Together with my colleague Nadine, I presented an updated version of our participatory budgeting approach. We had already shared the first version at a previous event, but since then we made significant changes that we wanted to share. A quick disclaimer: we did not invent participatory budgeting. We built on the materials from the SAFe (Scaled Agile Framework), so you will find the copyright on the relevant slides. If you want to introduce participatory budgeting in your own organization, we are happy to share our experience and always reference the original SAFe materials.
The Problem We Faced#
In Q3 2020, we had the following problem. Our portfolio (anonymized as the “fruit portfolio”) included several products, and we were organized with separate teams for each product line. These teams naturally wanted to run projects to be more innovative in the market, and for that they needed funding.
This funding was distributed by a portfolio committee that had a fixed budget of 100,000 francs (anonymized figure). Once a year, the committee would ask the portfolio what projects they wanted to pursue. Inevitably, the total sum of requested projects was about three times higher than the available budget.
What did the committee do? They looked at the requests, concluded that everyone had asked for double what they actually needed, and simply cut everything in half. Then, if someone happened to know the right people or had a persuasive conversation at the coffee machine, they might get a little more. At the end of the day, everyone was dissatisfied. The teams felt they received too little, and the committee knew they had not really distributed the money in the right way.
Discovering SAFe’s Participatory Budgeting#
We looked around for better approaches to budgeting and came across the SAFe framework. At the portfolio level, SAFe offers participatory budgeting (PB). One thing I always emphasize: SAFe is a framework that works like a toolbox. You open it, see all the tools inside, and you pick the one you need. That is exactly what we did. We took the participatory budgeting tool and adapted it to our needs.
Version 1: Our First Implementation#
For version 1, we took the original SAFe participatory budgeting material and applied it to our context. Our large organization has an overall budget, and a portion of that goes into the fruit portfolio. This portfolio has strategic goals and OKRs (Objectives and Key Results).
We reorganized our teams along value streams, following the principle of organizing around the flow of value. Each value stream sells its products within the portfolio. We then asked each value stream to prepare their epics. An epic, unlike a traditional project, has a hypothesis: something you want to test in the market. Teams had to fill out the SAFe epic template, writing down their hypothesis and how they planned to validate it.
In the participatory budgeting event, we distinguished between two budget categories:
- RTB (Run the Business): Everything needed to keep the value stream alive and continue selling existing products.
- GTB (Grow the Business): The epics for trying something new in the market to grow the business. For example, a hypothesis might be: “If we start selling plantains, we can increase our banana revenue by 10%.”
In the first round, we discussed and agreed on the RTB budgets. With the remaining money, we went into a second round to discuss which GTB initiatives (epics) to fund. Some epics did not make the cut, some were modified, and some received partial funding. The feedback from this first event was overwhelmingly positive.
What We Changed for Version 2#
After the first round, we collected feedback from participants, reflected on what went well and what did not, and sat down with management to introduce several changes.
Eliminating the RTB/GTB Distinction#
The distinction between Run the Business and Grow the Business was confusing and sometimes hindering. People tried to game the system by labeling things as RTB or GTB depending on what seemed more advantageous. We eliminated this distinction entirely in version 2.
Reducing Information Overload#
The information density was too high in version 1. Participants had to process too much in too little time. They needed more time to understand the epics from other value streams, evaluate trade-offs, and discuss options before the actual event.
Extending the Timeline#
In version 1, everything happened within a very short period: information, budgeting, and communication all in May. For version 2, we stretched the process across roughly four weeks instead of two:
- Introduction session (one month before): We explained what an epic is, how to write a hypothesis, and what a Minimum Viable Product means.
- Interim meeting: We reviewed which epics from the previous half-year had been successful and introduced the new epics for the upcoming period.
- Participatory budgeting event: Now more focused, with time dedicated to real discussion rather than information sharing.
- Communication: As before, results were communicated promptly after the event.
Pre-Calculating the Run the Business Portion#
Instead of debating RTB at the event itself, we pre-calculated each value stream’s RTB portion using an internal formula. This freed up the event entirely for discussing growth initiatives and strategic epics.
Introducing Guardrails for Epics#
We introduced three guardrails to keep discussions focused:
- Minimum viable size: An epic had to exceed a certain financial threshold to even be discussed at the event. Anything smaller could be funded by the value stream itself.
- Minimum scope: An epic had to affect at least three value streams. If only two were involved, they could arrange bilateral agreements without occupying the full group of 14 value stream leads.
- Openness to innovation: Despite the guardrails, we remained open to genuinely new and innovative ideas that deserved a hearing.
The Revised Event Agenda#
In version 1, the agenda included a management opening, two budgeting rounds (RTB and GTB), a confidence vote, and a feedback session. In version 2, we kept the management framing (opening and closing statements) but consolidated to a single budgeting round focused entirely on growth epics:
- Recap: Review of successful epics from the previous cycle.
- Forward look: Discussion and budgeting of new epics.
- Feedback and closing: With management confirmation that the agreed budget has their full support.
How the Discussions Worked#
We split into two rooms, each with about seven of the 14 value stream leads. Nadine and I moderated one room each, using a shared Excel template (adapted from the SAFe template). The spreadsheet showed the available budget, all proposed epics with descriptions, and their requested funding.
Participants proposed different options. For example: “Let us discuss an exploration option where we fund Mario’s and Luigi’s epics for new territories.” Or: “What about a momentum option where almost every epic gets partial funding?” We worked through multiple options, found agreement within each room, then reconvened in a plenary session. There, the two groups compared their proposals, discussed differences, and arrived at a unified budget.
What We Learned#
The feedback from version 2 was again positive overall, but several observations stood out:
Discussions Were Intense#
This time, the discussions at the tables were significantly more intense. Participants felt they needed even more time, and we agreed. The pre-event discussions that should have happened between value stream leads did not always take place. We need to find a way to encourage those conversations before the event.
Epic Owners vs. Value Stream Leads#
In version 1, epic owners and value stream leads were the same people. In version 2, some epics were proposed by epic owners who were different from the value stream leads. This caused confusion because the lead at the table sometimes did not fully understand the epic. The feedback was clear: epic owners should be available for questions, even if they are not the primary participants.
No Epic Was Rejected#
All epics received at least some funding. Nobody wanted to offend anyone, so budgets were reduced but never cut entirely. This was not ideal. I believe that in our next event (scheduled for May), the discussions will be harder and some epics will genuinely be stopped.
The RTB/GTB Confusion Persists#
Even though we removed the formal RTB/GTB split, the conceptual confusion remained. Some people tried to promote routine items as epics to get them into the growth discussion. We still need to improve understanding of what truly qualifies as an epic.
The Epic Template Works Well#
Everyone appreciates the discipline of thinking in hypotheses and leading indicators. The concept of a Minimum Viable Product resonated strongly with participants. One improvement request: show on the epic template which OKR or strategic goal the epic supports, so that epics without a clear strategic fit are immediately visible.
Time-Boxing Works#
Giving each group a fixed time to arrive at a budget proposal worked very well. It created urgency and focus.
Entrepreneurial Thinking#
The biggest positive surprise was the level of entrepreneurial thinking we observed. The value stream leads genuinely engaged with the company’s strategic goals, OKRs, and MOALs (Measures of Agile Leadership). They debated whether an epic truly advances a strategic objective or whether a different approach would be better. Management’s feedback was that the participants thought like real entrepreneurs. This was truly great to see.
Of course, when you discuss significant sums of money and strategic direction, emotions run high. People are passionate about their products and their ideas. Good moderation is essential to ensure that quieter voices are also heard.
Conclusion#
Participatory budgeting is something I can genuinely recommend. Our management is fully behind it. The greatest advantage is that at the end, everyone understands the budget: why we arrived at these numbers and what the priorities are. Everyone stands behind the result. This is not a budget dictated from above. It is a budget created participatively, where people have a voice and commit to the outcome. That makes all the difference.