Читать книгу Scaling Conversations - Dave MacLeod - Страница 8

A waiter asks a group what they'd all like to drink. First person answers: A beer. Next person: Sure, me too: a beer. Next few people follow suit and order a few more beers. One person orders a glass of wine.

Then, the final person says to the waiter: “A friend of mine said you make one of the best margaritas in town…and since it's the first hot day of summer, I'll take one of those.”

Everyone else at the table considers this critical new information.

Then the first person says: “If it's not too much trouble, I'd like to change my order. I'd also like one of those famed margaritas.”

Second person: “Me too.”

Third: “Yep, me too, and unless I'm mistaken, make it a pitcher so we can just do a round of margaritas?”

The wine‐ordering person is the only holdout. “I'll stick with my wine,” they say.

This type of interaction is at the core of human communication. We share ideas, listen to one another, change our minds at the drop of a hat, and ultimately forget which idea belonged to who in the first place.

 “I'm glad I thought to order margaritas!”

 “That wasn't your idea, that was mine…”

 “Was it?”

In any conversation with fewer than 10 people, this is, in a word, natural. Small meetings create dialogue and interesting solutions surface. A single idea can become the most important idea in a heartbeat. During small focus groups, or in small chats online, discussions take place and people change their views, combine thoughts, explore where there is consensus, and where there isn't. No problem.

But as soon as that group gets larger than 10 what happens? The most frequent thoughts are mistaken as the most important ones. In the margarita example, the waiter would never be able to discover the “margarita” thought with a large group of tens or hundreds, even if they used the highest standard of survey or polling technology. Every survey output, word cloud generator, person paid to codify feedback, and advanced natural language processing algorithm that clusters similar thoughts would do the same thing: Inaccurately emphasize that BEER! and words similar to beer are by far the best and most loved simply because that term was most frequently shared as a “first‐thought‐best‐thought.” When, of course, given the chance to participate in a conversation and consider the thinking of other people, no one in the group above even ordered a beer.

The curious thing about “margarita thoughts” is that, in a small group, this phenomenon of surfacing important ideas, instead of counting the frequent ones, is just so obvious.

Picture any meeting or group discussion you've had with six or seven people. As you try to solve a problem, everyone shares ideas as others react. Eventually, someone shares a thought that many people resonate with and that becomes an important thought that guides the actions you all take together. This is an extremely common and standard way of people working together. It's a conversation.

Now picture, in that same meeting, someone was listening to the conversation and counting the frequency of ideas shared. After the group arrives at agreement to take an action, that person interrupts and says: “Sorry, we can't go with that idea. It was only shared once by Julie and there are several other ideas that were shared more often early in the conversation. We need to use one of those earlier, frequent ideas. They are more important.” That is obviously ridiculous. No one would do it. But maybe they would. And more importantly, maybe you would. In fact, you probably do. I'll explain.

Mistaking frequent thoughts for important thoughts is how most organizations inform decisions affecting many people. The annual survey run through HR is now a staple of almost every organization in almost every sector. These surveys are supplemented by special issue surveys/polls/pulses on topics such as Diversity and Inclusion, Culture and Professional Development. Some organizations even “pulse” weekly to measure everyone. During this now familiar process, a quantitative survey of questionable scientific value is sent to a few hundred or a few thousand people and the results are, let's be honest, hard to interpret. And for good reason.

I recall speaking with one leader who had the opportunity to talk with an employee who had anonymously provided an extremely low mark on their internal NPS survey. The question was phrased something like this: On a scale of one (low) to ten (high) how likely are you to recommend our organization to a friend or colleague as a great place to work? The employee gave an extremely low rating. Fortunately, they spoke up as the results were discussed by the team. “I gave that a one because all of my colleagues already work here and none of my friends work in our industry.” They had interpreted the question as asking whether their friends would be suitable employees, not as a measure of their happiness.

In another, very similar scenario, a parent came forward after participating in a school district survey, which had asked a similar question: On a scale of one (low) to ten (high) how likely are you to recommend our school district to your friends and family? They had also given a rating of one. Their explanation: All of my local friends already attend this school district and my family lives out of town.

Ah.

 “But how do you feel about our school district?”

 “Me? I love it!”

It's worth repeating: Closed‐ended responses are hard to interpret. To remedy this, along with the closed‐ended questions most or all surveys now have at least a few open‐ended questions asking for more context and explanation. In an effort to unpack various high and low marks in their surveys, leaders look to these open‐ended responses for context. Modern survey platforms even help disseminate these open‐ended comments by sending pages and pages of open‐ended thoughts to managers and leaders in the areas of business related to the feedback. Facing this firehose of feedback, much of it directly contradictory, what do we do? Count responses. Theme them by frequency. Put them in word clouds to see which ideas are most…Common. The more frequent, the bigger they are in the cloud and the more influence the idea has.

If those people were all in the same room, you would instinctively know better than to count the number of times something was said. You would be far more interested in how things resonated with people, how they learned and changed their thinking after being exposed to the thinking of one another. You'd be interested in what emerges in a conversation as the most important ideas, which people agree on, and you would take note of the areas where people don't agree.

In early spring 2020, education leaders in the city of New Britain, Connecticut, were conducting a review of their curriculum, just as the COVID‐19 pandemic was gaining momentum. Jonathan Costa, Assistant Executive Director of EdAdvance, a Regional Educational Service Center, wanted to scale a conversation and get the district faculty's thoughts and feedback on their return to school in the fall so his team could better respond to their needs.

“I was thinking we were going to get some instructional guidance,” Costa shared with our ThoughtExchange team. However, as he quickly discovered, “If you don't feel safe, you're not going to be thinking about building a good lesson plan.”

When Costa saw the trending thoughts in the conversation with approximately 800 people, it was clear that curriculum instruction wasn't what they were looking for. “I could feel the intensity of everyone's personal concerns for health and safety—their inability to imagine how we could safely bring people back to school without a guaranteed vaccine or therapy.”

The surprising results from that online conversation gained the attention of Dr. Miguel Cardona, Commissioner of Education for the State of Connecticut. He chose to further scale the conversation about safety and the return to school amidst the pandemic to include the voices of teachers and parents across the state. Over a weekend more than 40,000 people joined a conversation where thoughts were considered by one another more than a million times. Ultimately, the Connecticut Governor, Ned Lamont, utilized the voice of tens of thousands to inform critical decisions that literally affected the lives (and, sadly, deaths) of many Connecticut residents. Schools were closed for the remainder of the year.

So, if your organization still counts responses from open‐ended surveys, analyzes text, and mistakes frequency for importance, or if you start a conversation expecting it to be about one thing and find it ends up being about something completely different, your organization has something to learn from the lonely and powerful margarita.