Swarmwise — The Tactical Manual to Changing the World: Chapter 2 Launching Your Swarm

Many birds in the air


Ok, so you have a provocative idea. You’ve done the math. Everything appears good to go. How do you gather a swarm around the idea?

A traditional method would be to go about an advertising campaign to generate interest. Working swarmwise, though, two words about the idea of an advertising campaign: forget it. If your idea doesn’t generate enthusiasm on its own, no amount of whitewashing is going to create the grassroots activism that you need to form a swarm.

On the other hand, a swarm will form as long as you present a compelling enough idea that people feel that they can be part of. You don’t need to spend ten million on an advertising campaign. It can be enough to mention the idea just once in passing in a semi-obscure chat channel.

To traditional economists, this sounds ridiculous. But that’s what I did to kickstart a brand that’s now well known in the IT sector worldwide and has local presence in fifty-plus countries.

When I started the Pirate Party in Sweden, I took its website online, and wrote two lines in a file sharing hub’s lobby chat. This was on January 1, 2006, at 20:30 CET.

Hey look, the Pirate Party has its website up after New Year’s.

The site had a manifesto which was rough, unpolished, but which came across as credible, tangible, inclusive, and world-changing. The site itself was just as rough and unpolished – which is the typical swarm way of trial and error, of putting a stake in the ground and evolving from there:

The first website of the Pirate Party, translated for reference. It is no overstatement that it was rough and unpolished.

And that’s it. Those two lines announcing the rough-looking site are all the advertising I ever did. The next two days, the site got three million hits. (Sweden has nine million people.) Media caught on quickly too. Worldwide. On the third day, my photo was in a Pakistani paper.

My point here is, if you’re thinking hard about how to gather a swarm for your idea:

Don’t worry about advertising.

Word of mouth is much more efficient than any campaign can ever be, but that requires that your idea – or rather, your presentation of it – meets four criteria: Tangible, Credible, Inclusive, and Epic.

  • Tangible: You need to post an outline of the goals you intend to meet, when, and how.
  • Credible: After having presented your daring goal, you need to present it as totally doable. Bonus points if nobody has done it before.
  • Inclusive: There must be room for participation by every spectator who finds it interesting, and they need to realize this on hearing about the project.
  • Epic: Finally, you must set out to change the entire world for the better – or at least make a major improvement for a lot of people.

If these four steps are good, then the swarm will form by itself. Quite rapidly, in the twenty-odd cases I have observed firsthand. Very rapidly. On the other hand, if these four components are not good enough, no amount of advertising or whitewashing is going to create the volunteer activist power that you want.

Let’s take a look at sample project plans. I’ve seen many examples of all of these three types.


Oh boy I am so starting a new project t0talli for Das Lulz!!11!!!oneone!!six!!11 lololol
I wonder what I will put in it?


We are seeking a synergy between results-oriented actitivies related to dynamic business intelligence and competitive social media. Particularly, we are pursuing a path of cost-efficient achievements in quality predictability and static client satisfaction, measured by coupons used and referrals given. The means of achieving synergy is to strive for interaction with consumer focus groups in the field of cross-brand social communication and with student specialist groups in a study of networking revenue potential. The goal of the project is to raise the quarterly operating profits by up to two per cent.


We will dropkick the politicians’ worldwide war against online anonymity by deploying one million anonymizing TOR exit nodes and get the corresponding TOR client into the default-install codebase of at least 25% of browsers used worldwide by user count.

We will do this in seven stages, increasing the number of TOR exit nodes by a factor of five every 60 days. One stage of installed exit nodes will commit to recruiting five of their friends for the next stage of exit nodes to change the world in this manner. We will provide worldwide network recognition for the best contributors.

Halfway through the project, in stage four, we get the Mozilla and Chrome developers to sign up on including the TOR client by default in the code base. If completed for deployment by stage five, everybody who wants to can be completely anonymous ever after.

We are going to change the world for the better and make it impossible for the stone age politicians to put the cat back in the bag. Want to be onboard for first stage of signups? Sign up HERE (link).

Now, we need to go back to our goals here. We want to gather tens of thousands of energized activists around an idea to change the world. Having an idea is not enough; the idea and its plan must energize people.

So don’t worry about advertising. Mention your idea and plan in a couple of places where your intended activists would typically hang out. That’s enough. If it’s good, people will pick it up and talk to their friends in turn. It snowballs very quickly from there. If it doesn’t energize, no advertising is going to change that.

If your idea is good and people can contribute, change the world, and see how it can be done, then you will have the first wave of hundreds of volunteers in less than a day. You will see hundreds of people holding out their hands, palms up, at you and say, “here, use my hands! I want to be a part of this! Give me something to do!” in the electronic channels where you announce your presence.

The idea doesn’t need to be polished. The important thing is to put that stake in the ground, start attracting people, and start working your way to the goal. In this, too much effort spent polishing the appearance of the idea rather than its own merits can even be counterproductive, as people can perceive it as glossed-over corporate whitewash.

This brings us to the next problem: taking care of these hundreds of people while they’re still interested. For they all turn to you, personally, and there’s just no way you would be physically able to give them all instructions on an one-to-one basis.

Surviving The Initial Impact

When your initiative hits the ground, and it is interesting enough to create a splash, then that splash will be unlike anything you have seen before. It can happen in many ways — it can be entirely word of mouth, it can become a major story in oldmedia, or most commonly, it can hit the front page on one of many social news sites (or several of them at the same time).

When that happens, you will go from having been alone to suddenly having hundreds of people who want nothing more than to help you out on your project in their spare time. But their attention span is short; you need to respond. If you don’t, they’ll shrug and your initiative will wane out of memory in less than 24 hours.

In order to retain these hundreds of people, you also need a focal point for their interest – something as simple as a signup page or a forum. Of course, that focal point needs to be ready and functional when the impact of the idea hits, or the activists will be lost.

With the focal point active and the idea launched, it’s said that one of the hardest steps you can take in a business is going from one person to two, as you recruit your first employee. When we’re dealing with a swarm, everything is on a different scale. Here, we go from one person — you, the founder — to three hundred or more in the first instant.

It goes without saying that it can be a bit tricky, and you have at most 24 hours to sort out the situation or lose the initiative to form a swarm around this idea. What’s worse, you can’t really do it yourself. There is no way you can give individual and meaningful instructions to 300 people in the attention span you have been given.

But the swarm can do it for you, it you let it. And you must.

The swarm’s very first task will be to self-organize, and it excels at such tasks. But it is you who must set the structure and explicitly give the swarm the task to self-organize.

This is where traditional organizational theory kicks in to some degree.

Initially, you will be able to coordinate at most 30 groups, so create a discussion forum with at most that number of subgroups. You’ll likely want to have people in streets and squares campaigning for the swarm’s cause before long, so subdividing your hatchling swarm by geography works well here – and when subdividing, create at most 30 subdivisions geographically. (Most countries have administrative divisions into counties, states, etc. that vary in number between 15 and 30 units. If you’re gunning for a Europe-wide movement, you can easily observe that the size of the EU plus a few hangaround countries fits the 30-state limit well, and so on. The United States with its 50 states would be trickier, as would North America. Just pick a way to divide it into at most 30 units.)

Your discussion forum can take many forms. It can be a traditional web forum, it can be a wiki, it can be an etherpad, it can be any kind of collaborative space where people can go uninvited and just start working with others. I prefer the traditional forum because of its well-recognized form.

You’ll need to make a judgment call on the approximate resulting group sizes, based on how many hands are at your disposal. Try to pick your geographical division so that the typical size is about 7 members, and that no subgroup has more than 30 members. Don’t announce this intent, as doing so would cause a distracting discussion about that action: just create the subgroups in a way that will cause this division to happen.

If you have more than a thousand people at your disposal in this initial splash, which can happen, then 30 subgroups of 30 people each will not be enough: that structure has a maximum of 30×30=900 people. In this rare case, you may need to exceed the 30-people-per-group limit and have as many as 150 people each in 30 subgroups. This is a rare case, though, and you are not likely to encounter this.

(The magic numbers 7, 30, and 150 are deeply integrated parts of the human social psyche – part of how we are wired. We’ll return to how people behave in groups of those sizes in the next chapter.)

Having set the initial structure, you need to tell everybody to go to the appropriate subgroup and meet with other people who go there. Tell everybody to introduce themselves to one another, and to select a leader between them for the subgroup. At this point, you can safely refrain from giving instructions as to how that leader should be selected; the subgroups will come up with different ways that each have legitimacy in their respective group, and that’s all that matters at this point.

As leaders get picked by the subgroups, contact those leaders in person – at least a voice or video call, preferably over beer or coffee if you live nearby – and introduce yourself, and get to know them more personally. You’ll be working closely with them in the near future, so you’ll want to get a feel for them as people and colleagues, and to allow them to get a feel for who you are as a person and colleague.

You’ll also want to set up a subforum where these subgroup leaders can discuss between themselves and with you. Make sure that other people can read it. Don’t keep secrets; rather, let everybody see the ongoing growth of your swarm.

This process takes a couple of days, but it kickstarts the swarm on all levels. You will have energized small subgroups of people who live reasonably close to each other, and they will have legitimate leaders – legitimate to themselves, anyway. The 30 leaders and you form an initial management team pyramid in the swarm’s scaffolding of officers, the swarm’s go-to people. Taken together, your subgroups form a comprehensive coverage of all the ground you intend to cover.

(A couple of weeks from this point, you will realize that you’ll need an intermediate layer of officers in-between you and these 30 – a few of them will have lost interest and gone radio silent, and you won’t have noticed, because 30 people are too many to keep track of to that level if they don’t contact you. Therefore, you will want an intermediate layer of five or six people between you and these 30 as the swarm grows. But don’t worry about that at this stage – that’s for the next chapter, and a couple of weeks out.)

The Swarm’s First Task

As the swarm organizes into these subgroups by geography, it needs to be given a task immediately that allows it to jell properly. If you just tell people to go to a forum, they will lose interest in a week if nothing more happens. These are people who wanted to help the swarm succeed with the work of their own hands, remember?

So in order to make this organization set and settle, there needs to be something to be done right away. In the case of the Swedish Pirate Party, that task was to collect 2,000 signatures from the public to support the party’s registration with the Election Authority. It needs to be a task that looks challenging but is doable for some hundred people, it needs to be a task where you can provide for internal competition between the 30-or-so geographic subdivisions that you have created, and it needs to be a task where everybody can see the clear benefit to the swarm on its completion. In the case of a political party, registering it with the Election Authority was an obvious benefit that everybody realized; you’ll need to have a similar task at hand that leads to such a goal.

What this does is cause the swarm to learn how to work together over the first four weeks or so of its existence, as this task is being carried out in a decentralized fashion. You should update the overall progress of the goal at least daily.

A swarm organization isn’t first and foremost reporting lines between boxes on an orgchart. A swarm organization is people who know other people and who choose to work together. Therefore, getting people to know other people should be an overarching goal of your activities at this point.

Do encourage people to meet, and be very clear that they should not make it formal. Do not meet in a protocolized formal meeting under any circumstance, but meet instead over beer, pizza, and laughs. Focus on creating opportunities for people to get to know people, and for new people to feel welcome to the group.

Once such meetings become regular, it becomes even more important to make sure that newcomers feel welcome. One method of accomplishing this can be to start every meeting with an introductory round where people present themselves briefly along with some piece of trivia, such as the latest thing they downloaded or shared: “Hi, I’m Rick, 40. I’m mostly known here for setting up an ugly website. The most recent thing I downloaded was an Ubuntu Linux release.” Seeing everybody present themselves helps newcomers immensely, and it provides for a convenient framing for the newcomers to introduce themselves, as well as for the regulars to learn the names of them. Also, the local leaders will need to pay particular attention to the newcomers in every meeting, personally welcoming them back to the next meeting.

The organization consists only of relations between people. For every new relation that is created, the organization grows.

Dealing With Attention Junkies

As the swarm has its initial successes, a very small amount of people will strive to join not because they sympathize with the swarm’s goals, but because they crave and demand attention to their person, and the visibility of the swarm seems to be able to provide this to them.

As the swarm is open, you cannot and should not try to keep these people out – but you can deny them the space and spotlights they crave. It can be hard to detect them, but one telltale sign is that these people will demand attention from you personally rather than trying to build the overall swarm with people who aren’t as visible yet. You will also notice that they think very much in terms of rank and hierarchy, whereas other people will think in terms of getting stuff done and changing the world.

A few particularly tricky people will work for the swarm’s goals very hard for the first couple of weeks, and then use the built-up credibility to cash in on attention. As this happens, the transparency of the swarm is the best conceivable antidote, as such people typically depend on other people not comparing the different versions of the story they’re being told.

This part of building a swarm is inevitable, it is tough to deal with, but you can rest assured that as long as you keep the swarm open and transparent, this kind of people won’t be able to hijack it for their own, personal visibility. They will eventually flush themselves out, sometimes in quite a bit of disruption.

This is the second chapter of Swarmwise, a book arriving this summer. Did you like it? It’s going to be free to share (it, like this excerpt, is CC-BY-NC), but you can also buy it hardcover. The first chapter is here.  The actual book is expected to be available by June 1, 2013.

Leave a Reply