How to improve employee engagement: what works, and in what order

In the UK, only one in ten employees feels engaged at work. Just 7% are actively disengaged, which is less than half the global average. The other 83% are simply 'not engaged.' This means the country doesn't have an “angry” workforce, but it does have the most indifferent one in Europe.

This is important because most engagement strategies focus on the visibly unhappy, like conflict resolution, tough conversations, performance management, and retention offers. In the UK, that's only about one in fourteen employees. The rest aren't unhappy; they're just not connected to their work, and helping them needs a different approach.

Another problem is the order in which you take action. Most guides list ten tactics without saying which are most important, and organisations often try to do everything at once. This creates a lot of activity but not much real progress. Engagement strategies usually fail because they are too complicated, seen as only an HR issue, or rely on surveys with no visible follow-up.

This guide puts the steps in order. It covers twelve changes, split into three stages over twelve months, and explains what to stop doing and how long each step takes to show results.

What does it mean to improve employee engagement?

Two different goals drive engagement improvement, and mixing them up wastes most budgets.

Raising the score means getting a higher survey number. Improving engagement means more people feel truly committed to the organisation's goals. These aren't the same, because most internal measures use averages instead of looking at who is actually committed.

The mechanism is well documented. Reporting "percent favourable" — the share selecting a 4 or 5 on a five-point scale — inflates scores and conceals problems.

A workforce where almost everyone is mildly positive and almost nobody is committed produces a healthy-looking average. That is the shape of the UK workforce, and it is why the main national surveys put British engagement around two-thirds while the strictest measure puts it at one in ten.

Before choosing your tactics, decide what you really want to change. If your goal is just to improve the score, the fastest way is to change the survey tool, but people will notice. If you want to boost real engagement, follow the evidence-based steps below.

What you actually get for it

Research comparingtens of thousands of businesses shows a big gap between the most and least engaged teams. This gap is why engagement matters for business results, not just for company culture.

Why engagement initiatives fail before they start

Six failure patterns recur, and five of them are structural rather than tactical.

The last failure pattern is specific to the UK and comes from the workforce distribution mentioned earlier. Programmes focused on conflict and counter-offers only target the smallest group of employees.

The depletion ceiling

There's a seventh failure pattern worth highlighting: noengagement programme can raise a workforce above the energy its people have left.

This is called “the depletion ceiling”. Engagement is about feeling connected to work, and that takes effort. People need to pay attention, care about results, and have some energy left at the end of the day. If employees are stretched too thin, they might still do their jobs well, but they won't feel connected. Recognition, purpose, or development won't help if the real issue is exhaustion, not motivation.

This is the main exception to the usual order. The twelve methods below put recovery last, because giving out gym memberships before managers have time for real conversations can look like the company is ignoring the real problem. But if exhaustion is the main issue, recovery should come first. Otherwise, stage two won't show results, and you might think the manager-focused work failed when it never really had a chance.

How can you tell which problem you have? Here are three signs you can spot without a survey:

• Absence and presenteeism. If you see more short-term absences, or the opposite—people working while clearly unwell—it shows your workforce is running on empty.
• Untaken leave. When annual leave builds up across a team, it's a clear sign of depletion—and you can spot it easily in your HR system.
• Cancelled one-to-ones. If regular meetings are the first thing dropped when things get busy, it means people are already at capacity.

If you notice two out of these three signs, start with method 12, then go back to the beginning. If none are present, follow the sequence as listed below.

How to improve employee engagement: 12 methods, in sequence

The order is important. Stages one and two remove barriers that would make stage three less effective. For example, offering a benefit that no one has time to use, or having a development talk with a manager who leads nineteen people, just adds cost without real progress.

Stage one: stop the losses (first 30 days)

None of these steps need budget approval. All three are things the organisation has already promised but not delivered, making them the cheapest engagement fixes—and the most damaging if left undone.

1. Clear the adjustments backlog you already owe people

This is the clearest causal finding in the UK evidence. National survey work finds that employees with long-term health conditions, neurodivergence or other protected characteristics report engagement 20% lower than colleagues — and that where proper support is in place, their engagement matches the wider workforce average. The gap is in the provision, not the people.

More than a third of employees who have disclosed a disability receive no workplace adjustments at all. Operationally, that means these are people who have already identified themselves, already made a request, and already been recorded somewhere. An adjustment agreed in a meeting and never implemented is worse than one declined, because the employee disclosed and was then ignored.

The same idea applies beyond formal adjustments. Inclusion efforts that stop at policy—a published commitment, a network with no budget, or a training module—show the same pattern: employees see promises that haven't been acted on.

Do this first: pull every adjustment request from the last twenty-four months and sort it into delivered, agreed-but-not-delivered, and declined. The middle pile is your first action list, and it is usually short, cheap, and overdue.

2. Publish the last survey's results, including what you are not going to change

The fastest way to lower engagement is to run a survey, find a problem, and then do nothing about it. Using too many surveys without follow-up is a common and avoidable mistake.

The Civil Service People Survey shows what happens when the main engagement score looks good, but communication is lacking. The engagement index is one of the highest ever recorded, but the number of staff who feel informed about organisational matters dropped by four points, and only 32% say change is managed well. survey found. What will change, with a date and a named owner. What will not change, and why. The third list builds credibility because it proves the first two are real.

3. Retire an initiative before you launch one

Making things too complicated is a real cause of failure, not just a matter of style. Most organisations trying to improve engagement already have wellbeing weeks, a recognition platform, an intranet campaign, a values refresh, and a mentoring scheme—many of which are inactive and not measured.

Start by listing every current people initiative, who owns it, its last measured outcome, and how much manager time it takes each week. Close anything without an owner or a clear outcome. In stage two, you'll need managers' attention, and that's a limited resource.

Stage two: fix the manager layer (months one to six)

About 70% of the difference in team engagement comes down to the manager. This stage is where the real work happens, but most programmes skip it because it takes up manager time instead of budget.

This group is also declining. Manager engagement has dropped by nine percentage points worldwide since 2022, with the biggest fall in the last year. Managers used to be more engaged than their teams, but now the difference is small. Arne Sjöström of Culture Amp told People Management that middle managers are "being squeezed from both directions" and warned that "if managers are depleted, the ripple effects are felt well beyond the manager population itself."

4. Count spans of control, then cut them

Before you invest in manager training, find out how many people each manager actually leads and what other responsibilities they have. After years of flattening and restructuring, many managers now have more direct reports than before. A manager with six people can meet with everyone every two weeks. A manager with nineteen can't, and more training won't change that.

In practice, map out spans of control, find everyone managing more than about twelve people, and either split the team, add a team lead, or take away non-management tasks. Share these numbers. If no one has counted a span of control, no one is responsible for it.

5. Replace the annual review with a fortnightly conversation

The advice that works is about how often you talk, not the format. Regular, meaningful coaching conversations matter more than a once-a-year performance review.

The NHS Staff Survey shows what happens when this is missing in the country's largest workforce. The staff engagement score fell for the second year in a row, and the number of staff who look forward to going to work dropped to just over half, the lowest level ever recorded.

In practice: meet every two weeks for thirty minutes, let the employee set the agenda, and have the manager record one commitment. That's the whole plan. If a one-to-one meeting is cancelled twice, it's a sign that work is being put ahead of the person.

6. Write down the expectations nobody has written down

"I know what is expected of me at work" is the first of Gallup's twelve engagement elements, and it fails most often not on the formal job description but on the unwritten rules: when a reply is expected, what "urgent" means, who decides, what good looks like this quarter as opposed to last.

This is also where hybrid work can quietly reduce engagement. Expectations that used to be picked up in person now need to be stated clearly.

In practice: each manager should write a one-page team operating note covering response times, decision rights, meeting expectations, and this quarter's main priority. Review it during one-to-ones. It takes about an hour and helps remove small communication issues that often show up in surveys.

7. Make recognition specific, fast, and unbudgeted

Recognition is the most common engagement tool companies buy, but it's one of the least effective when done through a platform. What works best is being specific and quick with recognition, not the size of the reward. Make recognition a regular agenda item and ask employees what makes them feel valued, since the answer is different for everyone.

This is also where a different idea comes in. Rewards help prevent dissatisfaction more than they build commitment, as Herzberg pointed out and as we cover in employee motivation. Recognition is about giving attention, not just handing out rewards.

In practice: name the behaviour and its effect within a week, and do it in front of someone the employee respects. You don't need a platform for this.

Stage three: widen the offer (months three to twelve)

Most guides start with these interventions. They work, but only after managers have the capacity and employees have received the adjustments they've requested.

8. Name the next step, even when it is sideways

Career progression drives engagement and is a common weakness in the UK. Fewer than half of civil servants now say they have career development opportunities. In flatter organizations, there are fewer steps up, so honesty matters more than ever.

In practice: every employee should know their next step and the two or three things that would help them get there. Moving sideways counts too—like switching functions, taking on a broader role, or developing a specialism. What doesn't count is a conversation that ends with 'keep doing what you're doing.'

9. Put connection inside the work, not beside it

Social connection drives engagement, and UK employees report higher rates of loneliness than those in other European countries. Standard responses like events, socials, and away-days tie connection to the calendar instead of the job itself, so attendance drops and the effect doesn't last.

In practice: working together on real tasks is better than just socialising. Try cross-team projects, paired problem-solving, peer learning groups, or rotating showcases—anything where people need each other to get the job done.

10. Apply flexibility consistently rather than generously

Inconsistency is worse than being strict. If a policy is applied differently across teams, or if some people get special exceptions, it looks like favoritism and undermines the fairness engagement depends on.

Our work-life balance guide covers details about flexible working—like hours, location, leave, and the right to disconnect. For engagement, the key is to publish the rule, apply it the same way for everyone, and have managers explain it consistently.

11. Hand AI adoption to managers, with time to do it

This is the newest and most overlooked way to improve engagement, and the evidence is clear. Nine out of ten firms have seen no productivity boost from AI  in the past three years, even though most use it. The best predictor of whether employees actually use AI, after it's set up, is whether their direct manager supports it—something fewer than a third of employees say happens. Companies that focus on technology instead of people are muchmore likely to miss their goals.

It's important to be clear: AI adoption doesn't cause disengagement. Instead, disengagement is the reason AI adoption doesn't deliver results.

In practice: set aside manager time along with buying licences, measure adoption by team instead of by individual, and don't expect employees to volunteer for redesigning their work unless they have a good reason.

12. Take the cost out of recovery

Engagement means feeling connected to work, and that requires having some energy left at the end of the day. British employees report the highest daily stress levels in Europe, and the Health and Safety Executive says 964,000 workers suffered from stress, depression, or anxiety caused or made worse by work in just one year.

This isn't just another wellbeing programme. It's the depletion ceiling mentioned earlier, shown as a number—and when that's the problem, recovery should come first. Recovery means having a real strategy to prevent burnout and support wellbeing, not just offering a list of benefits. A subsidised gym membership won't fix a manager having nineteen direct reports, and if it's offered before stage two, it just shows the company would rather spend money than pay attention.

How Wellhub helps UK employers improve engagement

The real benefit a wellbeing platform offers is structure. When support is offered randomly, people use it randomly. Usually, the gap between intention and habit comes down to access, not willingness. A single subscription that covers gyms, studios, classes, and mental health or nutrition apps lets employees make one decision instead of many—and removes the cost barrier that stops most people from trying in the first place.

The business case is better evidenced than it was. In Wellhub's own Return on Wellbeing 2026 research, 82% of HR and benefits leaders say wellbeing programmes sustain performance and 85% say they matter for retaining top performers — while 37% report rising absenteeism or presenteeism where mental wellbeing has declined, which is the depletion ceiling showing up in attendance data. Among organisations that actually measure the return, 95% report a positive one.

But none of this reduces a manager's span of control, delivers an adjustment someone asked for eighteen months ago, or turns a survey result into real change. A wellbeing benefit raises the ceiling, but it doesn't fill the room. Those tasks are still management's job, and that's where most UK engagement is lost.

What to stop doing

A list of things to stop doing is more helpful than just adding more ideas, because managers' attention is the main limit.

How to increase employee engagement in different industries

The sequence holds, but the binding constraint shifts by context.

How to improve employee engagement survey scores without gaming them

A higher engagement score only matters if it reflects real change. Four habits help keep the numbers honest.

Keep the measurement tool the same. If you change the scale, question wording, or reporting method during a programme, you can't track trends properly. If you have to make a change, publish both sets of results for one cycle.

Show the full distribution of results, not just the average. The key shift is from 'mildly positive' to 'committed,' and averages can hide it.

Break down the data where the gaps are biggest. Look at managers separately, since their engagement affects their teams. Also check employees with health conditions, neurodivergence, or protected characteristics, as described in method one. Review by tenure too, since new joiners and long-term staff disengage for different reasons.

Look at individual survey items, not just the overall theme. For example, in NHS data, the overall engagement score barely changed, but one key question hit a record low. Theme averages can hide important changes.

How long does it take?

Expectations are often where engagement programmes lose executive support. When managers are trained as coaches across the organisation, team engagement rises by 8% to 18% and turnover drops by 21% to 28%—but these changes take nine to eighteen months, not just a quarter.

If something promises to improve engagement in just a quarter, it's probably measuring enthusiasm, not real engagement.

How to measure whether it worked

Improvement and measurement go hand in hand, and the tool you use decides what counts as success. In short: use a threshold measure if you want to know how many people are truly committed, or an index if you want to track average sentiment—but never compare the two.

Use a measurement rhythm, not a single instrument

Three different survey rhythms serve different purposes, and using all three takes less effort than running one poorly designed annual survey.

The rule is the same for every survey rhythm: don't measure more often than you can act on the results. If you run a survey and don't follow up, people will stop taking it seriously.

Read the signals you already have

Sentiment is the slowest sign of change, but organisations often rely on it alone while quicker evidence sits unused in systems they already have.

• Voluntary turnover in the first twelve months of tenure — moves before any survey does, and isolates onboarding and manager quality from everything else.
• Absence and presenteeism patterns — the depletion ceiling, visible in attendance data.
• Untaken annual leave — accrual building across a team is a capacity warning that predates burnout.
• Internal mobility — people who see a future apply for it; a falling internal application rate is a progression signal.
• Benefit and programme participation — not as a success metric in itself, but as a read on whether people have the bandwidth to use what they have.

None of these signals require anyone to fill out a form, and you can break them down by team, which is where the real differences show up.

The bottom line

Improving employee engagement in a UK organisation isn't just about picking the right tactics. The best tactics are already well known. The real challenge is getting the order right and focusing on the right group of employees.

The right order is to keep existing promises, support the managers who have the most impact, and then expand your offerings. If you do this in the wrong order—like offering benefits before building capacity, launching platforms before setting routines, or providing training before reducing workloads—the same twelve steps will just add cost and activity without improving engagement.

Focus on the unattached majority, not just the unhappy minority. These employees aren't just looking for happiness at work—they just want to feel that the organisation cares about them. Most of the steps above are ways to show that the organisation does care.