Employee Engagement London: Change What Managers Do

Engagement scores measure an outcome, not a behaviour

Most engagement work follows the same loop. The annual survey lands, the score has slipped, a steering group is formed, communications go out, and twelve months later a near-identical number arrives with a different vendor's logo on it. The loop persists because it mistakes the dial for the engine. An engagement score is an outcome. It is the residue of a year of one-to-ones, of feedback given or dodged, of workload decisions, and of the way somebody reacted when bad news reached them. Measuring the residue does not touch any of it.

That makes the only useful brief for an employee engagement consultant a behavioural one. Not how do we lift the score, but which specific things do our managers do differently on a Tuesday afternoon, and how would we know they had. It is a less comfortable question, because it names moments rather than themes, and because it can be checked afterwards.

It is also the question the evidence supports. Gallup's manager research found that managers account for at least 70% of the variance in employee engagement scores across business units. If seven-tenths of the difference between your best and worst team sits with one layer of people, an intervention that leaves that layer untouched is not an intervention.

The numbers behind a London engagement brief

Gallup's State of the Global Workplace 2026 report, published in April 2026 with data collected through 2025, puts engagement in the United Kingdom at 10%. The Europe regional average is 12% and the global average is 20%. UK employers are operating at half the world average on the most basic measure of whether people are actually trying at work.

The same report shows where it comes from. Global engagement fell to 20% in 2025, its lowest level since 2020, and the steepest fall inside that figure was among managers. Manager engagement declined five points between 2024 and 2025, from 27% to 22%, and has dropped nine points since 2022. Gallup estimates low engagement cost the world economy approximately $10 trillion in lost productivity last year, or 9% of GDP.

Put the UK figure and the manager figure side by side and the shape is clear. The employee cliff tracks the manager cliff. The layer that decides what work feels like has been quietly switching off, and in the UK it has been doing so from a lower base than almost anywhere in the developed world. For a London employer, the engagement brief is rarely a communications brief. It is a management capability brief wearing a communications costume.

Why re-running the survey does not move it

Three responses dominate and none of them touches the mechanism. The first is to measure harder: a new survey vendor, pulse surveys, a live dashboard. Better instrumentation tells you faster that the same thing is still happening.

The second is to soothe. Wellbeing apps, an extra day of leave, a resilience webinar. None of these is a bad thing, but they treat a competence deficit as a comfort deficit. A manager does not dread Monday because of a shortage of perks. They dread it because there is a conversation they have been avoiding for three weeks.

The third is to cut. When margins tighten, the management layer looks like overhead, spans of control widen, and the managers who remain absorb larger teams with no more time. Engagement does not recover from a heavier load.

Underneath all three sits an untrained layer. Research by the Chartered Management Institute with YouGov, based on a survey of 4,500 UK workers and managers, found that 82% of people who move into management have had no formal management or leadership training. Of the 2,018 workers in that survey, one in three had already left a job because of bad management.

The behaviours an engagement score is made of

Strip the construct back and engagement is a short list of observable manager behaviours. Whether the person knows what is expected of them this week. Whether anyone has spoken to them about their progress in the last six months. Whether the person who leads them appears to care about them. Each of those is a conversation, held by an individual, under time pressure and usually unrehearsed.

This is where behavioural science earns its place. The COM-B model set out by Michie, van Stralen and West in Implementation Science in 2011 holds that behaviour requires three things at once: capability, opportunity and motivation. Most engagement programmes work on motivation alone, which is why they produce a good fortnight and then nothing. A manager who has never rehearsed a hard conversation lacks capability. A manager with eleven direct reports and back-to-back meetings lacks opportunity. Enthusiasm closes neither gap.

So the diagnostic question we put to clients early is simple and slightly rude. In your calendar, where does any manager actually practise the behaviour you say you want? When the honest answer is nowhere, the engagement score is not a mystery.

How we work with London organisations

Sidestream is based at 58 Malden Road in London. We build immersive theatre productions with professional actors and pair them with behavioural science, so that managers do not discuss a scenario, they run it. The actor playing the team member who has gone quiet pushes back, deflects and reacts the way a real person does, and the manager holds the conversation more than once, with feedback between attempts, until the skilled response is the automatic one rather than the effortful one.

The discomfort is the point. A manager's first reaction in a live conversation runs before conscious thought catches up, so it does not change by being told to change. It changes by being practised in conditions close enough to the real moment that the new response becomes the default.

The method has developed alongside academic work at UCL, the University of Cambridge and Bocconi, and has been applied with the Metropolitan Police, Goldsmiths, Imperial College London, HM Courts and Tribunals, TCS, WISE and the Innocence Project. Our production The Death of Jane Doe was recognised at the CorpComms Awards and The Accused was recognised at the Goldsmiths Public Engagement Awards. A typical engagement runs in four parts, and the shape is closest to our leadership training in London.

A short diagnostic that identifies which manager behaviours are actually driving your score, drawn from your own data rather than a generic model
Scenario writing built from the conversations your managers are facing this quarter, not from generic leadership abstractions
Rehearsal with professional actors, in cohorts small enough that every manager runs the conversation live and more than once
A ninety-day embedding period with structured peer practice, so the new behaviour survives contact with the calendar

Measuring the work at Kirkpatrick Level 3

Most training evaluation stops at the happy sheet. In the Kirkpatrick model that is Level 1, Reaction, which measures how favourable participants found the experience. Level 3, Behaviour, measures the degree to which people perform the critical behaviours in their environment and are supported in and accountable for that performance. It is a harder standard, and in engagement work it is the only one worth paying for.

We agree the observable behaviours before the programme starts, so ninety days later there is something specific to look for: the one-to-one that now happens fortnightly and has an agenda, the feedback given within the week rather than saved for the review, the concern raised in the meeting rather than in the corridor afterwards. Those are checkable. A satisfaction score is not, and neither is an annual engagement figure that moves for several reasons at once.

This also guards against a pattern we see repeatedly in our own measurement. Participants rate their own skill well above their observed performance. Managers do not always know what they cannot do, which is why self-report is a weak instrument here and behavioural observation replaces it. The same logic runs through how we measure behaviour change.

What to ask before appointing anyone

London is not short of engagement consultancies, and the differences that matter are visible in the first meeting if you ask for specifics rather than for a methodology.

Any provider who answers those five clearly is worth talking to, including the ones that are not us. Sidestream is priced per engagement and scoped after the diagnostic, because the honest cost depends on how many managers need to rehearse and how often.

Which behaviours will change, named as things a manager does, not as values or themes
Where in the programme each manager practises those behaviours out loud, and how many times
Who plays the other person in that practice, and how hard the pushback actually is
What is measured at ninety days, and whether it is behaviour or satisfaction
What happens between the sessions, since behaviour that is not rehearsed afterwards reverts

Employee Engagement Consultant in London: In Short

An engagement score is an outcome, and the behaviour that produces it belongs to managers: Gallup attributes at least 70% of the variance in team engagement to that layer. UK engagement stands at 10% against a global 20%, and manager engagement has fallen to 22% from 27% a year earlier. Sidestream, based at 58 Malden Road in London, changes that layer's behaviour through immersive rehearsal with professional actors, designed on COM-B and measured at Kirkpatrick Level 3 on observed behaviour ninety days out rather than on satisfaction.

Frequently Asked Questions

What does an employee engagement consultant actually do?

The useful version diagnoses which manager behaviours are driving your score, then changes those behaviours through rehearsal rather than communication. That means defining a small number of observable actions, giving managers repeated live practice at them with feedback, and checking at ninety days whether the actions are happening. The version that redesigns the survey and runs a campaign produces a report, not a different Tuesday.

Why is UK employee engagement so low?

Gallup's State of the Global Workplace 2026 puts UK engagement at 10%, against a European average of 12% and a global average of 20%. The proximate cause is the management layer. Gallup's manager research attributes at least 70% of the variance in team engagement to the manager, and research by the Chartered Management Institute with YouGov found 82% of people entering management had no formal management or leadership training.

Will a survey or a wellbeing programme lift engagement?

A survey measures, it does not intervene, and a wellbeing programme addresses comfort rather than capability. Neither reaches the conversation that decides whether somebody feels engaged on a Wednesday afternoon. Keep both, as an instrument and as a benefit, but do not mistake either for the work. Gallup's manager engagement figure still fell to 22% in 2025, down from 27% the year before.

How do you measure whether the work changed anything?

On observed behaviour, at Kirkpatrick Level 3, which measures the degree to which people perform the critical behaviours in their environment and are supported in and accountable for that performance. We agree those behaviours before the programme starts, so ninety days later there is something specific to check: the one-to-one that now happens, the feedback given within the week, the concern raised in the room rather than in the corridor.

Do you work with organisations outside central London?

Yes. Sidestream is based at 58 Malden Road in London and most of the work is delivered on client sites across the city and beyond, including with the Metropolitan Police, Goldsmiths, Imperial College London and HM Courts and Tribunals. Sessions run wherever the managers are, because the point is rehearsing the conversations that specific group is facing this quarter.

How long before it shows up in the engagement numbers?

Behaviour changes inside the room within a single session. Whether it survives depends on what happens afterwards, which is why we plan a ninety-day embedding period with peer practice and measure behaviour at that point. An annual engagement score is a lagging indicator and moves later, for several reasons at once, so we treat behaviour rather than the survey as the primary measure.

Related Sidestream Guides

Sources: Gallup, State of the Global Workplace 2026: global engagement 20%, manager engagement 22% (down from 27%), $10 trillion or 9% of GDP · Gallup, State of the Global Workplace: United Kingdom Country-Level Data (UK 10%, Europe 12%, global 20%) · Gallup, Managers Account for 70% of Variance in Employee Engagement (State of the American Manager) · Gallup, Employee Engagement indicator page (Europe 12% engaged; managers account for at least 70% of variance) · CMI/YouGov research on accidental managers: 82% with no formal management training, 4,500 UK workers and managers surveyed, one in three of 2,018 workers left a job because of bad management (reported by Fortune) · Small Business Charter summary of the CMI/YouGov finding: 82% of managers have had no formal management training · Michie, van Stralen and West (2011), The behaviour change wheel: a new method for characterising and designing behaviour change interventions, Implementation Science 6:42 (COM-B) · Kirkpatrick Partners, The Kirkpatrick Model: Level 3 Behavior definition