Your scoring is done. You have 15 ideas above the threshold, budget for three, and a review team where everyone has a different favourite. Leadership wants the shortlist by Friday.
This is where most innovation programmes get stuck — not because ideas are bad, but because nobody has a structured way to make a final call when multiple good options compete. This guide walks through a prioritisation framework, a 90-minute session that makes the call, and how to explain the decision without losing trust. Most programme owners we have worked with face this same pattern, where 5 to 15 ideas have to be reduced to a quarterly shortlist on a tight calendar.
Why is letting leadership pick a trap?
The instinct is to escalate: bring the top 15 to leadership and let them choose. It feels right — defers to authority, gets buy-in, spreads responsibility.
The problem is that leadership picking favourites without structure almost always surfaces the most politically visible ideas rather than the most impactful. Senior leaders are not closer to the problem than the people who generated the ideas — they are often further away. Asking them to choose from an unstructured list is outsourcing a decision they are not well positioned to make.
A better approach: bring leadership a structured recommendation. Here are our top 3, here is why, here is what it would take to act on each. That is a more productive conversation than "here are 15 ideas, which do you like?". The one-page innovation report is the format leadership tends to engage with most.
What is the difference between scoring and prioritisation?
Teams routinely confuse these two steps. Scoring decides which ideas are good enough to consider. Prioritisation decides which of the good-enough ideas you actually fund. Different inputs, different outputs.
| Step | Question it answers | Inputs | Output | Tool |
|---|---|---|---|---|
| Scoring | "Is this idea above the bar?" | Defined criteria (feasibility, impact, fit), per-reviewer scores | Shortlist of qualifying ideas | 3 scoring models |
| Triage | "Which ideas decline, advance, or need a second look?" | Score thresholds, time budget | Bucketed list (decline / advance / second look) | 2-hour triage method |
| Prioritisation | "Of the qualifying ideas, which 3 to 5 do we fund this cycle?" | Strategy, resource constraints, portfolio balance | Funded shortlist with owner per idea | This guide |
If you are trying to make a final fund-or-not call using only a scorecard, you are skipping a step. The scorecard tells you what is eligible. Prioritisation tells you what wins this cycle.
How does the Start, Stop, Continue, Explore framework work?
Before final prioritisation, place each idea in a broader context using four categories:
| Category | Definition | Typical examples | Resource profile |
|---|---|---|---|
| Start | Genuinely new for us; we are not doing this yet | Launch a new internal product, run a new campaign type, enter a new market | High effort, high uncertainty, longer payback |
| Stop | Stop doing something that is not working; frees resources | Sunset a low-value report, cut a redundant approval step, retire an outdated tool | Low effort, immediate payback |
| Continue and improve | Doing something we already do, but better | Faster onboarding, fewer SOP exceptions, tighter handoffs | Medium effort, medium-term payback, low risk |
| Explore | Not ready to commit; assign someone to learn more | New tech we have not tested, partnership we are unsure about, a market signal | Low cost (research only), defer the funding decision |
Most top-10 lists benefit from a mix of all four. A portfolio of all-Start ideas is exciting but hard to resource. All-Continue is safe but unlikely to produce meaningful change. A balanced split (1 Start, 1 Stop, 2 Continue, 1 Explore) is what most successful programmes converge on per cycle. Programmes that have run for several quarters typically use a similar split to spread risk across implementation timelines.
How do you run a 90-minute prioritisation session?
This works for 3 to 8 people reviewing 5 to 15 strong ideas. You need a shared list with scores, a facilitator (ideally not the person with the most stake in the outcome), and a decision-making authority in the room.
| Time | Stage | What happens |
|---|---|---|
| 0–10 min | Set the context | Review campaign objective and constraints. What were you solving? What resources are realistically available? What must be done by when? |
| 10–40 min | Rapid round | Each idea in 2 to 3 minutes. One person summarises. The group uses thumbs up, sideways, or down. No discussion. You are mapping the landscape, not debating. |
| 40–70 min | Discuss the contested ones | Focus only on ideas where the group is split. Unanimous up or down do not need extended discussion. |
| 70–80 min | Make the call | The decision-making authority decides the final shortlist. Not a vote. Not a consensus. A decision, made by the person accountable for the outcome. |
| 80–90 min | Document and assign ownership | For each shortlisted idea: named owner, concrete next step, deadline. Without these three, a prioritisation session produces a list that goes nowhere. |
Treat the agenda as non-negotiable. Sessions that drift past 90 minutes consistently produce worse decisions, not better ones. The pressure of the timer is a feature, not a bug.
How do you explain decisions to people whose ideas did not make it?
The most important thing is the actual reason, not a softened version of it.
"Not a fit with current priorities" is a real reason. "We do not have the resources to implement this right now" is a real reason. "This addresses a real problem but two other ideas address a bigger version of the same problem with more impact" is a real reason. People can accept all of these if you say them directly.
What people cannot accept is silence, vague platitudes, or finding out through the grapevine that their idea was dismissed without discussion. Honest, specific feedback pays back every time you run another campaign. The feedback that builds trust guide has the templates that work; in our experience, standardising these messages cuts admin time on idea handling significantly.
How do you balance speed and rigour in prioritisation?
Triaging quickly does not mean making bad decisions. It means deciding with the information available and the time allowed. A 90-minute session is tight, but enough if you are disciplined about focus. Separate rapid assessment (thumbs up/down) from deep discussion (only on contested ideas). This two-stage approach lets you move fast while still being thoughtful about close calls.
If sessions consistently overrun, the cause is almost always one of three things: unclear scoring upstream (so the session re-debates eligibility), too many ideas on the agenda, or no clear decision authority in the room. Fix the upstream cause; do not extend the session.
What are the most common mistakes in idea prioritisation?
Treating all ideas as equally important. "Is idea A better than B?" is hard to answer. "Does idea A fit our Start/Stop/Continue/Explore categories better than B?" is easier. Use the framework.
Letting the loudest voices win. If discussion is dominated by three people while 12 stay silent, you are not getting full input. The rapid round forces everyone to signal a view without discussion. Do not skip it.
Confusing votes with decisions. Voting sounds democratic but often produces committee thinking where nobody feels accountable. A clear decision by one person (the decision authority) is faster and creates accountability. That person can solicit input, but they make the call.
Prioritising without measurement infrastructure. If you do not track which prioritised ideas actually delivered the projected impact, your scorecard drifts over time. Close the loop a few months later: did the ideas deliver what you expected, and if not, adjust the criteria. The measurement guide covers how.
Skipping the portfolio check. Picking the top 3 by score alone often produces an all-Start or all-Continue portfolio. The Start/Stop/Continue/Explore lens forces you to balance risk before you commit budget.
Frequently asked questions
What is the best framework for prioritising innovation ideas?
For most mid-market programmes, a two-step approach works well: score each idea against a simple scorecard (impact, effort, strategic fit) to get a shortlist, then apply a Start/Stop/Continue/Explore portfolio check and a time-boxed 90-minute review session to make the final call. This combines objective scoring (to cut through politics) with structured discussion (to surface context the scorecard misses). The idea scoring scorecard covers the upstream scoring options (RICE, ICE, weighted criteria) in depth.
What if we use the framework and end up with too many in one category?
That is valuable diagnostic information. If you have 12 ideas in Continue and none in Start, that tells you something about your challenge framing or audience. You might not be inviting people who think beyond incremental improvements. The idea challenge framework covers how to phrase a question that produces a healthier mix; the 20-question diagnostic helps spot the upstream cause.
Should we involve the original idea submitters in the prioritisation discussion?
Sometimes. If a submitter can add context evaluators lack (effort estimate, the full problem the idea is solving), that is valuable. If they are just advocating, it creates bias. Decide based on what information you need, not on principle. A common compromise: invite submitters of the top 3 to 5 ideas to a 5-minute Q&A at the start of the session, then continue without them.
What if the highest-scoring ideas are not the ones the decision-maker wants to pick?
Have a conversation about why the criteria did not surface the ideas that matter most. Maybe the criteria need to change. Maybe evaluators misapplied them. Maybe the decision-maker has context evaluators did not. Have that conversation explicitly. Do not let the decision-maker override the score without explaining why; document the override and feed it back into the next cycle's criteria.
How often should you run a prioritisation session?
Match it to your submission cadence. Continuously open programmes benefit from a monthly or bi-monthly rhythm, which keeps the pipeline flowing without backlog build-up. Campaign-based programmes run a single session at the end of each campaign. Large multi-unit organisations typically run prioritisation at business-unit level rather than centrally; that distribution keeps cadence sustainable at scale.
Who should sit on the prioritisation panel?
Three to eight people covering the operational, financial, and strategic angles. A facilitator (ideally neutral), a decision-making authority, and a mix of line managers and subject-matter experts. Avoid stacking the panel with senior leaders only — it loses frontline context. Avoid stacking it with submitters only — it creates bias.
Should we use software to run prioritisation, or is a spreadsheet enough?
A spreadsheet works for one-off campaigns or programmes with fewer than 50 ideas per cycle. Beyond that, the audit trail (who scored what, when, with what comment) becomes hard to maintain manually. At that volume, you want a tool that captures per-reviewer scores, weights, comments and the final decision in one place — so you can defend a call to leadership in 10 minutes rather than spending an hour reconstructing the spreadsheet.
Related guides
- The idea scoring scorecard: 3 models for different situations
- How to triage 100+ ideas in 2 hours
- How to measure your innovation programme without lying to yourself
- How to give feedback that builds trust (even when the answer is no)
- The one-page innovation report for leadership
- Why is my innovation program stuck? A 20-question diagnostic
%2520and%2520Participative%2520Innovation-%2520The%2520Connection%2520That%2520Drives%2520Results.webp)
