Measuring stakeholder communication: a minimum viable plan

Start from the decision

Most advice on measuring stakeholder communication asks for more precision than the data can carry. Attendance counts, sentiment scores and dashboards arrive, and no decision changes.

Measure only what a decision will consume. Before choosing a metric, name the decision it serves: whether to escalate a relationship, whether to re-plan the work, whether to change who owns a conversation. If no decision uses the number, the number is decoration. The same discipline sits behind how to set KPIs and marketing objectives.

This matters most on small programmes. A register of 15 stakeholders and a quarterly review cannot support statistical claims. It can support consistent, comparable observations, which is enough to act on.

The 5 numbers worth starting with

This is a proposed minimum set, not an industry standard. I use it because each number is cheap to collect and each one has changed a decision for me.

Number	What it tells you	How to collect it
Coverage	The share of registered stakeholders you contacted in the period	Register checked against the communication log
Satisfaction	How content high-priority stakeholders are with the engagement itself	A 3-question survey, twice a year
Frequency	How often each high-priority stakeholder hears from you	Communication log, reviewed quarterly
Implemented feedback	The share of stakeholder inputs that changed something, or were declined with a stated reason	Decision log
Willingness to recommend	Whether a stakeholder would recommend working with your programme	One question, asked in person or by survey

The recommend question is Net Promoter, which Frederick Reichheld introduced in ‘The One Number You Need to Grow’, Harvard Business Review, December 2003. He proposed it as a predictor of company growth, built from customer surveys.

It was designed for customers of commercial firms, not project stakeholders. Treat the answer as a trend for each relationship rather than a score that predicts anything. A stakeholder who would not recommend working with your programme is telling you something specific. Use the number as the prompt for that conversation.

Baseline, threshold, cadence

Three things turn the numbers into a plan.

A baseline, recorded before you change anything. The first round of collection has no comparison target. Its only job is to exist, so that the next round has something to sit against.

A threshold for each number, agreed when you are calm, naming the action it triggers. For example: if coverage of high-priority stakeholders drops below every one contacted in the period, the next review starts with the communication plan rather than the metrics.

A cadence you will actually keep. A quarterly review that happens beats a monthly dashboard that does not. Put the review date in the calendar next to the decisions it feeds.

What the numbers cannot tell you

Coverage counts contact, not quality. Satisfaction measures the people who replied, who are rarely the quietest stakeholders. Feedback implementation rewards volume of suggestions. None of them will surface a stakeholder who has disengaged silently. Pair the numbers with one conversation per high-priority stakeholder per period and the blind spots shrink.

There is a deeper limit worth knowing about, because it applies to the whole category rather than to these 5 numbers.

Jonas, Kock and Gemünden, in 'Predicting Project Portfolio Success by Measuring Management Quality', IEEE Transactions on Engineering Management, 2013, took a longitudinal measurement of what they call management quality: information quality, allocation quality and cooperation quality, treated as one construct. They found it strongly predicts portfolio success.

The useful part for a measurement plan is what they chose to measure. Not the volume of interaction, but whether the information moving through it was any good, whether it changed how resources were allocated, and whether functions cooperated better as a result. Counting contact misses all 3, which is a reasonable explanation for why activity metrics so rarely move alongside outcomes.

Beringer, Jonas and Kock (2013) add the caution. Across 197 project portfolios they found the effect of engagement intensity to be phase-specific, and moderated by role clarity as a measure of management maturity. More engagement is not uniformly better, so a number that only goes up is not necessarily good news.

The methodological caveat matters here more than usual. Much of this research collects both the engagement measure and the success measure from the same person, usually the project manager, on the same survey. People who know a project went well tend to rate their own engagement generously in hindsight. Your own numbers carry the same bias, which is the practical argument for collecting them on a fixed cadence rather than after the fact.

Tools: use what you already have

A survey tool for satisfaction, a spreadsheet for the register and the log, and the project tool you already run for decisions. Do not build a dashboard before you have kept the cadence for 2 reviews. Tool selection is the easy part. Keeping the cadence is the hard part.

The whole plan on one page

A register, the 5 numbers with baselines, thresholds, and a quarterly review that feeds real decisions. Feed the review outcomes back into your engagement framework rather than into a separate reporting line; the framework is set out in what a stakeholder engagement framework is. For a worked example of the engagement work the numbers sit under, see creating a stakeholder engagement framework for a digital project.

Pick the one number that would change a decision

Choose the next significant decision your programme faces. Pick the one number of the 5 that would most change that decision if it moved. Collect it once this week, write the number and the date next to the decision, and set the threshold at which you would act. The other 4 numbers can wait until a decision needs them.