Giving non-profit the engagement intelligence that three years of surveys had never revealed

Situation

The organization had data. It had simply never looked at it all at once.

A mid-market non-profit had administered an annual employee engagement survey for three consecutive years. Each year, a report was produced, but the reports were never reviewed together. Year-over-year results, in isolation, meant trends, compounding signals, and emerging risk areas were never surfaced. The picture was always a single snapshot, never the full sequence.

That’s a costly blind spot. The most important signals in engagement data rarely live in any one year; they live in the movement between years. The organization engaged Growth Operators to conduct a full retrospective across all three years of results and to make recommendations on survey design and cadence going forward.

Execution

During our engagement, the Growth Operators team:

• Applied AI to conduct a three-year longitudinal analysis of engagement survey results — identifying trend lines, cohort-level shifts, and statistically significant changes across key driver categories at depth and speed not practical with manual analysis
• Surfaced insight themes that single-year reporting had masked, including sustained declines in specific engagement dimensions and post-reorganization shifts by department
• Used AI to draft the findings narrative, translating statistical patterns into plain-language leadership commentary ready for executive review
• Leveraged AI to design a new quarterly pulse survey, analyzing which dimensions from the annual instrument carried the highest signal value and structuring questions to maximize response quality
• Implemented the pulse survey alongside the annual instrument in a combined reporting structure, reducing total HR analyst time spent on survey analysis by roughly 50%

The AI Value

AI didn’t just accelerate this engagement — it changed the kind of intelligence the work could produce:

• Cross-year analysis: AI identified patterns across three years of data, simultaneously correlations and trend lines that would have taken weeks of manual analysis to assemble
• Insight generation: AI drafted narrative findings directly from statistical outputs, closing the gap between data and decision-ready communication
• Survey design: AI evaluated question-level signal strength across three years of historical responses to inform how the pulse survey was built
• Depth that changes decisions: Single-year analysis tends to produce directional findings; things are better or worse than last year. Multi-year AI analysis produced causal hypotheses: sustained declines tied to specific organizational events, and engagement patterns that signaled turnover before it appeared in attrition data. That’s a qualitatively different kind of intelligence, not a faster version of the same one
• Questions leadership didn’t know to ask: The retrospective surfaced themes that weren’t on anyone’s radar, precisely because no one had looked across all three years at once. AI doesn’t only answer the questions you bring to it — it finds the ones you didn’t think to ask
• An instrument grounded in evidence: Because the pulse survey was designed from AI analysis of three years of signal data, question selection was defensible rather than intuitive. Leadership had confidence that the pulse was measuring the right things, because the design was derived from what actually moved in the historical data, not from a generic best-practice template

Results

• Cut HR survey analysis time by approximately 50%
• Established a two-speed engagement intelligence system — a quarterly pulse for early signals and an annual survey for the full diagnostic
• Increased both the frequency and the relevance of the engagement data that leadership acts on
• Revealed multi-year trends, post-reorganization department shifts, and early turnover indicators that standalone reporting had never surfaced
• Delivered a pulse instrument whose question design was derived from evidence, giving leadership confidence that it measures what matters
• Validated the new cadence with initial pulse results — response rates held, and key themes aligned with the leading indicators the annual analysis had flagged

Client Success

AI doesn’t just do the work faster. It changes the ceiling on what the work can produce.

Three years of standalone reporting had answered the same question each time: was engagement better or worse than last year? Useful, but directional. Looking across all three years at once produced something different in kind — causal hypotheses, early warning signals, and themes no one had thought to look for. That’s the distinction that matters. The value wasn’t a quicker report. It was intelligence the organization had never had access to, drawn from data it had been holding all along.

HR leadership now operates with a system built for both horizons: a quarterly pulse that catches early signals and an annual survey that delivers the full diagnostic. The analysis runs in half the time it used to, and the data leadership receives is more frequent and more relevant. Most importantly, the pulse was designed from evidence rather than instinct — so leadership can act on what it surfaces with confidence. At Growth Operators, that’s the bar: not faster answers to the questions you already have, but a higher ceiling on what you can understand and decide.