Link Search Menu Expand Document
Countable Ops Manual

Making Decisions

Table of Contents

Purpose

It can be confusing how decisions are made, so we define different types of decisions and the processes we hold around them.

Scope

Document the ways in which we make decisions.

Decision-making Process

There are 4 types of decisions and they trade off between efficiency and buy-in from the team:

  1. Command decision - made by one person in charge of that area of responsiblity (AOR). This is the most efficient and fast but has the least buy-in. An example is when to send invoices. We don’t need to discuss that as it only impacts our finances probably.
  2. Consultative decision - made by one person in charge, but they must ask for feedback from their peers who are knowledgeable in that area and consider that feedback. This is probably the most common. An example is what clients to feature on our website. Julie decides but should probably ask team members what they are ready to show. Setting up a new process is usually consultative as well, someone proposes it and does the homework to get evidence and make a strong case, and gets feedback on their work doing so. When we disagree, determine who breaks ties on what we will try first, often the easiest thing to test.
  3. Democratic decision - we get the entire team’s feedback and take the popular choice. An example is our straw poll about what professional development activities to undertake.
  4. Consensus decision - everyone must agree. This has the most buy-in but is inefficient. We use it only for decisions that strongly impact everyone. As the team grows we can only approximate this as one dissenter can make a decision impossible. An example is when to host company events or what language is culturally appropriate. Anyone can generally veto these as long as they’re being reasonable.

Type 1 and Type 2 Decisions

Next and even more importantly, we have a 2nd dimension to categorize decisions: Type 1 and type 2.

You should research, and get help with Type 1 decisions, and make Type 2 decisions quickly on your own (provided it’s within your AOR).

More on this distinction

  • irreversible (also called Type 1 by Jeff Bezos) or difficult to reverse decisions. Requires committing an investment. These are more rare and great care must be taken. Hiring someone is a type one decision as it’s expensive to onboard someone. These decisions should be escalated if you come across them.
  • reversible (also called Type 2 by Jeff) decisons. Typical everyday decisions can easily be backed out. When in doubt, just guess and move forward.

For example: updating the logo on our website is technically type 2. Even though it could be very annoying on other fronts and make us confused about process if not considered carefully, it’s actually easy to change back.

Updating the logo on physical stuff is Type 1, assuming we can’t cheaply return stuff we order. This highlights that we can’t necessarily set broad categorical rules around decisionmaking, as even the same exact topic (changing a logo) can have different sorts of impact based on the context (physical printed goods).

A rule of thumb is, does it cost hundreds of dollars to change the decision back (in people’s time, etc)?

I hope this clarifies how we approach decision making and if any decision is unclear let’s categorize it in these ways.

Understanding how we operate, and having language for that, should help reduce friction and frustration.

Please give feedback! This document is a process proposal, which is a consultative decision so it may be amended if you raise anything about it.

When Decisions Are Made

  • When a decision is made in the meeting, include it in the minutes and inform any other team members outside the meeting.
  • Document the decision in this manual if it’s relevant to future work for more than one specific client.

Guidelines for Making Good Decisions

  • What is the biggest opportunity, biggest risk, and expected outcome of each path?
  • Consider whether your decision offers an incremental, geometric or exponential benefit.

Rough Consensus

The IETF has a number of principles for flat decision making, written in RFC 7282, and summarized here.

The key point is, make sure people speak up about critical factors and cases where the options are vastly different in value, and avoid discussing subjective factors or solutions which seem a little better.

Focus on the really important stuff, not small optimizations and opinions.

Straw Polls at Countable

This process is important because it lets us gauge whats important to the team while increasing buy-in on each action we take to improve.

The goal is to find the changes that will move us towards our mission most quickly.

Create a survey on slack #general using emoticon reaction. Provide 8 “straw man” suggestions that have been in the ideas section of our “Company Building” Trello board.

Script:

@channel there are a ton of things we can choose to work on internally(many in the Systems board in Trello). We want to find the ones that will move us towards accomplishing our mission fastest. However, we can’t do everything so here’s a list. Your job is to make regular yellow circle emoticon to show how you feel about each, I will exemplify. Also everyone please suggest the top thing you think we need to address that’s NOT in this list. Put an emoticon under 9 then. :smile_cat: or :smile: or :grin: or :simple_smile:, in order of how much you support it, high to low. Or use any emote that fits, like :partyparrot: if you’re so excited that you have to. Please react to max 5 of these. Please consider the items added by other team mates below this as well. (edited)

Any items with 25% or more of the team supporting it should be moved into Trello “Company Design” Board requests for discussion on how to implement. Mention all those who voted for the item in the card. Add the “staff pick” label to the card.

Items with 2 or more votes can be discussed briefly with the voters on how they’d approach it and have them articulate it better. If there’s a quick fix, go ahead.