How to use cross-team guilds to accomplish more than you ever thought possible
NOTE: I wrote this in February 2018 when I was working at Wise (formerly known as TransferWise). It was originally meant as an internal guide, but is applicable in any organisation with perhaps 150+ people.
It outlines the two key roles and clarifies the types of problems cross-team guilds can and cannot solve.
In a fast growing company, there’s an endless list of problems you can choose to throw yourself into. So many things could be better than they are. We’ve developed this concept of a “cross-team guild” that’s helped us do more than we could if we only relied on our teams. How to make them work isn’t well understood even within Wise. So this is my attempt to explain how and why they work.
A cross-team guild is a small group of employees, usually no more than 5-7, usually from different teams and offices, who work together to solve a business problem.
There’s a single person or small group (not more than three) who compose the “core” of the guild and other members who help out, particularly with execution.
Guild members commit 5-30% their time from their main role towards the guild’s objectives. They’re not created from above (like a task force) but emerge from the business as needed. Some are long-lasting, almost permanent virtual teams, others work feverishly for a few months until they’ve solved a particular problem and then dissolve.
Over the last four years I’ve been fortunate to have created and contributed to more than dozen cross-team guilds in Wise. They’ve covered a wide variety of topics: from managing our analytics stack, to team-planning, to leadership development. Some have failed spectacularly. I’ve helped bring back some from the brink of death. And others have wildly exceeded everyone’s expectations and are now central to how Wise works.
If they’re set up and run correctly, cross-team guilds can do things a team never could, with less effort and higher impact. So let’s now dive into how I’ve seen them work best.
The first lesson is to start small. You’ll eventually do tons of good work, but just like with a startup, you need to MVP it. Figure out the core problem you’re solving (for your customers) and solve it. It should feel like a small side project, not your day job.
Start with a small group of people. Search for people who are extremely motivated to solve the same problem you want to solve. A cross-team guild is an extra responsibility on top of their day job. You can’t strong arm them into helping you (not that this works anyways). Don’t get people who will help you just as a favour; their motivation tires fast.
Then help the guild members be successful.
You shouldn't put too much effort into setup. Treat it as an experiment: do the minimum to get it going and see if it’s worth anyone's time. But your ragtag group of initial guild members will need three things to have an initial impact:
A clear scope and purpose for their work. Those helping you will need to justify at least to themselves - but maybe to their lead/manager as well - why they’re spending time on this. Make this easy.
Tasks that are clearly aligned to that purpose with short-term measurable impact - they need to see their extra efforts pay off with clear impact. If they have even a small but clear impact, they might be willing to make a long-term commitment. But if not, there’s no chance.
Hold them accountable. Participating in a cross-team guild is an extra responsibility above what their team have asked them to do. But - assuming you have a clear scope and purpose defined - the work you do in the guild is important for the company. Guild tasks often fall into the “important but not urgent” category and hard to prioritise above “urgent but not important”. Guild members will need reminders about why it’s important and their impact. So plan the work together to get buy-in. Leave nothing unclear about ownership. Check-in regularly on how they’re progressing. Remove any roadblocks they encounter.
If you get the initial parts right, you’ll be on a good track. Most cross-team guilds I’ve seen not work out have failed to do these three basic things. That said, there are a few other pitfalls to watch out for.
Don’t ask for too much – A cross-team guild is essentially a volunteer organization. You’ll need to be patient spinning the team up. Be patient and wait for the right people to be sufficiently motivated before recruiting them. For instance, we wanted to build a guild around wellbeing (managing stress in a chaotic environment). We especially wanted to help offices far abroad where we weren’t, like in Tampa. But there was nobody in our Tampa office who was passionate about that topic. So we skipped Tampa for the first quarter. Make the case (with a clear vision) to people who want help from, but accept that it might not be a priority for them.
Don’t let standards slip - just because it's voluntary it doesn't mean its poor quality. In fact, it has to be the opposite. Quality contribution is even more important because you simply can't waste time. You have no dedicated resources. The default is to not do it. Use that constraint to focus your attention on doing only those activities that will clearly impact.
Don’t go broad - It’s tempting to try to change the world with your cross-team guild. But not only are you not going to (especially at the beginning), if you think you are, this will cause you’ll fail. But, you can absolutely deliver a small tangible impact if you define your scope correctly. Focus on solving a small problem that you and your colleagues face every day. Then, build your ambition over time as your list of impacts stack up. If you’re successful, others will be interested in helping you. Then, you’ll have a shot at changing how your company operates and what it’s able to do.
So, once your cross-team guild is set up, how should you work together as a group?
There are two groups in any guild. There’s a “core” person or small group and then there’s the other members of the cross-team guild.
Guild members will be taking actions in their own office or in their own part of the organization.
So, how do they work together? First they plan their activities and create commitments as any other traditional team would. In Wise, all teams plan quarterly and so do our cross-team guilds. Second, the core team will support before (to set commitments and mentor on setup) and after (to reflect on the outcome, and plan next steps). But during the activities guild members usually work independently.
There’s always a core group of individuals who lead the guild and take on a special responsibility. To lead the guild, this core person or team will need to do three tasks. They may also execute on the plans as would any other guild member, but these activities are what they need to do as leads. The guild won’t work if these things are not done well so it has to take priority. These are Coordinate, Clarify, and provide some Core Capability.
Coordinate – By its nature, the cross-team guild is loosely knit, so over-invest in communication. Clear and frequent communication is also the only way to hold guild members accountable. This is the most time consuming role for the core part of the cross-team guild. There’s going to be tons of activity going on if you’re effective as a cross-team guild; work hard to keep up and keep everyone in sync.
Clarify purpose and KPIs - everyone who has a day job will find it hard to prioritise this work. Reinforce and clarify why you’re doing this at all and take time to clearly measure the impact.
Core capability - if you’re starting or leading a cross-team guild, you often have more experience on this topic. You’ll need to guide others. That may mean doing some more technical tasks, writing best practice docs or just mentoring others.
Execute. They need to make whatever you’re doing as a cross-team guild happen. If you can’t execute, then you’re not a guild. You’re a discussion group. So what does it mean to execute? Its three things:
Commit to action - Every member of the guild needs to commit to doing something with a tangible output.
Develop the approach - The guild members need to figure out which actions to take, and how to run them. There are often many approaches possible. For instance, for a topic like burnout: is it better to get an external expert to speak? Or an employee who’s been through it? Or bring in experts to talk 1-1 with people. Or make a document? The guild members need to make a decision on how best to deliver the content.
Carry out - Then they need to do it. They need to take the concrete actions they committed to and that they figured out was the best way to do it. This is the most time consuming part for cross-team guild members.
Whatever work the guild members do is their own. They own it. For instance, if Britta hosts a Lunchwise (gather employees together to listen to an external speaker) in London, that’s hers. She’s figured out how to make the event work and she made it happen. Publicly celebrate all that they’re doing and motivate them to do even more. The only thing the guild has helped with is to figure out how to help Britta to do more events, and to do them a bit better. The core guild can build her confidence that she's having an impact (through mentoring, sharing insights about needs, showing impact on KPIs, etc.).
As a final point, guilds are ideally suited to certain types of problems. They tend to be a particularly good in two scenarios:
Diffuse problems. The type of problem that’s not worth any one team’s time but if you add it up across the whole organization it’s a big problem. Where the threshold of inaction prevents a team from forming around it. For instance, business intelligence tools are used across the business, but no one team uses them a lot. So maintaining these tools and making them work well for Wise is up to a guild called “self-serve data.”
Testing whether a team is needed - The second scenario guilds work well is when you don’t know if a dedicated team is needed or not. A cross-team guild is a good way to figure that out fast, without much cost.
There are also a few scenarios where guilds are inappropriate:
When a lot of people need to be involved (more than 7-10). Instead form a proper team around it, or set up multiple guilds - with different leads - who can work independently.
If you don't have the resources/capabilities to deliver something valuable with the resources in the guild. If you need engineering capability to have a meaningful impact, and you don’t have that, a guild won’t help you. You shouldn’t ever hire someone external into a guild: guilds repurpose existing energy only.
When you can’t lead it properly. If you can’t (or don’t want to) commit your own time, can’t clarify the problem, can’t manage the communication, then you can’t be successful. You’ll just waste your own and other’s time.
Why a group of cross-team volunteers, working part time, can help an organization achieve far more than traditional teams if set up correctly.
A cross-team guild is a small group of employees, usually no more than 5-7, usually from different teams and offices, who work together to solve a business problem. There’s a single person or small group who compose the “core” of the guild and other members who help out, particularly with execution.
Guild members commit 5-30% their time from their main role towards the guild’s objectives. They’re not created from above (like a task force) but emerge from the business as needed. Some guilds are long-lasting, almost permanent virtual teams, others work feverishly for a few months until they’ve solved the problem they wanted to and dissolve.
Cross-team guilds really shouldn’t work. You’re a group of volunteers – as a guild lead, you can’t tell anyone what to do. You’re usually spread across teams and offices – so you barely see each other. And you each give 10-30% of your time – so it’s hard to keep focused.
And yet, they work wonderfully well if set up correctly (see part 1). I’ve been a part of more than dozen cross-team guilds in TransferWise across a wide variety of domains.
So how can that be? I’ve found six reasons, which I’ll walk you through next.
This sounds like a problem, but it’s actually a feature. There’s a high bar for impact for a cross-team guild. You’re borrowing all of your resources from other people and other teams. If those people are doing good work in their main role, the work they do in the guild has to be fantastic or it won’t be worth their time to help you.
If it happens that you can’t deliver impact as a guild, actually, that’s ok too. Everyone just goes back to their (higher-impact) day job. There’s little disruption if a cross-team guild stops as compared to dismantling under-performing teams.
So cross-team guilds give an organization a fluidity that teams can’t. Guilds almost never get off-track the way teams do because they just stop if they’re not working, and people go back to their regular tasks. They also can’t create politics the way a team whose existence is threatened can.
A full-time employee - if they’re any good at their job - does an astonishing amount in a year. You need to accumulate an incredible backlog of tasks before you can justify hiring an additional dedicated resource. So you mostly don’t. You wait until you and your team is too stretched. And then it takes at least half a year to create the job spec, recruit, hire, and onboard. So years can pass by the time you accumulate a sufficient need and actually hire.
Many fast-growing startups are spread out across the globe to be closer to customers. For example, TransferWise has 9 offices at the moment and several of them are small (under 30 people). So, for instance, it’d be great to hire someone to focus on analytics in Singapore, they don’t need 1-out-of-30 people to focus on analytics.
But if you set up a cross-team guild, you can often get 10% or 20% of someone’s time in Singapore. Right now. And this can solve most of the problem (and delay hiring for a long time).
It doesn’t make sense to hire someone for analytics in Singapore right now. But it also doesn’t make sense to hire someone centrally - for TransferWise that means in Tallinn or London - to cover Singapore. First, in terms of timezone, because we’re 6-8 hours behind. Second, because we don’t really know what it’s like in Singapore. We’ll do a worse job of understanding their needs. So a 10-20% commitment from someone - in the Singapore office and supported by the analytics guild - is beneficial.
When organizations get to scale, there are hundreds of employees, a bunch of offices around the globe and dozens of teams. Individuals and even small teams can get lost and even isolated. And teams that get isolated can get wildly off-track and either just waste time or in the worst case start working against the organization.
But guilds help to bridge the gaps between teams. It’s a benefit to the organization that everyone in the guild is a part of another team, probably 70-90% of their time. If you’re leading the guild, it’s a huge pro to work with people from across the org who aligned with you on a topic you care about.
You’ll assemble a team of part-timers, all with slightly different interests than you. You can’t direct them in the same way as you could if they were your team and you hired them yourself. This sounds like a disadvantage, but the benefit is that you can learn a lot from their experiments. Then you can rapidly spread the good practices that emerge around the organization.
The jobs that people are hired to do often either get boring if the org is changing slowly or change beyond recognition if the org is changing quickly. Participating or leading a cross-team guild helps employees test out new career paths more and sooner. This is win-win-win for the employee, the guild and the organization.
Sometimes, this shift in focus will lead ambitious employees to move completely into to a new area of the business or form a permanent team. This is how I ended up moving from leading analytics to leading one of our people (HR) teams. And this is how half-a-dozen of our analysts moved from Customer Support into being full time analysts in one of our product teams.
But switching career paths is not the main goal and overall doesn’t happen that often. The main motivation for someone to join a guild is to solve a problem that bothers them. By joining the cross-team guild, they’re able to do it.
Cross-teams guilds are a super power we’ve developed in Wise. They allow us to solve problems years faster than if we hired dedicated resources. They shut down automatically when they’re not needed and allow best practices to spread around the organization with lightning speed. They give a new challenge to employees who join a guild and build bridges across teams and offices.
And fortunately, they’re not hard to set up: all it takes is a handful of motivated people, a clear scope and some alignment and coordination. Then they just need to start small, focus on quality and being patient to let it grow naturally as success stacks up.
So, if there’s a problem you think a cross-team guild could help solve, go do it. Let me know how it goes and reach out if I can help clarify anything. Let me know at mcclelland (dot) jeff at gmail.com