Cause and effect with fishbone (Ishikawa) diagrams
Cause and effect with fishbone (Ishikawa) diagrams
"Everything Starts Out Looking Like a Toy" (No. 65)
This week’s toy: a generator of random illustrations in different styles. It’s more than filler, though - it’s a way to provide consistent visual language in a product prototype or a content document. Go beyond gray boxes in your designs. Edition No. 65 of this newsletter is here - it’s October 25, 2021.
The Big Idea
What’s the key to getting better at operations? Understanding cause and effect, and helping the rest of the organization to see the impact.
Start with Simple
When you are trying to debug a problem that has happened in your operations world, start by starting. You will first need to know about what was intended to happen to determine what actually happened.
If there isn’t documentation about what is intended to happen, you might need to do two things in quick succession:
Find out the intended steps of the process to find the issue
Analyze each intended step to determine whether improvements might be possible (or logged for the future).
Here’s a quick example:
Imagine that your sales rep tells you that they don’t see an opportunity they’ve worked recently in a Salesforce report they use every day.
From their perspective: “my stuff is missing - how is this possible?”
Reviewing the report reveals that it filters opportunities with a close date in the next two quarters. When you look at the opportunity in question, you find that instead of being set with a prospective close date of December 31, 2021, the rep has instead set the close date to December 31, 2020.
What was expected?
We expect Sales reps to select the correct opportunity close date
If the date is wrong, the Sales rep expects an alert that there is a problem
What must have happened
There is no system control that prevents a past date from being selected
There is no affordance that lets the rep know there is a problem
This is a simple example, where it’s easy to see that you need to create a validation rule disallowing past close dates when comparing to the current relative opportunity create date or today’s current date. But at the time this workflow was created, perhaps this situation wasn’t anticipated.
Starting with the happy path
You’d like to follow the figurative “yellow brick road”, and that’s where many business processes start.
This is not a bad place to start, as you want to end up in the right place, so why not design the process where everyone takes the right steps?
“A complex system that works is invariably found to have evolved from a simple system that worked” - Arturo Rios
However, there’s usually a new case for almost every time someone encounters a system, and when you hit an error the root cause analysis would reveal that there was an unexpected action that wasn’t in the original happy path.
Switching the view to “what will go wrong”
Consider a Ishikawa (or “Fishbone”) diagram. This frame gives you a way to brainstorm all of the possible causes for the effect you are seeing, and to enumerate possible reasons for each effect. While this diagram doesn’t estimate how these factors work together or are isolated, it does give you a repeatable process to examine what happened, why it might be happening, and the multiple causes that are in play.
How do you build fishbone analysis into your ops game to get better? By treating the Ishigawa diagram as a another tool during process design and debugging.
Here are a few ideas to get started:
treat every operational process as a workflow product. What are the inputs, outputs, and expected actions? Who are the actors? What are the most likely things that go wrong?
intentionally feed bad data into the process. What happens when you don’t follow the yellow brick road?
create a checklist for the steps required for the output of this process to be an M.V.P. - a minimum viable process. What is supposed to happen? How long should it take to occur? How would we know things didn’t go well? Will the process try again?
What’s the takeaway? Getting better at process requires process analysis. It sounds meta, but it’s true. When you think about how you solve problems (whether things are going well or not), you find and resolve issues more consistently.
A Thread from This Week
Twitter is an amazing source of long-form writing, and it’s easy to miss the threads people are talking about.
This week’s thread: why those Spirit Halloween stores exist. Their financials are not that spooky, and they are a good model for other pop-up or seasonal businesses.
Links for Reading and Sharing
These are links that caught my eye.
1/ A search engine for 3d - What does it look like to build a way to search physical things based on 3-d models? This is a fascinating podcast about mapping the physical world. Imagine pointing your phone at that broken piece in your dishwasher and being able to order a replacement piece. That would definitely be better than today’s alternative.
2/ How do streaming services pick? - “Binge watching” some days looks much more like doom scrolling the trailers for shows or movies you’ve never seen rather than finding exactly that thing you want to watch. So how do Netflix and other services decide what to show you in their main pages? They try to maximize happiness for $ spent in their content budget.
3/ On being a counter-thinker - If you’ve wondered how to build a practice where you evaluate ideas, consider if they are worth pursuing, and then take action, “How to be a smart contrarian” will help.
Speaking of being contrarian, if you’d prefer the tl;dr from this article rather than reading it, here’s a summary:
On the Reading/Watching List
Watching: your moment of zen, or watching moths at 6,000 frames per second.
Reading: Arriving Today, by Christopher Mims. Reading anything recently about supply chain worries, store-stocking issues, and the “bullwhip effect” in the way we order and consume goods? Christopher Mims shares the details (and the big picture) on how items go from bits to atoms and how exactly they are shipped and delivered around the world. Whether you are buying your holiday presents locally or order everything from the “Everything Store”, this one will be worth a read.
What to do next
Hit reply if you’ve got links to share, data stories, or want to say hello.
I’m grateful you read this far. Thank you. If you found this useful, consider sharing with a friend.
Want more essays? Read on Data Operations or other writings at gregmeyer.com.
The next big thing always starts out being dismissed as a “toy.” - Chris Dixon