Home > The Lean Post> It's Not All About the Data on Value-Stream Maps: An Interview with Judy Worth
The Lean Post
Sharing how the world is making things better through lean.

It's Not All About the Data on Value-Stream Maps: An Interview with Judy Worth

by Cameron Ford & Judy Worth
May 26, 2016

It's Not All About the Data on Value-Stream Maps: An Interview with Judy Worth

by Cameron Ford & Judy Worth
May 26, 2016 | Comments (5)

Data is a critical component of value-stream mapping. But it's not the only component you should be concerned about. All too often we find mappers getting caught up in their efforts to gather data, which can lead to overload and inaccurate mapping. LEI faculty member Judy Worth has seen this many times - I recently sat down with her to get her perspectives:

Where does this problem come from?

I think we’re living in a culture right now that says you have to have data – lots of it – to make an informed decision. Bill Gates said if you have the measurement right, you can change the world. And now we’re just drowning in data. The question is, what’s meaningful data, and what quality and amount do we need in order to improve?

Why exactly are we seeing such a big problem with this?

Part of it comes from IT, which has given us the ability to collect and analyze large amounts of data.  We also live in a world where people are asking for transparency and the ability to show impact, or lack of it.

And you see this everywhere – in manufacturing, in healthcare, in office and service work, you name it. Everybody everywhere is looking to the data for answers.

What are the consequences of someone getting lost in the data?

There are quite a few. For one it can send you off in the wrong direction because you haven’t identified the key data yet - you’ve just gathered a lot of data. And another is data paralysis – people just feel like they can’t move forward without first gathering all the data there is.

Can you describe a time where you helped somebody deal with this problem?

Sure. Most recently I met someone at the 2016 Lean Transformation Summit out in Vegas who wanted my take on this. He was from a manufacturing plant and he said that his plant had been gathering data for a value-stream map for over one year and in that time they hadn’t begun any improvement work. He was really frustrated with the VSM process because he was under the impression that you couldn’t start without collecting all the data there was to collect.

So I shared with him that many of us out in the field have the experience of being able to use data estimates provided by good, experienced, frontline people who touch the work as a starting point, then start mapping and get the really hard, clean right data later. The key is recognizing that you need a baseline for improvement, but you don’t have to have that high quality of data to be able to figure out what the problems are, where they’re located, and which ones are the major heartburn and which ones are just little discomforts. You want only those metrics will help you deliver the customer’s needs and the business’s needs.

What would a VSM produced with an overload of data look like alongside one with a more focused approach to the data?

I don’t know that the maps themselves would look especially different, but here’s the thing – very often when we look at data estimates we look at ranges, so that we try to cover what happens 80 percent of the time, low-end/high-end range for things like process time.  Then we start to look at why that range exists. In contrast, somebody who’s overly focused on data often wants to include a single “correct” data point based on a huge collection of data, which can cause them to overlook some important information about where problems are occurring.

What advice would you offer for ensuring that one doesn’t get lost in the data-gathering when planning for a VSM?

I always remind people that not all data is equally important to tell the story of the value stream. Having perfect data before starting the mapping is not critical. Having  good quality data once the mapping is under way is critical because you need to show whether the changes you’re making are having an impact.  I would ask someone about to start a VSM, “What story are you trying to tell and what data would you need to at least have a good enough estimate of to tell the story?” If you narrow their focus they’ll be on their way.


What to do next:

The views expressed in this post do not necessarily represent the views or policies of The Lean Enterprise Institute.
Search Posts:
Learning to See / Creating Continuous Flow Set
By Mike Rother, John Shook, and Rick Harris
Learning to See / Seeing the Whole Value Stream Set
By Dan Jones, Mike Rother, John Shook, Jim Womack
Was this post... Click all that apply
19 people say YES
16 people say YES
11 people say YES
6 people say YES
Related Posts
5 Comments | Post a Comment
Ken Hunt May 26, 2016
1 Person AGREES with this comment

Thanks Judy. You have captured and explained some of the pitfalls of analysis paralysis. It's better to have some data, get started, and let the momentum build as you collect the NECESSARY data. You want the team to feel good about what they are accomplishing (Building the map) and where they will be going (Future State).


Reply »

Jørgen Frost May 27, 2016

I absolutely agree. We are very concerned about the process when doing VSM in hospitals and less about the data. The VSM is very useful to help understand the other parts in a process, and hence, to generate a lot of ideas for improvement. The VSM focuses on the value for the patients which is very easy to discuss. And you are not focused on guilt. After that we can take the best ideas for PDSA and then we decide which data we need for baseline and which goals to achieve and how we wnat to measure.

Reply »

Frank Sabala May 27, 2016

Thanks for the article.  Do you find this approach to be true for most activities we do where data is involved (process mapping, problem solving, etc...)?

Reply »

Calvin L Williams June 06, 2016
1 Person AGREES with this comment

Hi Cameron & Judy. Great post! Thanks for sharing.

I just wanted to add a technique that I've used in the past that has been effective as well. It's more of a project management technique that may help your friend from the event to get better results.

Start from the end and work your way backward. This means to create a map that starts with the end result and draw connections for all the inputs that are needed for the desired output. Then determine what specific analysis / datapoints would be needed to execute each input. Finally, allocate timeframes and owners to each piece of the puzzle. 

This method forces you to respect the constraints of time and other resources, which is a waste elimination technique in itself. Reguarding the quality of data, you try to get the best quality of data you can within the timeframe alloted. The key is to not use inaccurate data but make assumptions where you must. Just make sure to clearly display your assumptions. If they are still relevant in the end, you can always invest more time to sure them up at a later date.

As long as you're working with a cross-functional group, you can usually get pretty good estimates fairly quickly. As you mentioned, you just want to highlight where the most process waste is occurring, which will be fairly obvious if this is the first VSM exercise done on the process.

One last thing to understand about the data - it's constantly changing. Process data can become obsolete very quickly. Any data collected a year ago probably expired a long time ago. It's key to get sufficient numbers quickly so you can start making decisions and experimenting.

Calvin L Williams
Continuous Improvement Strategist
Percent Perfect Methodology®

Reply »

ejder May 14, 2017

Thanks Judy,

not only because you remind us the importance of VSM but also maybe more because of the appreciation on the action rather than being steady.

today many of people are lost in details and late for the action, lack of estimation and planning but really love to talk on the data, observation etc. which of all i agree but see no reason to stay on the start point and watch everyone running..

improvement is so-called "contiuous" so that we should start and step by step get the results, the data, the counteractions but all in all "improve"

Reply »

Search Posts:
Learning to See / Creating Continuous Flow Set
By Mike Rother, John Shook, and Rick Harris
Learning to See / Seeing the Whole Value Stream Set
By Dan Jones, Mike Rother, John Shook, Jim Womack
What Is Your Leadership Model?
"Too Busy to Walk the Gemba"
Please include links as plain text URLs only. Do not copy and paste directly from a web page or other document. Doing so may pick up additional HTML that will not function here.
URLs will be converted to functioning links when your comment is displayed on the site.
Here's an example:
See this article for more details: https://www.lean.org/whatslean