (like phone calls, verbal agreements, etc.) and do not feed the information back to the system. Probably because it is time consuming.

Our assumption for metrics is that a very user friendly app will lead to 100% utilisation, which is crucial for the app to function. Therefore, usability is a key component of the design.

Minimum Viable Product

Now it’s time to create a small prototype for testing. This section will be continue on Part 2/2 of the post. Coming soon.

Original image from the book.

0 replies

Want to join the discussion?
Feel free to contribute!

The Lean UX Method

Overview

Framing the problem

Fill the Grid and start the design loop

First Draft

Assumptions

Minimum Viable Product

Leave a Reply

Leave a Reply Cancel reply

The Lean UX Method

Overview

Framing the problem

Fill the Grid and start the design loop

First Draft

Assumptions

Minimum Viable Product

Share this entry

Leave a Reply

Leave a Reply Cancel reply