📘 High-velocity decision making

Happy Wednesday!

In case you missed it, in last week's newsletter, we kicked off with blind spot bias, talked about what to expect with the 50 biases we’re covering, and shared a nifty poster highlighting those biases.

Today we’re covering:

• 📈 Upskilling team decision making: Speed

• 🧠 Bias 2/50: Groupthink

• 🔄 Working in 'nominal groups'

Upskilling team decision making: Speed

There are plenty of resources available on how individuals and leaders can improve their decision making skills. The Blueprint will focus on how teams and organizations can build a foundation for sound decision making regardless of who participates.

How can we measure effective team decision making?

Looking at how a team or organization decides, we can break down decision making proficiency into three groups: speed, quality, and communication. We'll cover the spicier subjects of quality and communication over the next couple of weeks, but how do we know when to move fast and when to move slow?

“Speed matters in business – plus a high-velocity decision-making environment is more fun too.” - Jeff Bezos

Tools to improve decision making speed may be the most familiar, and many of the popular techniques were made popular at Amazon and through other influential, celebrity CEOs. Reid Hoffman’s Chief of Staff said that his first principle of decision making was ‘speed’

Most of the tools work to combat our strong desire for perfect information due to the ambiguity effect and our bias against risk taking. For example:

• The 70% rule: You should make your decision when you have 70% of the information you need in order to come to a conclusion.

• Decision types: Amazon categorizes decisions into two types, one-way and two-way door decisions. As their name suggests, two-way door decisions are reversible and should be taken faster than one-way door decisions that are much more difficult, if even possible, to unwind.

McKinsey, a leading strategy consulting firm, also recommends categorizing decisions by size and complexity to better understand what level of rigor is necessary in the process.

Outside of Amazon, many high-performing organizations adopt techniques to increase the speed of decision making, and though these learnings from Amazon tend to dominate the blog post content, here are a few more frameworks that we think are worth digging into:

OODA Loops

Developed by military strategist and U.S. Air Force Colonel John Boyd, the OODA Loop framework was made popular by Reid Hoffman after he claimed winners in business simply have faster OODA Loops.

“When you blitzscale, you deliberately make decisions and commit to them even though your confidence level is substantially lower than 100 percent. You accept the risk of making the wrong decision and willingly pay the cost of significant operating inefficiencies in exchange for the ability to move faster...

...These risks and costs are acceptable because the risk and cost of being too slow is even greater.” ― Reid Hoffman

The time-accuracy tradeoff

Annie Duke, author and cognitive-behavioral decision scientist, has a very practical book on decision making that covers decision speed (here’s an excerpt from her book ‘How to Decide’).

Though it's from an individualistic decision making perspective, you can see where aspects of this can help teams identify decisions that can take a faster or slower approach to decision making.

Particularly the 'final step' in the process that helps guard against the pursuit of perfect information. We tend to use more data and information as a security blanket. Annie Duke suggests explicitly identifying information that could change your mind and if so, understanding if it's worth getting.

Authority: Defining roles and responsibilities

Often for teams, no individual is always the primary decider for each decision. Otherwise, this would create a bottleneck. For speed in this environment, we need tools to clarify 'who does what?' in the decision making process.

We'll start this one with a core belief in the Blueprint: Consensus doesn't actually exist and the mirage of consensus-driven decision making is counterproductive and slow. The desire for consensus breeds complacency and complacency kills businesses.

The Blueprint takes the 'consent and commit' approach. The goal here is for teams to make the answer to the question 'who makes what type of decision under what circumstance?' clear and explicit.

"At many companies, decisions routinely get stuck inside the organization like loose change. But it’s more than loose change that’s at stake, of course; it’s the performance of the entire organization...

...The most important step in unclogging decision-making bottlenecks is assigning clear roles and responsibilities." - Harvard Business Review

There are two primary frameworks that are actively used to help teams clarify roles and responsibilities in the decision making process:

• DACI: an adaptation of the RACI roles and responsibilities matrix said to have been developed at Intuit, this framework is used at Atlassian, Twitter, and others to clarify roles in the decision making process.

• RAPID: This is a decision making framework developed by the management consulting firm Bain & Company used by Coinbase and LinkedIn among others.

Bias 2/50: Groupthink

When we all think alike, no one thinks very much. – Albert Einstein

Groupthink is our tendency to desire conformity over our own convictions. This can inhibit groups from making well informed, potentially contrarian decisions.

But group decision making isn't bad. In fact, research suggests that groups have the capacity to make better, more informed decisions than individuals. The reason isn’t surprising - there is more creativity, intuition, and experience to draw from.

But we are social and tribal creatures. We like to get along with one another or find 'our side' and vehemently oppose our challengers - neither of these are particularly effective in decision making.

So why do behavioral scientists warn against discussing options in groups? Because due to the effects of groupthink, groups tend to converge in two ways - to the mean, or to extremes.

According to Olivier Sibony, an author and thought leader on strategic decision making, groups tend to converge due to information cascades driven by anchoring bias and undersharing:

• Anchoring bias: Every new piece of information shapes our views in a linear way. Far more than we know.

• Undersharing: People share less down the chain as they fall victim to the information cascade. They are far less likely to share dissenting information than they may have been prior.

When we go around the table (or person-to-person on zoom) we fall victim to these effects without exception.

The outcome of this is a conversation that converges towards consensus around a weak, single point of view or resolves to repeated disagreements without surfacing the root problems.

To combat groupthink, behavioral scientists recommend a combination of asynchronous and in-person collaboration - typically referred to as 'nominal groups' or a modified Delphi Method.

Working in 'nominal groups'

The name comes from a method used in research techniques, but it's closely related to the mini-Delphi method used to collect forecasts from panels of experts. 'Nominal group' is easier to say, so even though the method is closer to a mini-Delphi, I'm assuming that's why it sticks.

Across all of the tools we will cover, this may be one of the most impactful changes a group can make to their decision making process for how relatively simple it is to implement.

The process is straight forward, but its effectiveness is in the details of how questions/decisions are framed and the facilitation of conversations. Here are the steps for working in nominal groups - modified specifically for decision making:

1. Frame the decision.

2. Have contributors formulate arguments on their own (surveys, docs).

3. Consolidate the feedback and surface where the group disagrees.

4. Bring the group together to discuss. Have contributors advocate for their disagreements without directly challenging the opposing point of view. The goal is not to persuade get people to agree.

5. Allow the contributors to update their points of view.

6. Summarize the outcome the the options, perspectives, and objections.

Through this process, even without a skilled facilitator you'll notice focusing on where the group disagrees keeps the time spent in conversation more valuable. There's no need to belabor the points we already agree on, yet we tend to spend an unnecessary amount of time on them. This allows the group to move faster, even on spicier decisions.

Lastly, allowing arguments and opinions to develop on their own before sharing them with the group keeps biases like the halo effect, anchoring bias, authority bias, and the bandwagon effect at bay.

We're curating a playbook of more tools like this in the Blueprint (over 30 of them so far 👀), so stay tuned!

“...groupthink is a danger. Be cooperative but not deferential. Consensus is not always good; disagreement is not always bad. If you do happen to agree, don't take the agreement—in itself—as proof that you are right. Never stop doubting. Pointed questions are as essential to a team as vitamins are to a human body.” - Philip E. Tetlock

Have feedback? Just reply to this email! It would be great to get in touch!