Supercharging decision making: 14 ways to Dory / Pulse
Share
Explore

icon picker
Supercharging decision making: 14 ways to Dory / Pulse

A simple yet impactful ritual that equalizes voices and removes groupthink in meetings.
A few weeks back I wrote about and received a lot of positive responses as well as a few questions / requests. For this post, I’ll focus on this question:
quote-left
Hey Shishir, I love the Dory / Pulse ritual. When I explained it to my team, they loved the concept but peppered me with many questions:
Is this just majority-rules democracy? Is it anonymous? And so on...
How would you respond?
For readers who are less familiar, is one of my favorite rituals from Coda. Dory is how we get our most important questions discussed and answered first. And Pulse is how we get the sentiment about a project or idea from a lot of people all at once. Together, they are a powerful decision-making tool.
I did some quick math, and in the past year, I’ve used Dory / Pulse over a thousand times. It’s a simple yet impactful ritual for supercharging decisions by equalizing voices and removing groupthink in meetings.
It’s also a ritual that is very commonly used by Coda customers and tends to be adopted (and adapted!) quickly. As I’ve worked with these teams, I’ve seen dozens of variations of the ritual—as each team has a slightly different set of circumstances and goals.
An example of our team using Dory/Pulse
So, to answer this reader’s question, I decided to catalog the many different ways I’ve seen teams use Dory / Pulse. As I polled teams, this list quickly grew from a handful of variations to 14 (BTW: if you have a good addition, send it to me!).
In each case, a team was solving a slightly different problem and thus arrived at different solutions. For example:
Teams scaling their company all-hands were drawn to techniques like and
Teams focused on product/design proposal reviews invented and
Teams struggling with hard prioritization choices landed on and
Teams challenged with moving beyond “majority rules” consensus-driven decision-making landed on and
And so on...
Many teams mix and match these techniques, so feel free to take a read and select the best parts for your specific situation.

: Ranking topics to focus your deliberations.

We’ve all been in a meeting where it seems like one person dominates the discussion with a low-priority topic, leaving no time for the group to discuss the most important topics.
If there’s one meeting ritual used more than any other at Coda, it’s what we call “Dory.” It’s a simple ranked Q&A tool that allows everyone to add questions and upvote/downvote them so that you focus on the most important topics first.
By the way, the name Dory comes from “the fish who asks all the questions.”
While the core concept of “ranked topics” seems simple and universal, it turns out that there are many meaningful variations of the Dory ritual. Here’s my list of the most common ones...

1. : “What’s most important to discuss?”

At Coda, Dory was born in our early days as a distributed team. One of our product leads got frustrated by our typical decision-making meetings where people would blurt out questions on Zoom, often talking over each other. This naturally led to obvious bias as the louder and/or more senior people dominated the discussions.
So we added a topics table with a voting button and called it “Dory.” As a reminder, it works like this:
This ritual changes our meetings — and our decision-making process — in a few key ways:
Equalizing the audience: In Coda meetings, we run through topics in order of votes, regardless of the seniority or volume of the person.
Discuss what matters: We go through questions in order, with high confidence that we're discussing the topics of most interest to the room.
Well-formed questions: We have a best practice to ask the person who suggested a question to read it out loud and offer any commentary ー this maintains the "human" element of the interaction. I've found that questions come out more factual and more respectful if they are written first.
Try it out: .

2. : “Ignoring authors and votes, what’s most important to discuss?”

This variation came from Ben Lee, a long-time Codan, and he wrote about it here: . I spoke to him about his insight:
Decision meetings can ramp up quickly. Read the write-up, add questions, vote on other questions, and begin the discussion—all in a span of 5 to 10 minutes. During that time, you’ll see a flurry of new rows, blinking cursors, author avatars, and increasing vote counts. The process feels busy and sometimes a bit overwhelming.
This causes a few issues.
First, on more than one occasion, I’ve found myself jumping to questions that were already receiving votes in case there wasn’t enough time to read all of the proposed topics. And I wondered if it would be better to just hide the votes till everyone is done adding their topics.
Second, while the point of Dory is to remove some of the systemic bias in the “normal meeting pattern” (loudest voice, etc), having the author of each topic be so visible could cause a lesser version of the same bias. Should I upvote my boss’ question?
Ben’s template quickly became a staple version of Dory used at Coda. It adds two new toggles: one that hides the vote count (and sorting), and another that hides the authors (except for yourself).
My recommendation is to use Unbiased Dory with these steps:
Have everyone add their topics.
Then toggle on Vote Counts and ask everyone to vote without showing the authors.
When you’re ready to discuss, toggle on the authors, so each person can ask their own question.
If you’re using Dory to make sure your decision-making meeting is discussing the right topics, it may be helpful to go one further step in removing the inherent bias of most deliberation formats.
Try it out: .

3. : “What do you really want to know (but don’t want to say out loud)?”

At one of our recent rituals dinners, Gayatri Iyengar from Doordash posed a question to the group:
Does everyone do Q&A at their All-Hands? Who does anonymous Q&A versus named Q&A?
This turned into a much more animated convo than expected and lasted over an hour. There were three clear camps:
Fully anonymous: One group felt strongly about anonymous Q&A. They felt that it was the only way to get to what was really on people’s minds.
No anonymity, read your question: A second group felt equally strongly in the other direction — that people should put their names next to their questions and stand by them. They viewed it as a cultural statement that it was a safe environment for people to ask the hard questions.
Anonymity with friction: And finally there was a group of pragmatists who settled somewhere in between. They wanted the primary behavior in Q&A to be attributable, but were ok with having questions submitted anonymously. . This Dory adds a little friction for the anonymous submissions by hiding a submission form under a collapsible heading.
As a side observation, it was also interesting to me that for this group, the concept of voting on topics was so directly tied to “all hands.” As they went back and forth on their opinions on anonymity, the reason became clear — the tools they used for voting were incredibly heavyweight. For example, one team accomplished anonymity through a Google Form which went to their HR lead who then copy/pasted those questions into the All Hands slide deck one by one!
Arvind Jain, the Glean CEO, gave me this example of how he runs an “anonymity with friction” Dory ritual in Coda:
For them, this achieves both purposes: a primary path where questions are named / attributed, but an easy path (with no extra overhead) for an anonymous submission as well.
And because it’s lightweight, it doesn’t have to be reserved for all-hands — it can be used in all your decision-making forums.
Try it out: .

4. : “You can only vote for 2 of these!”

This technique is inspired by one of the core rituals popularized in the book Sprint (and ) by John Zeratsky and Jake Knapp. They have a fantastic technique for a focused design exercise, but one of the fundamental rituals that they repeat many times is called “Note and Vote.” When done physically, it involves writing on post-it notes and then adding stickers to others’ notes as a way to rank the ideas. And one of the key constraints is that everyone has a limited number of voting stickers.
When I talk to teams using Dory, I commonly hear that participation can be a bit binary — sometimes no one votes for anything, and sometimes everyone votes for everything. Of course, for Dory to work, it needs to be a discerning signal on priority.
So, inspired by Sprint, I made a version that sets a maximum number of votes per person:
This format creates an interesting dynamic in both directions — it causes people to want to vote because their votes now feel like they matter, but it also prevents over-voting so the group gets a true signal on priority.
Pro tip from John:
We often adjust the number of votes to match the number of things we have capacity for. For instance, if we think we’ll have capacity to prototype three product concepts, we’ll give everyone three votes. From a facilitation standpoint, this feels better to participants than an arbitrary number, and has the same effect.
Try it out: .

5. : “Lots of interest in questions on topic X, let’s start there.”

When I think about great rituals for meetings, one of the books I reference most often is Gamestorming (by Dave Gray, Sunni Brown, James Macanufo), an end-to-end compendium of great meeting games. I have learned a lot about this technique from Dave Mastronardi who now runs the Gamestorming Group, and who also made a . When I asked him for his great techniques, one he described was “.” The core idea is similar to Dory, but with an important difference...categorizing those ideas into groups, before opening up the discussion.
This template makes a small change to Dory to mimic a similar behavior. In this case, everyone submits their ideas and votes as normal, but the driver fills in the category column, labeling the ideas into buckets. Then, when it’s time for discussion, you can focus on the view below, which is grouped by category and ranked by total number of votes in that category.
It’s worth noting that this technique gives the meeting driver a special power to use input from the Dory to drive discussion. By categorizing topics well, they can start the discussion with something like, “Lots of interest in questions on topic X, let’s start there.”
Try it out: .

6. : “I like, I wish...”

This technique is inspired by George Kembel, one of the founders of Stanford d.school. In an episode of the podcast, George gave a great introduction to his concept, ““.
So, it's almost like scaffolding for a way to give critique, because if you ask people, “what do you think?”, usually, two things happen: They want to be nice, so they just give positive things that actually doesn't help. They’re trying to be careful with the other human to not be too offensive. So, you just get nice things and you don't really know what doesn't work. Or they really go straight to the jugular like, "Well, this doesn't work, and that's stupid." And then, all of a sudden you get to the truth, but you can't receive it because it's not human.
And so, we found, how do we actually create a safe place where you can get the real feedback? And so, we scaffolded that with a ritual of “I like, I wish.”
This version of Dory allows the meeting driver to pick a set of sentence starters that encourage people to broaden the types of ideas they submit.
Note: Meeting driver will fill in Sentence Starters. Here are some common sentence starters that I’ve seen:
I like..., I wish...
Start... , stop..., continue...
Our users would love it if..., Given recent changes, it might be the right time to..., I've always wondered if...
Try it out: .

7. : My fav Dory rituals together, including AI.

For myself, I use each of these Dory techniques at different times. I’ve built myself a template of some of my favorite ones (#1, #2, #4, #5), all put together — see below. I’ve also added Coda AI, one of our most interesting new features. As you can see here, as the Dory is filled out, the AI summary auto-updates. This not only makes it easier for everyone to keep up during the meeting, but it also becomes a pretty good summary to send out after the meeting is done.
Try it out: .
Now that we’ve explored some of the variations of Dory, time to move on to Pulse!


: Structured feedback that leads to faster (and better!) decisions.

“OK before we make a decision, I’d like to go around the room and hear where everyone stands.”
This seems reasonable—after all, making a decision without everyone’s input seems like a bad idea. But in practice, this is not only time-consuming, it is susceptible to clear bias. If the first person agrees with the plan, and the second and third do as well, what are the chances that the fourth will speak up to dissent?
Another tool used frequently at Coda is Pulse: a “pulse check” that gets used to quickly see where everyone stands.
The Pulse table starts filtered so people can only see their own entry, and once everyone is done, the toggle is flipped to see everyone’s Pulse. This way, the decision maker has a way to see unbiased feedback and draw out latent concerns.
Pulse is a particularly powerful ritual, and the variations are vast. [FYI, the most controversial question tends to be whether this leads to a “democracy rules” consensus-driven culture—for which I would recommend reading ]. Here are a few variations of the Pulse ritual...

1. : “Where do you really stand?”

As a reminder, here's a quick example of how it works:
Like the Dory, this ritual has a big impact on our meetings:
Avoid groupthink: In going around a room to hear what everyone thinks, each person implicitly biases everyone who speaks after them. With the Pulse, reflection is done privately first, which leads to more honest and independent thinking.
Equalize hierarchy: A private Pulse check allows everyone to express their opinion at once without influence from the vocal minority.
Contextualize feedback: The Pulse clarifies the difference between "I really love this idea, and I'm asking tough questions to make it better" or "I really dislike this idea, and I'm asking tough questions to expose why.”
Draw out the latent concerns: The Pulse tends to give people space to describe what's on their mind, and I often find that it affects the agenda.
Try it out: .

2. : “Where does each part of the decision-making team stand?”

The most common question I get about Pulse is:
If you’re taking everyone’s opinion — does that mean it’s a democratic system where the majority decides?
My clear answer is: No. Pulse is about quickly gathering input, but it should assist the decision-maker, not make the decision.
When I talked to Surojit Chatterjee, the former CPO of Coinbase, he gave me a practical way to reinforce this idea. He said that, at Coinbase, they used a role-based decision process called RAPID, where each member of the decision-making group fit into one of five roles: Recommend, Agree, Perform, Input, Decide. And they decided to bring that role-based framework directly into their version of Pulse:
You’ll notice that there’s a new column for the role of each person giving feedback. I think Surojit did a very nice job of taking the spirit of open input while still reinforcing the roles of various disciplines. You can clearly weigh the input of each role differently, and, at the end of the day, the feedback next to the Decider is the final result.
Pro tip from Surojit: He would often put placeholders in the table with exactly the type of feedback he was looking for from each person. For example, he may want the Head of Finance to give an opinion on the budget and the marketer to give an opinion on the GTM timeline—each as key pieces of input to the final launch decision.
Try it out: .

3. : “Do you agree, and do you commit?”

Zac Hays, CPO at Luxury Presence, documented a great Pulse technique in his doc on . His primary insight was that in many decision-making processes, it’s both important to gather unbiased input and also to get explicit commitment after the decision has been made. In his view, it’s important for detractors to explicitly “disagree and commit.”
Pro tip from Zac: Have everyone fill in all parts of Pulse except the commit column. Then have your discussion and make a decision. At the end of the process, ask everyone to explicitly commit to the decision by filling in the last column.

4. : “Did you hear each other’s feedback?”

This technique is derived from a ritual Wade Foster, CEO of Zapier, implements in his staff meetings called . The main observation is that a well-constructed Pulse can lead to a collection of fairly deep opinions of your decision-making group. There’s no point in everyone writing down their view if we don’t take the time to read and listen to each other’s input. To make this process simpler, this version of Pulse adds a Read? reaction to every row and default filtered the table to only show the rows that each person hasn’t read. After you’re done submitting your own feedback, you then have a “feed” of Pulses from each person to read:
We adopted this in our own decision-making meetings at Coda. It’s a small change but it had a massive impact: not only does everyone feel like they have a chance to give input, it greatly increases the chances that that input is received as well.
Try it out: .

5. : “Do you prefer {option 1, option 2, or option 3}?”

Sagnik Nandy, President at Okta, told me about his pet peeve in decision-making:
The worst is when a reviewer rejects a proposal without an alternative suggestion. So we have a rule on my team at Okta: you cannot say NO without either proposing something new or endorsing another alternative.
I love this principle! While some decisions are reacting to a single proposal, it’s actually more common that we have multiple options to decide between, and it’s healthy to encourage reviewers to be specific on their preferred option. In those cases, I like to switch the sentiment column over to a select list of choices:
Once everyone has submitted their input, I feel it’s helpful to drive the discussion by pulling up the choices table and comparing the reasoning for people who selected the same option.
Try it out: .

6. : “Rank the options!”

This is another favorite from Dave Mastronardi’s . Similar to #5 (Pick an Option Pulse), this is most useful when there are multiple options. But in this case, instead of asking the group to pick just one option, you ask for a strict stack ranking between them. This Pulse will enforce that each person can only pick a single option, 1st, 2nd, 3rd, and so on:
Pro tip: If you want to remove bias in the voting process, you can configure each reaction column to display None and hide the Score column until you are ready to discuss.
Try it out: .

7. : “You have $100 to spend...”

One of my favorite Pulse rituals came from the . We would generate a list of options to pick from and give everyone a theoretical $100 budget to spread across each option. The key was that, along with your allocation, you needed to provide a reason.
I find that this technique forces the clearest prioritization and it quickly identifies misalignments. As an example, above, it’s clear that there’s one outlier on holding the event in LA, and it gives that person a chance to clearly explain their reasoning.
Try it out: .


Frequently Asked Questions

Here are a few more questions I’ve been asked about the Dory & Pulse technique:
Why do you have people read out their Dory questions?
TLDR: It’s more human. And it’s a chance to convey additional context.
As background, we didn’t start this way. The driver would go through the Dory, read out the question, and try to give a response. You can picture the meeting—it feels like the driver is the only one talking (quite the opposite of the feeling Dory should give). So we started asking the person who wrote the question to read it out. It definitely made the meeting feel less robotic, more human. But it had another interesting benefit—often the person would read the question, but add a little bit of context. It might be something that had changed in the discussion since they wrote the topic (“Well, I originally wanted to ask about X, but we already discussed Y, so I’d rather focus on this element of X”), or just some sentiment that wasn’t well captured in the question (“Just to be clear, I’m very happy about this direction overall but I would like to know more about X”).
Try it. It’s easily the most important tip I give for making Dory work in your team.

Does Dory / Pulse lead to an overly “democratic” / “just count the votes” decision-making style?
The single most common question I get about Dory / Pulse! Are you encouraging a “just count the votes” decision-making style?
Short answer: NO. That’s a very lazy interpretation of Dory / Pulse.
Remember: The alternative to Dory / Pulse is people barking out questions and opinions (with the loudest voices rising to the top and the others being suppressed). The goal of Dory / Pulse is to provide a way to gather topics and feedback in an organized, efficient way. How you then make a decision from that is up to the Driver.
I’ve definitely seen Drivers take the lazy path—“OK, looks like 3 votes to 2 for Option A, so I guess we’ll do that”. Frankly, without a Dory / Pulse, they probably would have done the same thing, except they only would have heard 2 of the opinions instead of 5.
A few tips for Drivers to make sure that Dory / Pulse is an efficient organization of input, not a substitute for clear decision-making by the Driver
Pre-seed the Dory with topics and/or pre-seed the Pulse with the types of topics you want each person to weigh in on. See for some ideas.
Categorize and override: Following the template, categorize all the input and then drag them to the order you want to discuss them in. It’s totally fine to say “5 of these comments are about Topic X, but I’m going to ignore that topic because I think Topic Y is more important.”
Be clear that your Pulse is the final say (and write it after everyone else is done): is a good starting template for this. Run the Dory / Pulse process and once the discussion is done, go to your row in the Pulse and say “OK, I’ve heard all the input and I’m going to write my decision down here”

How do you make sure everyone actually reads each other’s pulse?
First off, this is a real issue in many cases—and definitely worth addressing. You get everyone to write thoughtful Pulse’s. And then immediately jump to discussing. But no one actually reads each other’s viewpoint.
My suggestions:
Carve out time for this: “We’re going to take 5 minutes for everyone to write a Pulse and then another 5 minutes to read each other’s.”
Make an explicit “I’ve read it” action: See the for a template.
Can I use Coda AI to help with Dory / Pulse?
Yes definitely!
Two highlight use cases for me:
Summarization: See an example in . You can do something very similar for Pulse.
Categorization: See an example in . Let AI help you group topics together.
Does Dory / Pulse work for multi-choice decisions?
Some discussions are “yes/no” but the vast majority of discussions have multiple options. Depending on what I need, I use one of these 3 templates:
: When I just need a signal on a single choice from each person (“I prefer Option A”).
: When I need each person to give a comparative opinion (“I prefer Option B, then Option A”).
: When I need each person to also give a reason and a weight for their opinion (“I prefer Option B 50 times more than Option A, and here’s why”).

How do you handle anonymity?
To be honest, in most situations, anonymity isn’t required or desirable. Especially in small trusted groups, Dory / Pulse should be naturally attributed to encourage direct discussion and feedback. But in cases where anonymity is required, I pick between these options:
If I just want anonymous voting but the discussion can be attributed, use .
If I want the actual topics to stay anonymous, then use .
There are also some advanced techniques in the page (e.g., if you want to pre-approve questions, etc.).
When do you use inline comments vs Dory?
This is a common trade-off. For teams used to “1-way writeups” (if you’re unfamiliar with the term, check out Lane Shackleton’s great piece on ), inline commenting is a natural behavior.
My general recommendation is to treat them differently:
Use inline comments for contextual questions/suggestions/clarifications that the author an handle async.
Use Dory for the list of discussion topics you want to see prioritized, discussed, and resolved.
For example, you might use an inline comment for “I’m curious, what’s the image we used for this test?” but use Dory for “I think the results of the variation A vs B test are not positive enough, should we try variation C?”
Does this work asynchronously?
ABSOLUTELY! One of my favorite side effects of Dory/Pulse is that most of the interaction can happen asynchronously.
In fact, we’ve incorporated this into our main decision-making forum, Coda Catalyst. David Kossnick wrote about it in :
Often the Driver will send out a pre-read, gather and address feedback async, and gather enough input to make a decision before the meeting has even started. In those cases, the Driver is encouraged to cancel the meeting—this is a win!
What if people don’t contribute to the Dory?
As powerful as the Dory technique can be at equalizing an audience, it only works if the team participates. Baljeet Singh (product exec from Google, YouTube, Invoy, Livongo, etc) had a great suggestion:
At one point, we were suffering from the fact that a lot of the folks in the field didn’t feel like they had a voice and very few felt empowered to weigh in on product feedback. There was also a bit of intimidation when the CEO was on the call. So we did a few things: 1) actually required that each member contribute at least one question of feedback on the roadmap / plans and 2) ensured that each question / feedback item received some recognition or response from either the head of product or the CEO. We actually wouldnt end the meeting until every member had contributed something.


Give it a try!

I’ve included templates for each of the rituals in this doc, feel free to get started with or .
Huge thanks go to the following folks for reviewing and contributing to this doc. Their comments have made it immensely better (and any remaining shortfall is entirely my fault!): Aagya Mathur, Abishek Viswanathan, AJ Asver, Ankit Jain, Arianna Huffington, Ariel Bardin, Arvind Jain, August Bradley, Baljeet Singh, Ben Lee, Bukola Ekundayo, Ceci Mourkogiannis, Dave Mastronardi, David Singleton, Diya Jolly, Gayatri Iyengar, Gokul Rajaram, Ilyssa Russ, John Zeratsky, Karandeep Anand, Lenny Rachitsky, Mads Johnsen, Nikhyl Singhal, Oji Udezue, Rob Goldberg, Rushabh Doshi, Sagnik Nandy, Sarah Guo, Sean Cosier, Soma Somasegar, Spencer Swan, Sriram Krishnan, Surojit Chatterjee, Tracy Chan, Vivek Ravisankar, Wade Foster, Yuhki Yamashita, Zac Hays.
And if you want to learn about more rituals, head over to and/or signup for my newsletter at .

Share
 
Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.