Question about User Centered Design in combination with prioritizing requirements


#1

I’m not sure if this is the right topic. If this is not the case then it should just be changed to another part of the discussion board.

Hopefully someone can help me with this issue. I’m a student graduating in Holland.

A while ago I conducted some research with the User Centered Design “method”. Various requirements were brought to the surface by customers, the business and my own experience for a new application.

I’ve read articles on the internet and read a couple of stories about prioritizing requirements with various techniques but non mentioned the imput of my own research (there was only the stakeholders imput on requirements).

Now my plan is to: ask a few members in the company to prioritize the requirements. They all have a 1 “score” for a Must or a Should. As a researcher I want my own opinion to count double for the design I’m making. If a score (on 3 employees) and me is more / or 4 (max. is 5) then I’ll implement the Must-Have or the Should have. Is this a “correct” method and can or is it proven acceptable?

Any good help would be very much appreciated.


#2

Why should your opinion count double? Shouldn’t everyone’s opinion count double?


#3

Hello jmspool. My opinion should count double because I conducted the research with the customers. In other words (as my teacher says) I should know best what the customers want because I did the research.

He (my teacher) says that I should only prioritize and discuss with the company the requirements and that I’m the “expert”. I look at the whole picture as a rather more democratic process.


#4

I think your “opinion” isn’t any more valuable than anyone else’s, so should count the same. I think your teacher has misled you.

If you did research, then you have observations. Those aren’t opinions.

Having observations doesn’t make you an expert. It just makes you the person who has the research observations.

And if nobody else has those observations, then you’re hoarding information that would be useful to the team. The role of the researcher is to surface the observations so everyone has access to them.

What have you done to surface your observations? How can everyone on the team base their thinking off of the same observations?


#5

Hi jmspool. First of all thank you for your posts. I’m really not to sure if my observations are just an “opinion”. I’ll explain why.

Obviouslly I did the “observations” or “research” as you may call it. Since the group of interviewed customers was very small (5) I would like to say that their opinions and things they need in the new application are a sort of general things (the group is representative for general “ideas” and opinions) so the requirements are things that they need in the new application.

I presented all the requirements with an example and the problem hence where it came from. Everybody got this same piece of information. So everybody from support to development knows the same about things that came from the research.

Now comes my opinion based on the facts. I know the research report upside out and as a designer I want some kind of influence (design push) on the whole requirement prioritization.

Overal I find the prioritization very complicate. Who should you “choose” to prioritize, what is their balance in the company. I find myself almost stuck in this part because eventually you are still doing concessions with the “big boss”. Keep in mind that should 3 employees say: “I want this” that without any consideration a customers need could be lost. I find it a very unbalanced view as I may say so. Still I have the time and furthermore I’m looking more into this by subscribing to ixda.org…so another step towards the “way” perhaps :smile:


#6

Hi Wallace,
Like Jared I’m a little confused by your set up here.
If you gave your team all the information as you described, shouldn’t you then be asking why aren’t they “voting” to prioritize the same way you think they should?
I agree with Jared, that your role as both researcher and designer doesn’t give you any more weight in setting priorities. The Product Manager has other understandings that you don’t, and so do the developers. Your piece of the equation is not any more valuable than theirs.
If you disagree, you should seek to understand where your teammates are coming from. If they disagree with you, you should ask them why, seek to understand, so that maybe you can have an opportunity to clarify your points and find out what they value, so you can be more influential.

So a reminder of a few truths:

  1. You probably don’t have all the information and that is OK.
  2. User-centered approaches are not the only viable approaches, but is one when coupled with others works really well. (respect differences of POV and learn from them)
  3. Once you start taking the position of presumption, you will loose all hope of influence.
  4. Even with the best reported data things like incentives, metrics, and just personal bias among other things will keep people from seeing your data the way you do. Your team has goals and motivations and so you need to be both respectful of them, as well as you need to design for them (it’s another UCD problem space).

I’d be interested to hear from your professor/teacher and have them explain to me their thinking about why the designer’s opinion is more important than other people’s. I’ve never heard that in some 20 years of doing this work professionally. Ultimately respect and influence are earned, not given.

Hope this helps a bit.
BTW, 5 subjects can be valuable in research if the redundancy of the data is high and if it can be supported with heuristics and principles.

– Dave


#7

For all research, there’s a process:

Observation -> Inference -> Opinion -> Recommendation

It starts with raw observation.

  • “The user never clicked on the “Confirm” button.”
  • “The user said they’d like a feature to reconcile balances with the statement.”

These are what you observed during the research. Anyone else who paid attention to the research would get the same observations.

From observations, you draw inferences.

  • “The user didn’t see the confirm button.”
  • “Reconciliation is a necessary function in balancing accounts.”

Inferences aren’t observed. They are added to explain the observation. Different observers would bring different inferences to the table.

From inferences, you come to design opinions.

  • “The design of the confirm button is hard to see.”
  • “Reconciliation is a feature customers would upgrade for.”

Opinions are where you integrate your experience as a designer, product manager, or other key member of the team to factor in the importance of a future change to the design.

Based on opinions, you make recommendations.

  • “We should clean up the visual design of the buttons.”
  • “We should build a reconciliation feature during the next sprint.”

Recommendations are the last step before a decision is rendered.

(All of this is described in more depth in The Road to Recommendation.)

You said you think your opinions are worth more than others. You haven’t explained why. Just being the keeper of the observations doesn’t automatically grant you the extra value in your opinions or recommendations.

If everyone has equal access to the observations (meaning they have seen them and understand them like you do), then their opinions (based on their inferences from that set of observations) should be equally waited, plus whatever other experience they may have in the business.

If you know the observations better because you did the research and they didn’t, then that’s a failure in your distribution of the observations. (Observations observed are higher quality than observations reported.)

You don’t automatically get more weight because you conducted the research and they didn’t. It sounds like if you want them to take the research under consideration more, you need to get them involved in it more, up front. (That’s my recommendation, based on what you’ve said and the inferences I’ve drawn from your comments.)

This may help: Fast Path to a Great UX - Increased Exposure Hours


#8

Good morning,

Thank you both for the help. I really appreciate it. Since I now have little time and want to read your info carefully I will respond this weekend.

My explaination why my opinions are worth more then others is the following:

For my education I’m researching two applications on usability. My “research” has brought up a Research Report for the business. I’m advizing the company in this Research Report to make a renewed design with the customers remarks etc.

When reading your info a lot was made clear to me. For instance: My advice has nothing to do with the prioritization of the requirements. For now it’s still a “little” unclear what the weight is from people in the company. I quess an ICT Operation Manager and ICT Manager including a Customer Support Agent will do…since there are a total of 9 employees I quess the weight of the prioritzation is sufficient.

Excuse me for being a little confused. I was aiming to fast/far at the Research Document. What my teacher problably meant and I’ll ask him tonight in a Skype session is that a Research Documentation should (at least) contain some advice for a direction for the company and NOT how a design will come to look like. I really have to check this issue. For now I think I really misunderstood that part.

Still leaves me (a little) with the question which stakeholders are most representative and how this “representation” is absolutely correct. For me it seems a bit “misty” behaviour as it is almost a feeling. For example. Mister Boss has everything to say about the look and feel of the designed application. He is not responding to my requirements list so I thought doing a center approach on other “important” employees is “better”. I’m not to clear on this perhaps…but in other words how do you measure the balance of the persons who are prioritizing the requirements in the Scope? Is there a mathematical rule for “the way”.


#9

Hi Dave and Jsmpool,

I’ve looked at some remarks of my teacher. First of all I read the info wrong. There was quite some feedback in the Research Report and maybe that’s why I was a little confused about it. His only remark is actually that I should not ONLY let the company decide about the requirements. My influence is just as important as the employees is.

Still leaves me with the question how to find the right “structure” for the priorization but I quess there are hardly any rules for that as long as the people in the process have sufficient influence and decide about the “final” design and implementation of it.

I think creating a foundation for your design (and requirement prioritization) is a very important key issue in this yourney after the step of prioritization. So saying that out of 4 people (including me) 3 from 4 is sufficient to go throught to the design is “democratically” approved. You just have to make some kind of decission based on the information and the points. 3 (from 4) is more then 2 is the “deciding factor”.


#10

Good evening Dave and Jsmpool,

I took another look at your remarks and found them absolutely correct. Thursday evening I spoke with my teacher. It’s OK to prioritize the way I was planning to. So everyone get one point (from 4 including me) and when 3 or four have been reached the requirement will make it to the design.

There was another way and I think somebody also explained this to me. I could organize a meeting with the members and discuss about the prioritization. For now I’ll just keep it with the first manner.

I wasn’t to sure about how requirements should be prioritized. What my teacher also said that the second way (discuss in a meeting) would be more easy when there were a lot of big differences. For now this is just not the case.

So my next step would be making the Use Cases. I seems like an endless work but stil with 3 months to go I’m just gonna do this design step ofcourse. After four years of school I’m so tired about it. Hope I can give it my last best shot to graduate. Will come back in the forum from ofcourse to see what’s going on!


#11

A bit late to the party and just chipping in with my approach to world domination

Any report I have ever written has the raw data from observational testing. This is either a written transcript of what the user has said or a video of the testing session. I will report a finding, back it up with data and recommend something. I’ve never had any problems with fellow employees, clients etc because the evidence is right there in front of folks


#12

As always, Jared’s comments hit the point really well. I would add, though, that opinion might not be the best way to prioritise.

The opinion and recommendation phase here is very much the Lean UX hypothesis concept. It’s an opinion that needs to be tested. Testing means that it may very well be flawed. So not only should opinions be weighted equally, they should all be testable and the holders of those opinions should be open to being wrong.

When I work with clients to prioritise changes to existing designs, or to implement new designs, I get them to score different pieces in terms of urgency and importance. Those two scores then get plotted on a matrix so we can easily see that the business thinks needs to happen.

I favour prioritisation on a an approach of business needs rather than individual opinion.

I hope this helps too.