This article is an adapted excerpt from my book Work with Source.

Both startups and established initiatives need an approach to bring about change and development. The standard model of change taught at most business schools is to design a future state up-front and then roll out a top-down org change process that attempts to drag people along kicking and screaming until, hopefully, the destination is reached. If you’ve ever been through such a change initiative you will know how painful and ineffective they can be. Even if you finally get there, the carefully thought-out new design may not be so great in practice.

The alternative is to think of development and change as much more like tending to a garden. You engage in planting, weeding, pruning and watering. You work away diligently, season after season and watch the garden flourish in ways that might surprise you.

Creating change in this way happens by letting go of the desire to produce a grand design of an end-state, and instead work in smaller, short-term iterations. If you work in software development or modern product or service design you may recognise this more agile approach.

The change process should be opt-in so people participate if they really want to. There must be reassurance that there will be no negative consequences for anyone deciding not to participate. This means you’ll have an engaged group where everyone is treated like an adult, not a child. When others see the group having fun, making progress and learning, they are likely to join future rounds.

The change loop

The process (based on Jason Little’s Lean Change Management Cycle) is a loop with three simple yet powerful stages.

Let’s explore each step of the loop.

Step 1: Sense

The process begins by tuning in to what’s working and not working right now. Identify the tensions people experience as they go about realising their part of the vision, and seek out potentials of what might be possible by reaching higher than today.

My former colleagues at org transformation firm The Ready like to open up a dialogue about tensions with the question “What’s stopping you doing the best work of your life?” This is a great opener to surface crunchy issues that are getting in people’s way. My colleague at Greaterthan, Francesca Pick, likes to ask: “What’s stopping the highest potential being reached here?” This is a powerful question that reaches beyond the work being carried out and connects to the vision being realised.

The sensing phase can be carried out in a variety of contexts in the initiative. It works with just a small group of co-founders; a team working on something specific; or as part of a much larger meeting of hundreds of colleagues. With good facilitation, you can let all voices be heard and surface the top 3–5 tensions or potentials, perhaps by generating a long list and then voting to establish the most pressing issues. Once you have them, you’re ready to move on to exploring how they might be addressed.

Step 2: Explore

Now you can begin a creative process to explore ideas for how existing organisational elements might be modified, and consider new ways of working. They do not have to be major changes; less is often more. Look for small things that may help. You can scale up what’s working later, once you have proof.

Fortunately, you don’t have to reinvent the wheel. There are libraries of ideas and inspirations from many progressive initiatives available for free online, including: Prototyping.work, Corporate Rebels, and NOBL Academy. At the more technical end, I also like Sociocracy 3.0 which has a somewhat steep learning curve, but I love its modular, rather than formulaic approach to using elements of the Sociocracy system which has been used in highly participatory organisations since the 1970’s.

At this point in the process, things can feel exciting, and there will be a big temptation to dive right in and start changing things. A mistake I made many times earlier in my career was to say “OK, we’ve had a great idea, let’s give it a try and see how it goes.” I learned the hard way that this is unreliable and unnecessarily risky. Instead, disciplined experimentation is needed.

Step 3: Experiment

To put the change into practice, an experiment should be thoughtfully designed, committed to and seen through. This is the single best way to capture real learning and know if your efforts are actually helping or not.

Fortunately, this doesn’t have to be a drawn-out process. A one-page sheet where the experiment is designed is usually sufficient. Here are some prompts for what to include:

• The tension or potential being addressed.
• The hypothesis of what you think will help.
• What exactly you will do to test it.
• A fixed timescale for the experiment.
• Any guardrails needed to ensure the experiment can’t cause significant harm.
• Who will work on it and what their roles will be.
• The working rhythm, such as a weekly check-in call.
• Success/fail criteria: Deciding in advance (when you can be a lot more objective) how it will be evaluated.

Try to resist the temptation to skip some of these elements. Each serves a particular purpose and you won’t save time in the long-run by cutting corners — you’ll just lower your chance of learning and success. Here are some tips to help you design your experiment:

• If the team is already very stretched and has little room for more work, that itself is a tension you can experiment with, perhaps by testing what happens when you reduce the number of recurring meetings for a few weeks, or run meetings in half the time using better facilitation.
• Make sure experiments are within the authority of the people involved so they can really make it happen.
• Ensure the timescale is fixed, not open-ended, and usually limited to a maximum of eight weeks. If this doesn’t seem possible, don’t extend the timescale (or stress everyone out by taking on too much), and instead reduce the scope of the experiment. Longer timescales usually push back the real learning, and in any case, you can always do a follow-on experiment immediately afterwards.
• An experiment should always do something real. Writing a plan to do something in the future is not an experiment.
• Get the regular check-in meetings in everyone’s calendars right away and have a simple format such as: “What happened last week? What’s happening this week? Anything blocking you or support you need?” You can easily check in with four experiments in a 45-minute call. Also schedule a retrospective meeting for the end of the experiment period.
• When defining roles, have the experiment team nominate one person as the lead. They are not anyone’s boss but provide the main point of contact and coordination for check-in calls so the whole team doesn’t have to join. Additionally, give acknowledgement and credit to the person who first took the initiative to launch an experiment. Their energy for it will be a powerful asset.
• Run multiple experiments in parallel. It creates healthy competition as teams progress. But not too many.
• Encourage teams to “work out loud” by sharing what they are doing with other teams and others in the broader initiative. This creates buzz and helps recruit more people into change efforts in the future.
• Before moving ahead with an experiment, check that the whole team agrees it is safe to try. If not, find ways to reduce the risk before going ahead. It may also be wise to seek advice more broadly if there is unknown risk and other people will be affected by the experiment.

Now you’re ready to run the experiment and evaluate it at the end using the success/fail criteria that you define. This brings you back full-circle to sensing where you can explore whether the tensions and potentials have been resolved successfully. It’s common to see some tensions diminish while new ones pop up. No need to be discouraged when this happens as it’s a natural part of the process. Don’t be too quick to abandon ideas that aren’t working perfectly. Often you need a cycle or two more to refine it and iron out any problems.