Counting ideas is the worst innovation KPI

"Number of ideas submitted" is the most common innovation KPI and the least useful. It measures how loudly an organisation asked for ideas, not whether anything changed for a customer, a citizen or the budget. Keep the count as a health check on participation if you must, but report flow, learning and value instead: how many ideas moved forward, what you learned from testing them, and what reached real use.

The belief, stated fairly

The case for counting ideas is not stupid. It usually comes from a leader who wants to prove that people are engaged, that the new innovation programme has reached beyond a small team, and that staff feel invited to speak. An idea count is easy to collect, easy to compare month on month and easy to put on a slide. In the first weeks of a programme it does tell you something: whether the call reached people at all.

The trouble starts when the count becomes the target. Once a unit is judged on the number of ideas it produces, the number goes up and the quality goes down. That is not cynicism; it is how people respond to any measure that is tied to recognition or a performance review.

Why it fails in practice

It rewards volume over thought. Ten ideas written in five minutes count for more than one well-framed problem that took a week of interviews to understand. People learn quickly which one the system prefers.

It hides the real bottleneck. In most organisations the shortage is not ideas. It is the capacity to evaluate, test and implement them. A campaign that produces 900 submissions and a review panel that meets once a quarter has not created innovation; it has created a backlog and a lot of disappointed contributors.

It punishes the people who took part. Every idea that disappears without a reply teaches its author not to bother next time. A high count with a low response rate is a measure of future cynicism.

It says nothing about value. The board does not care how many ideas were submitted. It cares whether the organisation is getting better at turning uncertainty into results. An idea count cannot answer that question, so the conversation drifts to anecdotes.

It distorts the portfolio. Volume campaigns pull towards small, safe, incremental suggestions, because those are quick to write. The bigger, riskier bets that need framing and sponsorship never show up in the funnel.

What works instead

Replace the single count with a short set of measures that follow an idea through its life. You do not need many. Four or five, reviewed every quarter, are enough.

1. Flow: how many ideas move from one stage to the next. Track the share of ideas that receive a decision within an agreed time, and how many move to testing, to pilot and to implementation. Flow tells you where the system is blocked.
2. Response time: how long contributors wait for an answer. A clear "no, and here is why" within two weeks does more for culture than a "maybe" that lasts six months. This is the measure that protects participation.
3. Learning: how many assumptions were tested, and what changed as a result. A team that ran six small experiments and killed four ideas has done better work than a team that built one pilot nobody asked for. Count tested assumptions and decisions made from evidence, not prototypes built.
4. Value realised against value expected. For ideas that reach use, record the benefit you predicted when you approved them and the benefit you can show now: time saved, cost avoided, satisfaction, revenue, risk reduced. Be honest about the gap.
5. Portfolio balance. Look at where your active work sits: improvements to today's services, extensions into adjacent areas, and a small number of bolder bets. A portfolio that is entirely incremental is a choice, and leaders should make it knowingly.

If you need to show participation, show it as reach and quality, not volume: how many units and roles contributed, and how many contributors received a response. That keeps the original intent of the idea count without letting it run the programme.

International guidance points the same way. ISO 56008:2024, part of the ISO 56000 family, gives guidance on tools and methods for measuring innovation operations. It treats measurement as something you design around your objectives, not a list of counts to collect. You do not need the standard to apply the principle.

Illustrative example: a logistics company in Rotterdam

Illustrative. A mid-sized logistics company ran an annual idea campaign and reported proudly that submissions had risen every year. When a new operations director asked what had come out of the previous campaign, nobody could say. Most ideas were still marked "under review". Several warehouse supervisors had stopped contributing.

The innovation lead replaced the headline count with three measures. First, every idea had to receive a decision within 15 working days, and the share that did was reported monthly. Second, each idea that passed the first screen was turned into one or two testable assumptions, and the team counted how many had been tested and what they concluded. Third, for anything that reached the depots, the team recorded expected and realised benefit.

The next campaign produced fewer ideas. It also produced the company's first measurable result from the programme: a change to loading sequences at two depots, tested for four weeks before roll-out. Participation from warehouse staff recovered because people could see their ideas being answered.

Common mistakes

Do

• Agree the measures with the people who will act on them, starting from the objective and working down. A KPI tree helps you link each measure to the result it serves.
• Report stopped ideas as a sign of discipline, with the reason and what was learned.
• Review the measures every quarter and drop any that nobody uses to make a decision.
• Keep the count of ideas, if you keep it, at the back of the report as context.

Don't

• Set idea quotas for teams or individuals. You will get the number and lose the point.
• Tie individual rewards to submission volume, or rank people by how many ideas they entered.
• Launch a campaign before you have the people and time to review what it produces.
• Report "ideas in the pipeline" without saying how long they have been there.

Questions people ask

Is "number of ideas" ever a useful metric? In the first month of a new programme, yes, as a quick signal that the call reached people. After that it should become context, not a target. What matters is how many ideas received an answer and how many moved forward.

What are the best KPIs for an idea management programme? Decision time, stage-to-stage flow, assumptions tested, and value realised against value expected. Add a simple measure of reach, such as the number of units contributing, if participation is a stated objective. Our guide to how to measure innovation covers the wider set.

How do I convince leadership to drop the idea count? Do not ask them to drop it; ask them what decision it helps them make. Then show the same quarter reported both ways: one slide with the count, one with flow, learning and value. The second slide usually wins because it answers the question the board is actually asking.

Where to go next

Start by building a KPI tree from your innovation objective, then use portfolio mapping to see the balance of what you are working on. A hypothesis test turns ideas into assumptions you can count honestly, and objectives and key results help you set outcome targets rather than activity targets. For the bigger picture, read our guide to assessing innovation maturity, and see how to put these measures in front of your board in reporting innovation on one page.