People Create Your Culture
Last week I ran a trial run of a session Don Gray and I are working on at XP Toronto. This session was a result of a session we did at AYE this year about MBTI and corporate culture.
Hypothesis: Is there a way to increase the odds of a successful change by understanding your organization culture from Schneider’s culture model and how the MBTI (temperament and function pairs) of the people involved in the change fit into each of the four culture types.
Here’s how I ran the session:
- Brief introduction to MBTI so people can become familiar with the model. I use my type (ISTP/INTP) to describe the model. For this session we feel function pairs (how people process data and make decisions) are better suited than temperament. The four function pairs are ST, SF, NT, NF.
- Exercise to help people figure out their MBTI function pairing. I assumed most people wouldn’t know their type so 4 statements were given and people decided what statement most closely defined their stance.
- People split into 4 groups based on their function pairing and created a mission statement and described what a successful ‘agile adoption or transformation’ would look like.
- Each group de-briefed
- Each person wrote their function pairing on a sticky note and posted them on the Schneider culture quadrant where they felt most reflected the type of organization they would want to work in. The culture types were not given, only descriptions of them.
- Group discussion
Overall the goal of this exercise was to see how MBTI function pairings and temperament related to organizational culture. The exercises were planned to not introduce bias by giving participants the labels of the function pairs and organizational cultures.
Here’s a picture of the Schneider culture quadrants and each participants function pairing sticky (click to enlarge):
Observations:
- Crowd bias: given this was an XP Toronto meetup, the group agreed crowd-bias came into effect where most skewed towards Cultivation and the ‘possibility’ axis.
- During the exercises, each function pairing group seemed to come up with a mission statement that aligned with the values of their function pairing. For example, the NF group’s statement was “am someone who is guided by my passions and beliefs, has a sixth sense about people, and works to ensure harmony in the workplace” Their mission statement was “In our organization we work to common goals, make the best possible work environment to maximize our people’s potential“
- The “NT” group didn’t finish composing their mission statement. The NT pairing is all about ideas and it’s common for lots of idea generation without a firm decision compared to other groups.
- The “ST” group had the least number of ideas and they were more firm and concrete. The ST group also finished all the exercises much quicker.
- The “ST” group debated about having highly specialized people vs more general specialists.
Virginia Satir Change Model:
We also mapped temperaments to the Virginia Satir change model and discussed how different temperaments are affected differently by change. SJ (sensing/judging) want to remain in the status quo to protect the group. NT’s (Intuitive/Thinking) tend to want to progress through the change as fast as possible and react with more ideas to changes that have yielded no results yet. SP’s (Sensing/Perceiving) want to move through the change as fast as possible to find the next problem to solve. NF’s (Intuitive/Feeling) want to make sure everybody is ok while the change is happening.
Combining the Schneider Culture model, MBTI and Virginia Satir change model can be an effective way to create awareness around an organization and it’s people to increase the odds of a successful change. I’ve often heard the Agile community say things like “change is hard”, “culture is important” and “one-size-agile doesn’t fit all”. I think the intersection of the these models is the “why” behind those statements.
I want to offer this workshop again to collect more data, if you are interested, please contact me!
How Much Can a Missing Test Cost You?
I went to a local electronics place with a friend at lunch today so she could buy a TV that was on sale. When the person at the counter rang it in, the price of the TV was ok but the “environmental fee” was $2,500 instead of $25. Oops. So naturally the people in the store blamed the idiot who entered the environmental fee data. Being the nut I am, I started wondering what could have happened. As I see there, there could be quite a few problems here:
- the data may have been entered correctly
- the data input screen may have made the data entry person believe they entered it correctly (perhaps there was validation that wouldn’t allow the operator to enter “25.00″ so they entered “25″ and there was some error that transposed the zeros incorrectly
- maybe the person entered “2500″ and forgot the “.” and there wasn’t a check to say “are you sure you want to add a “$2500 fee”?
- maybe once the item was scanned, the fee being displayed was being rendered incorrectly
- maybe the person entering the data was working on 12 things at the same time and simply messed up
- Maybe the data was added with a bulk script and there was no test to validate it
- maybe there wasn’t a test in place that could have caught the error before it was deployed
- maybe there wasn’t a conversation that a clearly out-of-boundary error wasn’t a big enough deal to worry about (I’ve never heard of a $2500 fee for electronics before!)
- maybe it was the data entry person’s last day and he really wanted to mess up this company
There could be many more possibilities, my point is there is much more going on than “the idiot who entered the fee incorrectly”. There could be numerous causes and a possibility for a large loss of sales revenue.
It gets worse. The people in the store didn’t add that fee, it was added automatically when they scanned this item. The people in the store cannot correct the environmental fee price. The people in the store called head office and “the guy was on lunch”, which is where we would have liked to have been.
So in my brain I see software doing something the user didn’t intend for it to do (add a fee), wouldn’t allow the operator to correct it and gave the operator no course of action to correct the problem with a customer who was prepared to plunk down a few hundred dollars right there. Were other stores affected? This was a chain store so I will assume their POS devices are accessing a centralized system. How many items do each of these stores sell that have this fee and how many customers would have been turned away until the problem was fixed?
And what really happened anyway? Was it a data input error? Was it a system problem? Who knows and it’s un-likely we’ll find out. All we know is we need to go back and pick up the item in a pain-in-the-ass-area of the city to get to and we wasted our lunch.
I’m probably nuts but I know how I would handle this problem, what would you do?
I’m Not Here to Be Your Friend
I was recently talking with a friend who was having some problems with adopting Scrum on a couple of their teams. They were describing symptoms I’ve seen before in organizations where the QA folks struggle to keep up with the rest of the team. Because I enjoy inflicting help on people, I offered to talk to them on my own time to offer some stories and experiences that might help them.
Some of the symptoms included not finishing the sprint work, disagreements about what done meant to the developers, testers and business people, struggling to finish regression, missed deadlines and releases. It’s always tough to gather enough context in a short amount of time to offer some advice. I could see the main challenge was getting the team to buy into a shared goal and removing the silos. Having said that, these guys seemed quite collaborative, I think they were simply having problems adjusting to Agile given they’ve only been trying it for a couple of months. I think they’re on the right track and seem committed to improving which is just great. Anytime people want to give up their own time to learn is a good thing in my books.
As we talked about ideas about how to find balance I suggested one way to help manage these symptoms is to pull less work. Of course every action has some impact and I mentioned that this can create idle time, especially in cases where “the development work” is low effort but the “testing work” is high. You could have developers idle! Nooooo! I could see an adverse reaction to that suggestion so I made a joke about making sure “resources” are 100% utilized! The reality is, people need slack, especially to learn a new process and move forward.
I asked the product owner flatly that if they are missing releases now. He said yes. I asked if they have quality problems as a result of in-sufficient testing. He said yes. I then asked what’s the harm of slowing down? Your missing releases with lower-than-desired quality now. You’re only fooling yourself if you think slowing down is going to cause problems because you already have those problems now.
I could feel a bit of shock from a couple of people in the room after I said that.
The point is, when you are looking for a coach or consultant, they’re not there to be your friend. They are there to help you figure out how to solve, manage or cope with your problems and you are not going to like some of the answers you get. A good coach is going to tell you either the brutal or kind truth and the best you can do is establish personal safety by having an agreement and shared understanding about what is expected from the coach from your perspective and the same from the coaches perspective. Once you can get that agreement in place what unfolds can happen in the name of progress.
