Vesting schedules, cliffs and back-loaded vesting

1. The core idea#

Vesting turns an equity grant into something earned over time. The near-universal startup default is four years with a one-year cliff: nothing vests in the first 12 months, a quarter vests at the first anniversary, and the rest vests monthly (or quarterly) over the next three years. It protects the company from giving equity to people who leave early, and gives employees a reason to stay.

Defaults are useful because candidates and investors understand them instantly. But they are a choice, and a few variations are worth knowing:

• Cliff length: one year is standard; six months is sometimes offered to senior hires who are giving up unvested equity elsewhere.
• Vesting period: four years is standard; three years appears in competitive hiring; five or six years has been suggested for deeptech, where the road to an exit is longer. Back-loaded vesting: more vests in later years (for example 10/20/30/40%), used by Amazon and Snap, rare in startups.
• Front-loaded or no-cliff vesting: sometimes used for refresh grants or early hires who have already worked unpaid.
• Milestone or performance vesting: tied to a technical or commercial milestone rather than time; useful for some executives and for founders.
• Acceleration: faster vesting on a sale (single-trigger) or a sale followed by termination (double-trigger) (3.06).

For deeptech teams the main design question is alignment with a long, uncertain timeline: the core scientists matter most in years three to seven, exactly when a four-year grant runs out. The answer is usually not a longer initial schedule, which candidates dislike, but a planned refresh programme (5.05).

Who this is for: founders writing their first option grants or equity policy; anyone asked by a candidate to change the cliff or schedule.

Pairs with: 3.06 (founder vesting and acceleration), 5.02 (plan design), 5.05 (refreshes), 5.09 (leavers and exercise windows).

2. What the evidence says#

Four years with a one-year cliff dominates. Carta reported in 2023 that over 90% of employees it sees receive a four-year vest with a one-year cliff. Ravio's European benchmarking data shows the same schedule with monthly vesting as the most common, used by 62% of companies in its dataset. Carta's guidance also describes graded monthly or quarterly vesting as the most common pattern after the cliff.

Most employees don't finish their grant. Carta's Peter Walker reported that over 90% of startup employees on Carta do not stay for the full four-year vesting period. Ravio's 2024 data put median tenure in European tech startups at 1 year and 9 months. Practitioner summaries of Carta's 2025 data estimate that about 22% of employees leave before the one-year cliff and forfeit their entire grant.

What that means: the cliff does a lot of work. It returns a meaningful share of options to the pool, and it filters early mismatches. But the four-year period is mostly a ceiling, not a typical outcome, so the long-term retention of key people depends on refresh grants rather than the initial schedule.

Back-loading works for big companies with liquid stock. Amazon vests 5% in year one, 15% in year two and 40% in each of years three and four; Snap uses 10/20/30/40. These schedules work because the stock is liquid and new grants are made every year. In startups with illiquid options and uncertain outcomes, back-loading looks like a bait-andswitch to candidates, and Carta and others note it is rare in venture-backed startups.

Longer vesting is under discussion. Peter Walker has argued that the data support moving toward five- or six-year vesting schedules to match how long building a company actually takes. That argument is strongest for founders and executives in deeptech, where exits commonly come eight to twelve years after founding. For individual contributors, a four-year grant plus refreshes usually serves the same purpose with less recruiting friction.

Each grant vests on its own schedule. Carta's example: a four-year grant in 2025 and a second four-year grant in 2027 are not fully vested until 2031. Layered refresh grants are how companies keep unvested value in front of their best people.

3. The method#

Step 1: Set a default and write it down#

Adopt one standard schedule for new-hire grants, written into your equity policy and offer templates:

• 4 years, 1-year cliff, monthly vesting after the cliff (quarterly is fine if your cap table tool or Indian administration prefers it).
• Vesting starts on the employee's start date, not the grant date, so delays in board approvals don't cost the employee.
• Unvested options return to the pool on leaving.

A default avoids negotiating every offer and is easier to administer.

Step 2: Decide when to deviate#

Avoid back-loaded schedules for new hires in a startup. If you want stronger retention in later years, use refresh grants instead.

Step 3: Handle acceleration deliberately#

• Employees: generally no acceleration; the plan lets the board accelerate on a sale if it chooses.
• Executives and key scientists: double-trigger acceleration (for example 100% or 12 months' worth of unvested options if terminated without cause within 12 months after a sale) is common and investor-acceptable.
• Single-trigger acceleration (vesting on the sale alone) worries acquirers, who want the team to stay. Avoid it except in rare cases.

Step 4: Explain the schedule#

Show each employee a simple table: how many options vest when, and what happens if they leave at months 6, 12, 24 and 48 (5.11).

Worked example#

An Indian quantum-sensing startup is hiring:

• A senior photonics engineer from a large company, leaving ₹40 lakh of unvested RSUs. She asks for a 6-month cliff. Indian ESOP rules require at least one year between grant and vesting (section 4), so the company keeps the 1-year cliff on her 0.6% grant and instead pays a one-time cash sign-on bonus, repayable pro rata if she leaves within a year.
• A research scientist who worked part-time with the founders for 9 months before the seed round. Her grant is made with the shareholders' approval at the seed close, and vests in yearly tranches over 3 years instead of 4, recognising the earlier work while staying within the one-year minimum.
• The head of hardware, hired after seed. 1.5% on 4-year vesting, with 20% of the grant vesting on a milestone, the first sensor shipped to a paying customer (no earlier than one year after grant). Double-trigger acceleration of 12 months on a sale.
• Two years later, the company introduces a refresh programme: top performers whose grants are more than 50% vested get new grants on a 3-year schedule with yearly vesting.

If the company had a Delaware parent and granted options there, the 6-month cliff and backdated vesting start would be possible for US-plan grants, but options given to Indian employees would still be taxed under Indian rules at exercise (5.08).

4. Country notes: India, US, UK#

India.

• Minimum vesting of one year. Rule 12(6)(a) of the Share Capital and Debentures Rules requires at least one year between the grant of options and their vesting. A cliff shorter than a year, or crediting earlier service by backdating, does not work for Indian ESOPs. Use a shorter total schedule, or cash, to recognise prior work.
• Vesting can be time-based or performance-based, and the scheme must say which, along with any lock-in after exercise.
• Death and permanent incapacity: the rules require options to vest immediately on death or permanent incapacity, and give legal heirs or the employee the right to exercise.
• Founder vesting is usually done through the SHA (reverse vesting or a buyback right on unvested shares) rather than the ESOP, because founders typically hold shares, not options (3.06).

US.

• No legal minimum. Schedules are entirely contractual.
• Early exercise lets employees buy unvested shares (subject to repurchase if they leave) and file an 83(b) election within 30 days, starting the capital gains holding period early. Useful when the strike price is low.
• ISO limit: if a schedule accelerates vesting so that more than $100,000 of options (by grant-date value) first become exercisable in a year, the excess becomes NSOs. Acceleration on a sale often triggers this.
• Section 409A generally does not apply to options granted at fair market value, so changing vesting is usually tax-neutral, but modifying a grant can create a new grant date for ISO purposes.

UK.

• No legal minimum for EMI or unapproved options.
• EMI options are often exit-only: many UK startups make EMI options exercisable only on a sale or IPO, in addition to time vesting, to avoid employees needing cash before a liquidity event. With the 15-year exercise limit from April 2026, exit-only options can stay valid longer.
• Leavers: good and bad leaver definitions in the articles and plan govern what happens to vested options; EMI relief is lost if options are not exercised within 90 days of leaving (5.09).

Elsewhere. Some countries tax options at vesting (or grant), so a vesting event can create a tax bill without liquidity. Check before granting to employees abroad (5.18).

5. Scripts, mistakes, related playbooks, sources#

Scripts

Explaining the standard schedule in an offer:

"Your options vest over four years. Nothing vests in the first year; on your first anniversary a quarter vests, then the rest vests every month for three years. If you leave before a year, the options go back to the pool. After that, what's vested is yours to exercise under the plan's rules."

When a candidate asks for a shorter cliff:

"We keep the same schedule for everyone so it's fair across the team. What we can do is recognise what you're leaving behind directly, with a sign-on bonus [or a separate small grant], rather than changing the cliff."

Introducing refreshes:

"Most of your original grant has vested. Because we want you here for the next stage, we're giving you a new grant that vests over the next three years, on top of what you already have."

Common mistakes#

1. Back-loaded schedules for startup hires, which read as unfair.
2. Ad hoc schedules negotiated per person, creating resentment and administrative mess.
3. Vesting from the grant date when board approval was delayed months after the start date.
4. Ignoring India's one-year minimum when promising short cliffs.
5. Single-trigger acceleration for everyone, which complicates acquisitions.
6. No refresh plan, so key scientists are fully vested and uncommitted in year four.
7. Milestones that are vague or subjective, leading to disputes.
8. Forgetting the ISO $100,000 limit when accelerating vesting.

Related playbooks#

• 3.06 Founder vesting, cliffs and acceleration
• 4.11 Advisor agreements (FAST template)
• 5.02 ESOP plan design: India (Companies Act rules) vs US (ISOs and NSOs)
• 5.05 ESOPs for second-line employees: grant bands and refreshes
• 5.09 Exercise windows and what happens when people leave
• 5.11 Explaining equity to employees: offer letters and equity education

Sources#

• Carta (Peter Walker), "Fewer than 10 percent of startup employees complete 4-year vesting" (September 2023); Carta, Vesting guide and UK guide on overlapping grants.
• Ravio, "What is the most common vesting schedule in European tech?" (62% use 4- year/1-year/monthly; median tenure 1 year 9 months; Amazon and Snap schedules).
• Peter Walker via Sierra Ventures on longer vesting.
• Practitioner summary of Carta 2025 departure data: Stealth Agents.
• India: Companies (Share Capital and Debentures) Rules 2014, rule 12(6). US: IRC ss.83(b), 422(d). UK: ITEPA 2003 Sch.5.