# How to design a bonus scheme

> Design a UK bonus scheme step by step: purpose, eligibility, pool funding, measures, gatekeepers, discretion, deferral, malus, clawback, leavers and.

Source: https://lumihr.co.uk/guides/how-to-design-bonus-scheme
Author: David Whitfield (https://lumihr.co.uk/about/david-whitfield)
UK law as at: 20 September 2026
Review by: 6 April 2027

---
*By David Whitfield · Reflects UK law as at 20 September 2026 · Review by 6 April 2027*

> **In short**
> - A discretion in a bonus contract must be exercised rationally and in good faith [2].
> - Bonuses go through PAYE and attract Class 1 NIC in the period paid [8].
> - In the lumi reward benchmark, 61.2% of organisations have a gatekeeper metric that must be met for any payout (n = 196).
> - By the end you will have a decision on each element, a payout model and draft rules.

## Before you start

**Legal basis.** Contracts and scheme rules set the terms, within equality law [4][5], wages law [6], tax rules [7][8][9], pay gap reporting [10][11] and case law [2][3].

**Who to involve.** Reward, HR, finance, legal, payroll and the remuneration committee.

**What you need.** Current rules and contracts, headcount by grade, working pattern and sex, the budget, and past payouts.

## Step 1: Set the purpose

Agree what the scheme is for; later choices follow.

In the lumi benchmark, retention is the common primary aim of incentives, at 31.7% (n = 268). Cost control, at 23.1%, is an alternative. Combination (17.9%), performance (16.4%) and not defined (10.8%) are rare ([full findings](/research/bonus-scheme-design-2026)).

**◆ Decision: primary purpose**
- *Performance.* Pay follows results; a nil year can affect retention.
- *Retention.* Steadier payouts for staying; a weaker link to results.
- *Cost control.* Pay flexes with affordability; employees carry more risk.
- *A combination.* Several aims; state which wins in a conflict.

## Step 2: Decide who is eligible

In the lumi benchmark, 36.2% of organisations make 75% or more of employees eligible for a bonus (common) and 26.9% make none eligible (alternative) (n = 268).

Two legal points apply:
- **Part-time staff.** Less favourable treatment for working part time must be objectively justified, and the pro rata principle applies unless inappropriate [5].
- **Maternity leave.** A woman's bonus must cover time before her maternity leave, her compulsory maternity leave and time after she returns [4].

**◆ Decision: who is in**
- *Everyone, one scheme.* Shared focus on results; weak individual line of sight.
- *Selected grades or roles.* Spend where influence is greatest; can feel two-tier.
- *Separate schemes by population.* Better fit; more rules to keep consistent.

## Step 3: Choose how the pool is funded

In the lumi benchmark (n = 145, N/A excluded), 39.3% fund the main pool bottom-up from targets (common), 33.1% top-down as a profit share (alternative) and 27.6% as a hybrid (alternative).

**◆ Decision: funding route**
- *Top-down.* A set share of profit. Affordable by design; weaker link to individual effort.
- *Bottom-up.* Targets times scores. Clear line of sight; cost less predictable without a gatekeeper or cap.
- *Hybrid.* A target pool scaled by profit. Balanced; harder to explain.

## Step 4: Choose measures and weightings

In the lumi benchmark (n = 151; more than one answer allowed), 83.4% use company financial performance, 67.5% individual objectives and 47.0% a performance rating. 27.5% include ESG, safety or people measures, as a modifier or underpin or as weighted measures (n = 266).

Define each measure's data source, threshold, target, maximum and payout. Weightings add to 100%.

**◆ Decision: collective or individual weight**
- *Mostly collective.* Simpler to calibrate; individual effort less visible.
- *Mostly individual.* Strong line of sight; more room for inconsistency.
- *Non-financial measures.* A weight gives a fixed share; a modifier or underpin adjusts or blocks the result.

## Step 5: Set the gatekeeper and model payouts

A gatekeeper is a profit, safety or compliance condition that must be met before anything pays. In the lumi benchmark, 61.2% of organisations have one (common) and 38.8% do not (alternative) (n = 196).

Here, on-target bonus is 10% of an illustrative £48,000 salary, capped at 150% of target, and group profit must reach 90% of budget.

```python
TARGET, CAP, GATE = 0.10 * 48_000, 1.5, 0.90   # salary illustrative
W = {"profit": 0.5, "revenue": 0.2, "individual": 0.3}

def curve(a):  # 0 below 90%, 50% at 90%, 100% at 100%, 150% at 110%+
    if a < 0.90: return 0.0
    if a <= 1.00: return 0.5 + (a - 0.90) * 5
    return min(CAP, 1.0 + (a - 1.00) * 5)

def bonus(profit, revenue, individual, gatekeeper=True):
    s = {"profit": curve(profit), "revenue": curve(revenue),
         "individual": min(CAP, individual)}
    score = sum(W[k] * s[k] for k in W)
    if gatekeeper and profit < GATE: score = 0.0
    return round(score, 4), round(TARGET * score, 2)
```

In every case, unit revenue is 97% of target and the individual score is 120%.

| Case | Profit vs budget | Score | Bonus |
|---|---|---|---|
| A | 104% | 113% | £5,424.00 |
| B | 88% | 0% | £0.00 |
| B, no gatekeeper | 88% | 53% | £2,544.00 |
| C | 92% | 83% | £3,984.00 |

The gatekeeper removes £2,544 in case B. B and C show the cliff edge at 90%. In the lumi benchmark, the median reported average payout in the last 12 months was 75% of maximum bonus potential (n = 179, N/A excluded).

**◆ Decision: form of gatekeeper**
- *Hard gate on the whole bonus.* Protects affordability; creates a cliff edge.
- *Gate on the financial part only.* The individual element still pays; less cost protection.
- *Modifier.* Payout scales down rather than stopping; more complex.
- *None.* Payouts follow the scorecard, even in a weak year.

## Step 6: Define discretion after the formula

In the lumi benchmark (n = 267, N/A excluded), 38.2% let a committee adjust the formula outcome with documentation (common) and 31.8% let managers adjust it (alternative). For 19.1% the formula is final, and 10.9% adjust informally without documentation (both rare).

In *Braganza v BP Shipping*, the Supreme Court held that a contractual discretion must be exercised rationally and in good faith: considering relevant matters, ignoring irrelevant ones, and not reaching an outcome no reasonable decision-maker could reach [2]. For bonuses without set criteria, Lord Hodge saw little scope for intensive scrutiny [2]. Record who may adjust, on what grounds and why.

**◆ Decision: who can adjust**
- *Nobody.* Predictable; blind to events the formula misses.
- *A committee, documented.* Consistent and auditable; slower.
- *Managers.* Close to the work; harder to keep consistent.
- *Informally.* Flexible; no record if challenged.

## Step 7: Decide on deferral

In the lumi benchmark, 76.5% of organisations do not defer bonuses (common, n = 268). Deferral for executives (10.4%), senior leadership (9.7%) or a wider population (3.4%) is rare. Where bonuses are deferred (n = 63), 38.1% defer into shares (common), 31.7% into 1-year cash and 30.2% into multi-year cash (both alternative).

Cash is taxed at the earlier of payment and entitlement [7], so state when entitlement arises.

**◆ Decision: deferral**
- *None.* Simple; nothing held back for malus.
- *Deferred cash.* Retention and room for malus; employees wait.
- *Deferred shares.* Links value to share price; needs a share plan and tax advice.

## Step 8: Draft malus and clawback

Malus reduces unpaid awards. Clawback recovers paid ones. In the lumi benchmark, 42.9% of organisations use malus (alternative) and 57.1% do not (common) (n = 196). 78.5% use clawback (common) and 21.5% do not (alternative) (n = 158, N/A excluded).

Drafting points:
- **Triggers and period.** List events, such as misstatement or misconduct, and how long they apply.
- **Discretion.** Applying malus or clawback is a discretion, so the *Braganza* duty applies [2].
- **Recovery.** A deduction from wages needs a prior written contract term or prior written consent, and cannot cover earlier events [6].
- **Gross or net.** Say which you will recover.
- **Penalty rule.** Clawback triggered by breach may be tested for being out of all proportion to your legitimate interest [3].
- **Leavers.** Say whether the provisions survive employment.

## Step 9: Set payment timing and leaver rules

In the lumi benchmark, 55.5% of organisations pay annual bonuses within 3 months of year end (common), 35.1% over 3 to 4 months (alternative) and 9.4% over 4 months (rare) (n = 265).

**◆ Decision: leavers**
- *Employed, not under notice, at payment.* A retention hook; leavers lose a bonus for time worked.
- *Good and bad leavers.* Pro rata for defined good leavers; definitions cause disputes.
- *Pro rata for all.* Simple; weaker retention.

**Tax and NIC.** Operate PAYE at the earlier of payment and entitlement, and work out NIC on total pay in the period the bonus is paid [8]. For 2026/27, employers pay 15% above £417 a month; employees pay 8% from £1,048 to £4,189 a month and 2% above [9]. Case A's £5,424, paid with an illustrative £4,000 monthly salary, costs £813.60 employer NIC and £119.82 extra employee NIC, as only £189 falls below £4,189. After the P45, use code 0T non-cumulatively [8].

**Warning: directors.** That example uses a monthly earnings period. Company directors have an annual, or pro rata annual, earnings period, whatever their pay frequency: NICs are worked out each time on total earnings paid so far in the tax year against annual thresholds, less NICs already paid [13]. Executive bonuses will therefore give a different result from the figure above, so model any scheme covering directors on the annual basis.

## Step 10: Check the gender pay gap and communicate

Employers with 250 or more employees report mean and median bonus gaps and the proportions of men and women paid a bonus [10][11]. Bonus pay includes performance, incentive and commission pay [10] in the 12 months to the snapshot date. Bonus figures are not pro-rated [11], so pro-rated part-time bonuses lower the averages for groups with more part-time staff. Shares count when they create an income tax charge [11]. Model the gap before finalising Steps 2, 4 and 7.

Tell eligible employees in writing how the scheme works, whether it is contractual, and what they could earn. Each year, explain results and adjustments. Changing a contractual scheme needs agreement, and fire-and-rehire protections start in January 2027 [12].

## Checklist

- [ ] Purpose agreed
- [ ] Eligibility checked for part-time staff and maternity leave
- [ ] Funding route, measures and weightings set
- [ ] Gatekeeper set and payouts modelled
- [ ] Adjustment rights and record-keeping written in
- [ ] Deferral and entitlement dates set
- [ ] Malus and clawback drafted and reviewed
- [ ] Leaver rules, payment date and NIC costs set
- [ ] Bonus gap modelled
- [ ] Employee summary issued

## FAQ

**Can an employer decide a discretionary bonus however it likes?**
No. A contractual discretion must be exercised rationally and in good faith [2].

**What is the difference between malus and clawback?**
Malus reduces an award before payment. Clawback recovers it after.

**Can we deduct a clawback from salary?**
Only with a prior written contract term or prior written consent [6].

**Do employees on maternity leave get a bonus?**
Yes, for time before leave, compulsory maternity leave and time after return [4].

## Related

- [Bonus scheme design: UK benchmark 2026](/research/bonus-scheme-design-2026) — the benchmark findings behind the figures in this guide.
- [How to set up an all-employee share plan: SAYE or SIP](/guides/how-to-set-up-all-employee-share-plan)
- [How to run an annual pay review: process and governance](/guides/how-to-run-annual-pay-review)
- [How lumi works: methodology](/methodology) — how these figures are collected and calculated.

## Sources

1. lumi reward benchmark, collection window 2026 H1, national figures: [full findings](/research/bonus-scheme-design-2026)
2. Supreme Court, *Braganza v BP Shipping Ltd* [2015] UKSC 17, 18 March 2015. https://caselaw.nationalarchives.gov.uk/uksc/2015/17
3. Supreme Court, *Cavendish Square Holding BV v Makdessi* [2015] UKSC 67, 4 November 2015. https://caselaw.nationalarchives.gov.uk/uksc/2015/67
4. legislation.gov.uk, *Equality Act 2010*, s74. https://www.legislation.gov.uk/ukpga/2010/15/section/74
5. legislation.gov.uk, *Part-time Workers (Prevention of Less Favourable Treatment) Regulations 2000*, reg 5. https://www.legislation.gov.uk/uksi/2000/1551/regulation/5
6. legislation.gov.uk, *Employment Rights Act 1996*, s13. https://www.legislation.gov.uk/ukpga/1996/18/section/13
7. legislation.gov.uk, *Income Tax (Earnings and Pensions) Act 2003*, s18. https://www.legislation.gov.uk/ukpga/2003/1/section/18
8. HMRC, *2026 to 2027: Employer further guide to PAYE and National Insurance contributions* (CWG2), 7 July 2026. https://www.gov.uk/government/publications/cwg2-further-guide-to-paye-and-national-insurance-contributions/2026-to-2027-employer-further-guide-to-paye-and-national-insurance-contributions
9. HMRC, *Rates and thresholds for employers 2026 to 2027*, 1 September 2026. https://www.gov.uk/guidance/rates-and-thresholds-for-employers-2026-to-2027
10. legislation.gov.uk, *Equality Act 2010 (Gender Pay Gap Information) Regulations 2017*, SI 2017/172. https://www.legislation.gov.uk/uksi/2017/172/contents
11. Government Equalities Office, *Gender pay gap reporting: guidance for employers*, 21 May 2026. https://www.gov.uk/government/publications/gender-pay-gap-reporting-guidance-for-employers
12. Department for Business and Trade, *Plan to Make Work Pay and Employment Rights Act: timeline update*, 25 August 2026. https://www.gov.uk/government/publications/implementing-the-plan-to-make-work-pay-and-employment-rights-act/plan-to-make-work-pay-and-employment-rights-act-timeline-update
13. HMRC, *CA44: National Insurance for company directors*, 2026 to 2027. https://www.gov.uk/government/publications/ca44-national-insurance-for-company-directors

*This guide is general information about UK employment law and practice as at the date shown. It is not legal advice. Take advice on your own circumstances.*

**Want to see how your policy compares with other UK employers?** [Book a call](https://lumihr.co.uk/demo)