Engagement, satisfaction, happiness

TL;DR

Engagement, Satisfaction, Happiness: Which One Actually Predicts Performance?

  • Satisfaction is contentment with the deal; engagement is energy in the work; happiness is how the days actually feel.
  • Satisfaction-performance correlates at about .30; the arrow runs both ways.
  • The BT call centre study is the nearest thing to causal proof: 13% more productive in happy weeks.
  • All three are outputs. The input - fit between person and work - is the thing a leader can actually change.

Ask a board how the workforce is doing and you will hear three words used as if they were one: engaged, satisfied, happy. They are not one. They are three different measurements, with three different evidence bases, and confusing them leads to spending money on the wrong one.

What is the difference between employee engagement and job satisfaction?

Satisfaction is contentment with the deal - pay, conditions, colleagues, the job as it is. Engagement is energy directed into the work - involvement, absorption, willingness to give discretionary effort. A satisfied employee may be comfortably coasting; an engaged one is invested. The two overlap and are often measured together, and critics such as Rob Briner note the engagement construct borrows heavily from older ideas. The practical distinction: satisfaction predicts staying, engagement is the better frame for effort.

The coasting case makes the difference vivid. Picture an employee who likes the pay, likes the team, finds the work undemanding and plans to stay for years. Highly satisfied. Minimally engaged. Every organisation has these people, and a satisfaction survey files them as a success story.

What the meta-analyses found

Does job satisfaction predict performance?

Moderately. The landmark meta-analysis by Judge, Thoresen, Bono and Patton - 312 samples, around 54,000 employees - put the corrected correlation between job satisfaction and job performance at about .30. That is meaningful but far from destiny, and the causal arrow runs both ways: performing well also makes people more satisfied. Later work by Bowling argued part of the correlation reflects underlying personality. Satisfaction matters; it is simply not a lever you can pull directly.

A correlation of .30 explains under a tenth of the variance in performance - real enough to care about, small enough to warn against slogans. The more commercially interesting numbers sit at business-unit level. Harter, Schmidt and Hayes's meta-analysis of 7,939 business units found unit-level satisfaction and engagement related to customer satisfaction, productivity, profit, turnover and safety - the analysis that effectively launched the engagement industry. Gallup's later quartile comparisons put flesh on it: 21% higher profitability, 41% lower absenteeism in top-quartile units.

The caveat that keeps recurring in this series: these are correlations. High-performing units may generate engagement rather than be generated by it. Which is why one British call centre study matters more than its size suggests.

Are happy employees more productive?

The best causal evidence says yes, measurably. Bellet, De Neve and Ward tracked around 1,800 BT call centre workers weekly and found employees were 13% more productive in weeks when they reported being happy - same people, same targets, different weeks, which strips out most alternative explanations. Krekel, Ward and De Neve's meta-analysis across 339 studies and 1.88 million employees found the same direction at scale: wellbeing correlates with productivity, customer loyalty and profitability, and negatively with turnover.

The BT study's design is its strength. Because each worker was compared with themselves across weeks, stable differences between people - ability, role, temperament - drop out of the equation. What remains is the within-person effect of a good week: more calls, more conversions, no more hours worked. Happiness, in that study, was not a reward for performance. It arrived first.

Three gauges, one engine

So: satisfaction predicts retention moderately, engagement tracks unit performance, happiness has the cleanest causal link to weekly productivity. Which should an organisation measure? The question contains its own mistake - all three are gauges on the same engine, and none of them is the engine.

What sits upstream of all three is fit: whether the work, as actually experienced day to day, runs with or against how each person prefers to operate. A person whose role fits reports satisfaction when surveyed, shows engagement when observed, and has more happy weeks when tracked. A person whose role grinds against their preferences shows up in every gauge eventually - absence first, then the scores, then the resignation.

Reading fit directly is what Sariio MAPS does. A ten-minute survey reads 60 work preference pairs on continuous scales; the AI maps them for the person, the team and the organisation in one vocabulary; and the survey is retaken at least twice a year, because preferences move with context and the delta between readings is itself a signal. Sentiment gauges tell you something moved. A preference map is the diagram of the engine - it shows which specific fit, for which specific person, needs attention. The gauges and the map work best together, which is the argument of do employee engagement surveys actually work.

Read the engine, not just the gauges - the individual survey is free.

Sources