Building the Cost Case for Workplace Connection Without Overselling It

A finance director will ask one question about a connection programme: what does it cost us not to run it. If you cannot answer with numbers you can defend under scrutiny, the programme gets cut in the first budget review that goes badly. Most cost cases for workplace loneliness interventions fail not because the underlying problem is small, but because someone reached for a bigger number than the evidence supports.

This is a guide to building the version that survives scrutiny.

The number everyone reaches for, and why it’s the wrong one

Cigna’s 2020 workplace survey found that 61% of U.S. adults report feeling lonely sometimes or always, up seven points on the year before, and that lonely workers miss work roughly twice as often due to illness and five times as often due to stress. Those absence multipliers are the numbers that end up in slide three of most internal pitches, multiplied by headcount and average salary to produce a headline figure.

Treat that multiplication with caution. The Cigna figures describe an association in a self-report survey, not a causal effect of loneliness on absence that a programme is guaranteed to reverse. Lonely people may also be dealing with other things — poor management, high workload, ill health — that independently drive absence and are not fixed by a peer-connection scheme. If you present the multiplied figure as “what this programme will save,” and absence doesn’t move, you have spent credibility you needed for the next budget cycle.

The honest move is to separate two different claims: the size of the problem, and the size of the effect a programme can plausibly claim. These require different evidence and different levels of confidence.

What the mortality and cardiovascular literature actually gives you

The strongest, most replicated evidence on social disconnection is about long-run health risk, not workplace productivity. Julianne Holt-Lunstad’s 2015 meta-analysis found that social isolation carries an odds ratio of 1.29 for early mortality, loneliness 1.26, and living alone 1.32 — effects that held after adjusting for existing health status and were more predictive of death in samples averaging under 65. The American Heart Association’s 2022 scientific statement put the increased risk of heart attack, stroke, or death from either at roughly 30%, and explicitly flagged the absence of intervention evidence as the field’s central gap. The 2023 U.S. Surgeon General’s advisory summarised the disconnection risk as comparable to smoking up to 15 cigarettes a day.

These figures are real and well-supported for the outcome they measure: long-term mortality and cardiovascular risk in populations, not attendance or output at a given firm over a fiscal year. They belong in a cost case as context for why disconnection matters at all — a way of establishing that this is not a soft or marginal issue — not as an input to a return-on-investment calculation. If a consultant offers you a dollar figure for “healthcare cost savings” derived directly from these odds ratios, ask them to show the chain of assumptions. There usually isn’t one, and the AHA statement itself says the intervention evidence isn’t there yet.

What this means in practice: use population-level mortality and cardiovascular evidence to establish that disconnection is a serious problem worth budget. Use workplace-specific survey data, cautiously, to estimate scale. Do not chain the two together into a single savings figure — that chain does not exist in the literature, and a sharp finance reviewer will find the seam.

What you can defensibly claim, and what you can’t

Claim	Evidence status
Disconnection is associated with markedly higher mortality and cardiovascular risk	Strong — large meta-analyses and an AHA scientific statement
A meaningful share of your workforce reports loneliness	Plausible to estimate from survey data (Cigna found 61% of U.S. adults overall, rising to over 80% of employed Gen Z), but self-report and time-bound
Lonely workers show higher absence in surveys	Documented association, not demonstrated as a programme-fixable causal chain
Running a connection programme will reduce healthcare costs by a specific dollar figure	Not supported — the AHA statement names this exact gap
Structured, purposeful group activity produces more perceived benefit than passive social contact	Reasonably supported in social prescribing research, though evidence base is qualitative and heterogeneous
A given programme design will replicate published effect sizes in your organisation	Unknown until you measure it locally

That fifth row matters for programme design as much as for the cost case. A 2021 systematic review of social prescribing found that all nine included studies reported positive individual impacts, and three reported reductions in downstream service use — but the review’s authors were working with a small, heterogeneous evidence base, and “positive impact” in that literature usually means self-reported wellbeing and confidence, not hard cost avoidance. If you’re citing social prescribing evidence to justify a workplace programme, be clear with your funder that you’re borrowing from an adjacent field, not a workplace-specific trial base.

The named failure mode: the borrowed baseline

Call this the borrowed baseline problem. It happens when a programme’s cost case imports an effect size from a population study — mortality risk, absence multipliers, healthcare cost averages — and applies it directly to a specific workforce, as though the mechanism that produced the number in the source population will operate identically in a 400-person office. It rarely works that way. The source populations in the mortality literature are broad, often older, and followed for years. Your programme is running for one fiscal year with a workforce that is younger, already employed, and subject to a dozen other simultaneous interventions.

The fix isn’t to abandon the external evidence — it’s the best evidence there is for why the problem matters — but to keep two ledgers. One ledger holds the case for why disconnection is worth addressing at all, built on the strongest available population evidence. The second ledger holds your own baseline and your own measured change, built from data you collect before and after the programme runs: self-reported loneliness or connectedness on a validated scale, absence records, retention, and engagement survey items your organisation already tracks. Only the second ledger belongs in next year’s budget request.

Building the case, step by step

1. State the problem using population evidence, cited plainly. Use Cigna’s workplace figures and the Surgeon General’s advisory to establish scale and severity, without implying they predict this year’s savings.
2. Take a baseline measurement before launch. A short validated loneliness or connectedness scale, plus whatever absence and attrition data you already hold. Without this, you have no local number to compare against later.
3. Set one primary outcome and name it before you start. Absence, retention, or an engagement survey item. Resist the temptation to report five outcomes and highlight whichever moved.
4. Cost the programme fully, including staff time and manager time spent supporting it, not just the vendor invoice.
5. Report the change against your own baseline, not against a borrowed effect size. If absence falls 2%, say 2%, and say it is against your own prior year, not against Cigna’s fivefold stress-absence multiplier.
6. State what you didn’t measure. If you have no clean way to isolate the programme’s effect from other changes that year — a reorganisation, a pay rise, an economic downturn — say so. A funder who trusts your caveats trusts your numbers.

What this does not solve

A cost case, however honest, will not tell you whether the people who need connection most are the ones showing up. Workplace programmes reliably reach engaged employees who already attend optional events; they are structurally weaker at reaching people working remotely, on night shifts, or already withdrawing from workplace social life — often the same people the mortality and cardiovascular evidence suggests are most at risk. A rigorous financial case built on a self-selected sample can be internally sound and still miss the population it was meant to help.