Have you ever conducted remote brainstorming sessions the classic way, where a few people are genuinely engaged, half the participants sit with cameras off doing who knows what, the rest generate random ideas just to get you off their backs, and after two expensive hours for your company, you’re left with a bunch of raw ideas you don’t know what to do with?

I’ve participated in such brainstorms many times and conducted them myself just as often. The “time spent/results” ratio always seemed appalling to me.

I see several problems with classic brainstorming:

1. People don’t have time to think - the quality of ideas from all participants is higher when they’re given time to reflect.
2. There are always a few people ready to run with any ideas who monopolize the microphone, and brainstorming’s loose structure only makes this problem worse. The rest aren’t engaged, feel shy, or consider their ideas worthless.
3. But most importantly - people don’t take brainstorms seriously, don’t see idea generation as their job, and typically have plenty of other responsibilities. At best, brainstorming is a chance to have fun, but usually people overloaded with meetings are skeptical about yet another meeting with a vague agenda, and only have energy for banalities and jokes. It’s a vicious cycle: silly ideas breed unserious attitudes, and unserious attitudes breed even more silly ideas.

Meanwhile, Google Design Sprints worked excellently at our company, even remotely, producing quality solutions to genuinely complex problems. I see the difference in that GDS participants come to idea generation already immersed in the topic, with time for focused preparation, bringing thoughtful ideas to the group discussion.

I’m not proposing a new way to conduct brainstorms. Instead, I suggest doing more prepared and meaningful meetings. My method gives people the opportunity to think and allows everyone to discuss ideas together, creating synergy.

Process stages:

1. A short introductory meeting, where the facilitator prepares and presents all necessary materials;
2. Asynchronous idea generation;
3. Facilitator preparation for the main meeting;
4. Discussion and clustering;
5. Voting;
6. Decision making.

This format is largely inspired by the GDS framework, which is good in itself but designed for more complex tasks requiring a full week of work, resulting in a user-tested prototype. I highly recommend familiarizing yourself with this framework and using it when appropriate.

What I’m proposing is intended for narrower and quicker remote team idea generation sessions. But an important idea is inherited from GDS - we need a decider, someone who makes the final decision based on the idea generation session results. If in a full GDS this might be a founder, for our short sessions the decider could be, for example, a product manager. This is important: the goal of this meeting isn’t to collectively choose the best idea, but to provide quality material to the decider.

1. Introductory Meeting

The facilitator should prepare:

• Clear description of the question that needs ideas;
• Additional materials that might be useful to participants and provide context;
• Working frame in Miro (or another similar tool) containing:
  • description of the original task,
  • relevant links to supporting materials,
  • pack of stickers for capturing ideas;
  • instructions for asynchronous completion;
  • pool of red dots for voting in the final stage.

Conduct a short introductory meeting so participants can ask additional questions. This usually doesn’t take much time but significantly increases engagement. Be sure to record the meeting so participants can review it calmly.

Time: facilitator preparation - 1-2 hours, meeting - 20 minutes

2. Asynchronous Idea Generation

Participants asynchronously think about the posed question and add their ideas to the Miro board. In my experience, it works well when participants have at least a day to think about the given question, plus dedicated time before the meeting to input ideas on the board (for example, schedule 30 minutes in everyone’s calendar).

It’s important that participants bring somewhat thoughtful ideas, not just generate random stuff.

Time: thinking time - at least 24 hours, card creation time - 30 minutes

3. Facilitator Preparation for Main Meeting

You’ll need to assess how many ideas were generated, how similar they are to each other and whether they can be clustered, and understand how much time to allocate for discussing each idea. This will be difficult the first time, but with experience you’ll better anticipate discussion dynamics and estimate time.

You’ll need to decide how to structure the idea discussion. Two options are suggested in the next section.

Time: 30 minutes

4. Discussion and Clustering

By the start of the meeting, all stickers are added to the board - no time is allocated for generating new stickers during the meeting itself.

The main task of the call is to discuss ideas, cluster them, and possibly generate something synergistic. Deciding which ideas go forward is not the purpose of this meeting.

Depending on time and the number of cards generated before the call, you can take two approaches:

Option 1 (recommended): Go through all cards - the author of each card briefly explains the idea, others ask questions and suggest additions. If someone thinks one of their cards repeats or complements the discussed idea, they drag it to the discussed card. This creates clustering and reduces the number of cards requiring discussion. Comments and additions are written on additional cards and placed nearby. During intensive discussions, the working board quickly takes on the appearance of a detective’s cork board with lots of stickers and connections, but this discussion and enrichment of ideas is the main value of the meeting.

Option 2: If there are too many cards and not enough time to discuss them all, give participants time to read all cards and mark those where the idea isn’t clear, there are questions, or they want to discuss something in more detail. Clustering can be done here as in the first option. After this, discuss only the marked cards.

I don’t recommend the second option because then not all ideas get the chance to be enriched by others, which is the most valuable part of our process.

Time: about 1.5 hours

5. Voting

At the end of the meeting, participants read the enriched and clustered cards again and vote for the ones they like. Everyone takes X red dots from the pool prepared by the facilitator on the board and places them on preferred cards. You can vote for your own cards, put all dots on the same card - there are no restrictions.

How many votes each person gets (the X value) you’ll also learn to determine through experience, but to start I suggest using three votes if there are few cards, and five if there are many.

This voting doesn’t influence anything by itself - no one makes decisions based on its results. But it helps the decider get something like a heat map of ideas, drawing attention to what most participants find interesting.

Time: up to 15 minutes if there are many cards

6. Decision Making

So the decider has a bunch of interesting ideas from which they can choose those that seem promising for further development. Remember: we’re not trying to reach consensus - decisions are made by one person responsible for the results.

It would be great if the decider explains to participants why they chose these specific ideas and what will happen with them next. This way participants see that their work results didn’t go to waste, motivating them to take future sessions seriously.

Time: no time limit, but better to do this right after the meeting while discussion details are fresh in the decider’s memory.

---

Preparation, calls, decider, clustering - sounds complicated! This is certainly more complex than conducting a traditional fun (or torturous) brainstorm, but in my experience the quality of resulting ideas is an order of magnitude higher, and they much more often make it to implementation.