New webinar on 3rd March 2026 | 9:00 am PT
In this webinar, discover how Sopact Sense revolutionizes data collection and analysis.
Learn how to define and track survey metrics and KPIs that matter. From participation and data quality to engagement and outcome measures, this guide.
A nonprofit program director opens her year-end report. The headline reads: "We delivered 240 training hours to 187 participants across 18 workshops." The numbers are accurate. The report is detailed. And the funder still asks the same question she hears every year — "but what changed?" Hours, participants, and workshops are activity metrics. They don't answer her funder's question. They can't.
This is The Scorecard Swap — when teams report activity metrics (workshops delivered, people reached, hours spent) as if they were outcome metrics (skills gained, behavior changed, lives improved). The two measure completely different things. Swapping them feels harmless. Until someone asks "compared to what?" and the whole report comes apart.
Last updated: April 2026
Most guides treat survey metrics and KPIs as a long list of formulas. This guide treats them as choices — a handful of specific numbers, each tied to a specific decision, each answering a specific question. Pick well and every month's report becomes useful. Pick badly and no dashboard will save you.
Survey metrics are the specific numbers you track from the answers in a survey. They come in three main forms — counts (how many), ratings (on a scale), and percentages (out of a whole). A good survey metric is specific, repeatable, and tied to a decision someone will eventually make.
Survey metrics are not the same as all the data a survey produces. Most surveys collect dozens of fields. Only a few of those become metrics — the ones you'll track over time, compare across groups, or use to make decisions.
Survey KPIs (Key Performance Indicators) are the handful of most important survey metrics — usually 3 to 7 — chosen because they show whether a program, product, or service is working. Every KPI is a metric. Not every metric is a KPI.
A survey KPI answers a business or program question directly. "Confidence score for specific job tasks" can be a KPI if it predicts employment outcomes. "Number of survey responses collected" usually isn't a KPI — it's a process measure about the survey itself, not about what the survey is supposed to reveal.
Survey metrics are any numbers tracked from survey responses. Survey KPIs are the small, specific subset of those metrics that directly tie to the decisions the program needs to inform. You can have hundreds of metrics. You should only have a handful of KPIs.
Think of it as zoom level. Metrics are the full picture — everything the survey can tell you. KPIs are the zoomed-in view — the 3–7 numbers a leader checks first when they want to know "is this working?" Teams that confuse the two end up drowning in data and starving for signal.
There are four main types of survey metrics, and they map to four different levels of what a program is actually doing. Most reports mix them up. Good reports label each one clearly.
1. Activity metrics — what you did. Hours delivered, workshops held, meetings attended. Useful for operations, dangerous as the headline.
2. Output metrics — what was produced. People trained, forms completed, materials distributed. Still mostly about your team's effort, not the participant's change.
3. Outcome metrics — what changed. Skill gains, behavior shifts, attitude changes, confidence scores. This is where funders, boards, and leaders start paying attention.
4. Impact metrics — what ultimately happened in the world. Jobs obtained, income increased, communities strengthened, mortality reduced. The hardest to measure — and the one most funders are actually buying.
The single biggest mistake in survey reporting is reporting activity or output metrics and calling them outcomes. That's the Scorecard Swap. For the deeper reference point behind each number, see our baseline data guide.
Quantitative KPIs are KPIs based on numbers you can count or rate — percentages, scores, counts, and ratios. Things like "78% completion rate," "4.2 out of 5 confidence score," or "342 participants served." They are measurable, comparable across groups, and fast to track on a dashboard.
Most organizations start with quantitative KPIs because they're easy to chart. The limitation: a number on its own rarely tells you why it moved. A 4.2 confidence score tells you where people stand — not what drove it up or down. That's why the strongest reports pair quantitative KPIs with one qualitative KPI each.
Qualitative KPIs are KPIs based on themes and patterns from open-ended answers — not raw numbers, but structured categories you can count. Examples include "top three reasons participants dropped off," "most common barrier reported," or "theme of top 10% satisfied responses." Yes, KPIs can absolutely be qualitative. And the strongest programs run them alongside quantitative KPIs.
The trick with qualitative KPIs: the open-ended answers need to be coded into themes before they become countable. That used to take weeks of manual work. AI-assisted coding — the kind built into modern survey platforms — now does it in minutes with source citations, which is what makes qualitative KPIs practical at scale. Related: open-ended survey questions and qualitative survey.
A survey measure is the underlying thing you're trying to capture — confidence, satisfaction, frequency, readiness. A survey metric is the specific number you calculate from survey answers to represent that measure. The measure is the concept. The metric is how you count it.
For example, "participant confidence" is a measure. "Average confidence score on a 1–10 scale" is the metric that operationalizes it. Both words get used interchangeably in everyday conversation, but the distinction matters when you're designing a survey: first pick the measures, then decide the metrics.
[embed: bestpractices]
Here are three to four sample KPIs for each of four common program types. Each KPI is specific, repeatable, and decision-linked — the traits that separate real KPIs from busy-work metrics.
Training and workforce programs — confidence gain per participant on 5 specific job tasks (pre-post score change); percent of participants employed within 90 days; top 3 barriers reported in open-ended exit surveys (qualitative KPI); completion rate at 80% or higher.
Nonprofit service programs — housing stability score at 12 months (rating scale); number of repeat service touches per client (behavior); most common follow-up need reported across all intake surveys (qualitative KPI); percent of clients still engaged at 6 months.
Impact funds and portfolios — investee growth score across 5 pillars (scale); percent of portfolio companies hitting their quarterly milestones; top 3 blockers reported in quarterly pulse (qualitative KPI); board composition diversity metric.
Customer experience teams — Net Promoter Score (NPS); percent of detractors whose issue was resolved in 30 days; top 5 themes from open-ended feedback (qualitative KPI); time to first response for critical tickets.
Notice the pattern: 3–4 KPIs per program. One or more qualitative. All tied to a specific decision. No team needs 30 KPIs.
Picking the metrics that matter is three decisions in order.
Decision 1 — Name the outcome. What single thing must this program change? Write it as one sentence. If you can't, you're not ready to pick KPIs — you're still figuring out the program.
Decision 2 — Pick 3–7 metrics that directly show whether the outcome is happening. Not activity. Not output. The actual change in the people you're serving. Include at least one qualitative KPI if you want to know why, not just what.
Decision 3 — Pair every metric with a decision. For each KPI, answer: "if this number moves, what action follows?" If there's no answer, remove the KPI. It's not earning its spot on the dashboard.
These three decisions are where most reports start going wrong — not at analysis time, but at selection time. Fix the selection and the analysis almost takes care of itself. Related: survey methodology covers the full methodology framework.
Mistake 1 — Reporting activity metrics as outcome metrics. "240 training hours delivered" is not an outcome. It's an activity. The Scorecard Swap starts the moment you report one as the other.
Mistake 2 — Tracking too many KPIs. Once you pass 7 KPIs, nobody focuses on any of them. A dashboard with 30 numbers is a dashboard with zero signals. Cut hard and keep cutting.
Mistake 3 — Quantitative-only scorecards. Running 8 rating scales with no open-ended questions. The numbers move, and nobody knows why. Pair at least one qualitative KPI with the numbers.
Mistake 4 — Swapping scales mid-study. A 1–5 scale in Q1 and a 1–10 scale in Q2 are different measurements. The metric is broken the moment the scale changes. Lock the scale. Don't edit mid-study. Related: baseline survey covers why locking instruments matters across waves.
Mistake 5 — No decision attached to the KPI. Tracking "overall satisfaction" with no plan for what to do when it moves. If there's no action on the other side of the number, the KPI isn't a KPI — it's a vanity number.
Mistake 6 — Confusing a measure with a metric. "Participant confidence" is a measure. "Average 1–10 confidence score on spreadsheet tasks" is a metric. Saying you're "measuring confidence" without defining the metric leads to different teams measuring it differently every quarter.
Survey metrics are the specific numbers you track from survey answers — counts, ratings, and percentages. Each one should be specific, repeatable, and tied to a decision. Not every piece of survey data becomes a metric. The ones that do are the numbers your team will actually use to spot change, compare groups, or make decisions.
Survey KPIs are the 3 to 7 most important survey metrics that directly show whether a program, product, or service is working. Every KPI is a metric. Not every metric is a KPI. KPIs should be chosen because they answer a business question — not because they're easy to collect. Sopact Sense tracks KPIs live as survey responses arrive.
Survey metrics are any numbers tracked from survey responses. Survey KPIs are the smaller, carefully chosen subset of metrics that directly tie to program decisions. Metrics are the full data picture. KPIs are the handful of numbers a leader checks first. A good scorecard has many metrics available but only a few KPIs in focus.
Yes, KPIs can absolutely be qualitative. A qualitative KPI is based on themes and patterns from open-ended answers — for example, "top three reasons for drop-off" or "most common barrier reported." The answers get coded into themes that you can then count. AI-assisted coding makes qualitative KPIs practical at scale.
No, KPIs do not have to be quantitative. While numbers are the most common form, KPIs can be qualitative — counting themes from open-ended answers, for example. The strongest scorecards run both: quantitative KPIs to track what is changing, and qualitative KPIs to track why. Mixing them tells a richer story than either alone.
Quantitative KPIs are KPIs based on countable numbers — percentages, scores, counts, and ratios. Examples include completion rate, average rating score, and number of participants. They're fast to chart and easy to compare. Their limitation: a number on its own rarely explains why it moved. That's why strong programs pair them with qualitative KPIs.
Qualitative KPIs are KPIs based on themes and patterns from open-ended survey answers. Examples include "top three drop-off reasons" and "most common barrier reported." The answers get coded into themes and then counted. Qualitative KPIs tell you the why that the numbers can't. Sopact Sense codes them automatically as responses arrive.
Survey measures are the underlying concepts a survey tries to capture — confidence, satisfaction, frequency, readiness. Survey metrics are the specific numbers you calculate to represent those measures. Measures are the concept; metrics are the count. Pick the measures first (what you want to know), then decide the metrics (how you'll count it).
The Scorecard Swap is when teams report activity metrics — workshops delivered, people served, hours spent — as if they were outcome metrics like skills gained or lives changed. The two measure completely different things. Swapping them makes reports look strong while failing to answer the question every serious reviewer eventually asks: "what actually changed?"
Measure success in a survey by picking 3 to 7 outcome KPIs tied to the decisions you need to make, then tracking them across waves with the same people. Include at least one qualitative KPI for the why. Compare results against your baseline to show change — not against activity numbers, which tell a different story.
A survey should have 3 to 7 KPIs at most. Fewer than 3 and you miss important angles. More than 7 and nobody focuses on any of them. A 30-KPI dashboard is a 0-KPI dashboard in practice. Cut hard, keep only the ones tied to decisions, and revisit the list each year.
Sopact Sense tracks both quantitative and qualitative KPIs in one system. Open-ended answers get coded by AI as they arrive, themes count themselves into qualitative KPIs, and every number on the dashboard carries its compared-to inline. Permanent participant IDs link KPI values across waves, so change tracks per-person, not just per-survey.
