The first time most engineers or operators get a stock grant, the document is 47 pages long and the email summarizing it has three acronyms in it: ISO, NSO, RSU. The recruiter usually says “this is a standard package” and moves on. The new hire signs it, files it, and doesn’t think about it again until either the company sells, the company struggles, or it’s time for a refresh grant.
By then, the decisions that mattered most — when to exercise, whether to file an 83(b), how to manage the AMT exposure — have either been made or quietly missed.
This post is for founders setting up their first equity plan, for senior hires evaluating an offer, and for finance teams trying to explain to employees what they actually have. The goal: a clear, operator-grade view of how ISOs, NSOs, and RSUs each work, what each one is good and bad for, and what to actually offer at Series A and B.
The three instruments, in plain English
Almost every equity grant at a venture-backed startup is one of three things.
Incentive Stock Options (ISOs) give the employee the right to buy shares at a fixed price (the strike price) within a defined window. They’re called “incentive” because they get preferred tax treatment if held long enough — but that treatment comes with the AMT (Alternative Minimum Tax) trap that catches many employees off guard. ISOs can only be granted to employees, not contractors or board members.
Non-Qualified Stock Options (NSOs) also give the right to buy shares at a strike price. The difference is purely tax: NSOs are taxed as ordinary income at the moment of exercise, on the spread between strike price and the then-current 409A value. No AMT trap, but no preferential treatment either. NSOs can be granted to anyone — employees, contractors, advisors, board members.
Restricted Stock Units (RSUs) are different in kind, not just in tax. An RSU is a promise to deliver shares (or cash equivalent) when certain conditions are met — usually a vesting date plus, at private companies, a liquidity event (an IPO or acquisition). There’s nothing for the employee to “exercise.” When the conditions hit, the shares show up, and the employee owes ordinary income tax on the full value at that moment.
Three instruments, three tax regimes, three different timing considerations. The choice between them isn’t cosmetic.
ISOs: the default at Series A and B
For most venture-backed startups in the Series A/B stage, ISOs are what you grant to full-time employees. The reason is straightforward: when used correctly, ISOs produce the best after-tax outcome for an employee who holds long enough.
How ISOs work mechanically:
The employee is granted the right to buy a certain number of shares at the strike price set at grant (which is the 409A fair market value at that date). The grant typically vests over 4 years with a 1-year cliff — meaning the employee gets nothing until they’ve been there a year, then the remaining 75% vests monthly or quarterly over years 2–4.
The employee can exercise vested options at any time before the expiration of the grant (usually 10 years from grant date, or 90 days from termination if they leave the company — more on that below). To exercise, the employee pays the company the strike price multiplied by the number of shares being exercised. The company issues the shares.
The tax piece that makes ISOs valuable:
If the employee holds the shares for at least 2 years from grant date AND at least 1 year from exercise date, the eventual gain on sale is taxed at long-term capital gains rates (currently 15–20% for most employees) rather than ordinary income rates (up to 37%). This is the “qualifying disposition” — the magic outcome ISOs are designed for.
If the employee sells the shares before meeting both holding periods, it becomes a “disqualifying disposition” and the gain is taxed as ordinary income, just like an NSO would be. The tax advantage disappears.
The AMT trap:
This is the part most employees miss. When an employee exercises ISOs, the spread between strike price and the then-current 409A value (the “bargain element”) doesn’t trigger regular income tax — but it does count as income for AMT purposes.
A simple example. An employee exercises 100,000 ISOs with a strike price of $1 when the 409A is $10. The “bargain element” is $900,000 — the difference between what they paid ($100K) and what the shares are worth on paper ($1M). For regular tax purposes: $0 tax owed. For AMT purposes: that $900K gets pulled into the AMT calculation, and the employee may owe AMT in the six figures even though they haven’t sold a single share.
For private company employees, this is brutal: AMT is owed in cash by April 15th, but the shares can’t be sold to cover it because there’s no public market. We’ve seen employees take out personal loans, sell other assets, or skip exercising altogether because of AMT exposure they didn’t understand at grant time.
When ISOs make sense:
For most full-time employees at Series A and B startups, ISOs are the right instrument. The tax advantage is real, the AMT risk can be managed with planning, and the alignment between long-term holding and the company’s outcome is structurally healthy.
When ISOs don’t work:
- Contractors, advisors, and board members can’t receive ISOs (employees only). They get NSOs by default.
- Grants over $100,000 worth of ISOs vesting in a single year get reclassified as NSOs above that threshold (the “$100K rule”). Most senior executives at later-stage companies hit this and have a mix of ISOs and NSOs in their grant.
- Employees who plan to exercise and sell immediately (rather than hold) get no benefit from ISO treatment — they trigger a disqualifying disposition and get NSO-equivalent tax treatment anyway.
NSOs: the default for non-employees and large grants
NSOs are the simpler instrument from a tax perspective. There’s no holding period magic and no AMT trap — but also no preferential treatment.
How NSOs work:
Mechanically identical to ISOs: granted at a strike price, vests over time, can be exercised after vesting. The difference is what happens at exercise.
When the employee exercises NSOs, the spread between strike price and current 409A value is taxed as ordinary income, immediately. Same example as above — 100,000 options, strike $1, 409A $10, spread $900K — the employee owes ordinary income tax on that $900K in the year of exercise. At a 35% effective rate, that’s $315K of tax owed in cash.
The company also has a corresponding tax deduction for the same amount, which is one of the rare upsides for the company in granting NSOs vs. ISOs.
After exercise, the employee owns the shares with a cost basis equal to the 409A value at exercise. Subsequent gains are taxed at long-term capital gains rates if held more than a year, short-term if not.
When NSOs are the right choice:
- Contractors, advisors, and board members (ISOs aren’t available to them)
- Grants above the $100K-per-year ISO limit (the excess automatically becomes NSO)
- Employees in countries where NSO treatment is more favorable than ISO
- Situations where the employee plans to early-exercise immediately at grant, when the spread is zero (more on this below)
The early-exercise NSO play:
Some startups allow employees to “early exercise” their NSOs — meaning they can exercise the unvested portion at grant, when the strike price equals the 409A value and the spread is zero. The employee pays the strike price upfront, files an 83(b) election (more on that next), and now owns restricted stock that vests over time instead of options.
The economic logic: if you exercise at zero spread and file 83(b), there’s no income tax at exercise, no AMT, and the long-term capital gains clock starts immediately. Every dollar of future appreciation is taxed at long-term rates.
The risk: the employee pays the strike price up front, in cash. If they leave the company before vesting, they get their unvested shares bought back at the strike price — but they’ve tied up cash in the meantime. If the company fails, the cash is lost.
This is a sophisticated play that makes sense for early hires with cash to deploy and high conviction in the company. It is not the default move.
RSUs: when to use them and when not to
RSUs are common at large public companies and at pre-IPO companies preparing for a public exit. They’re less common at Series A and B startups, and there’s a reason for that.
How RSUs work:
Instead of granting the right to buy shares, RSUs grant a promise to deliver shares (or cash equivalent) when conditions are met. There’s nothing to exercise, no strike price, no money for the employee to pay.
Standard RSU vesting at a public company is time-based: a 4-year vest with quarterly tranches, for example. At each vesting event, the employee receives shares, and the full market value of those shares is taxed as ordinary income at that moment. The company typically withholds shares at vest to cover the tax (called “sell-to-cover”).
At a private company, RSUs usually have a double-trigger structure: time-based vesting plus a liquidity event (IPO or acquisition). The shares don’t actually deliver — and the tax doesn’t hit — until both conditions are met. This solves the problem of employees owing tax on illiquid stock.
When RSUs make sense at startups:
- Late-stage pre-IPO companies (typically Series D or later) where an IPO is plausible within the vesting horizon
- Senior executive hires where the company wants to grant a large amount of equity without the AMT and exercise-cost complexity of options
- Public companies (where stock options are largely obsolete in favor of RSUs)
- Companies that have grown enough that the 409A value is high relative to the strike price that would be set — at that point, options become economically less attractive than RSUs
Why most Series A/B startups don’t use RSUs:
- Options give employees more leverage. A $1 strike on a $1 409A is a 100% upside-only bet. RSUs are just shares — they’re already worth the 409A value at grant.
- Tax timing is worse for employees at private companies, even with double-trigger structures.
- The dilution math is different — RSUs deliver shares with no cash inflow to the company, while option exercises bring cash in.
For most Series A and B startups, options (ISOs primarily, with NSOs for the exceptions noted above) are the right choice. The shift to RSUs happens later, typically Series C or later, when valuation has scaled enough that options become uneconomical for new hires.
The 83(b) election: the form that costs nothing and saves everything
This one comes up constantly and most employees don’t know it exists until it’s too late.
What it is:
If an employee receives restricted stock (either through early-exercising options or through a direct grant of restricted stock), they can file an 83(b) election with the IRS within 30 days of the grant or exercise. The election says: tax me now, on the current value, instead of taxing me later when the stock vests.
For early-stage startups where the current value (strike price = 409A) is low, this means the employee pays tax on a tiny amount today rather than a potentially massive amount as the stock appreciates and vests. And the long-term capital gains holding period starts immediately, not at each vesting event.
Why missing it is catastrophic:
Founders who early-vest their own stock at incorporation, employees who early-exercise options without filing 83(b), early hires who get founder-style restricted stock grants — all of these scenarios depend on the 83(b) election to work economically. Miss the 30-day window and you’ve turned what should have been a small one-time tax bill into a series of much larger ones every vesting event.
There’s no extension. There’s no late-filing relief. The IRS does not care. 30 days, postmarked, certified mail, with proof.
The mechanics:
Print the form. Sign it. Mail it to the IRS service center listed in the instructions, certified mail with return receipt requested. Keep a copy with your tax records. Give a copy to the company. That’s it. The whole thing takes 15 minutes and the postage costs about $8.
The cost of getting this wrong is sometimes six figures of unnecessary tax over the life of a grant. The cost of doing it right is $8 and a trip to the post office.
Other terms that matter
A few mechanics that don’t fit neatly into the three-instrument framework but matter to anyone with a grant.
4-year vest with 1-year cliff. The industry default. Employee gets nothing until the 1-year anniversary, then 25% vests at the cliff and the remaining 75% vests monthly or quarterly over years 2–4. Some companies are experimenting with 3-year, 5-year, or back-loaded vesting; the 4/1 standard is what employees and recruiters expect.
Post-termination exercise window. Standard is 90 days from the last day of employment to exercise vested options. Miss this window and the vested options are forfeited. For employees with substantial vested-but-unexercised ISOs, this can mean choosing between exercising into a six-figure AMT bill or walking away from years of equity. Some startups offer extended exercise windows (3, 5, 7, or 10 years) as a recruiting advantage — Pinterest, Quora, and Coinbase notably extended theirs years ago. Worth offering if your stage permits.
Single-trigger vs. double-trigger acceleration. What happens to unvested equity if the company is acquired? Single-trigger means equity automatically vests upon acquisition. Double-trigger means equity accelerates only if the employee is terminated after the acquisition. Most founders get double-trigger for themselves; some senior hires negotiate it. Standard employees typically don’t get acceleration at all.
Refresh grants. The original 4-year grant is the front-loaded incentive. Refresh grants are what keep employees engaged through year 5 and beyond. The market norm is annual or biennial refresh grants of 25–50% of the original grant size for strong performers. Without refresh, your senior employees see their unvested equity dwindle to zero in year 4 and start interviewing.
What we recommend at Series A and B
Pulling it all together, the standard equity comp playbook for a venture-backed startup at Series A or B looks like this:
For employees:
- ISOs for everyone, by default
- Standard 4-year vest with 1-year cliff
- Strike price = current 409A value (refreshed at least annually, and after any material event)
- 90-day post-termination exercise window, OR an extended window if you can afford the dilution implications
- Double-trigger acceleration for executives, no acceleration for line employees
For contractors, advisors, and board members:
- NSOs (since ISOs aren’t available)
- Vesting tied to deliverables for advisors and board members (often shorter than 4 years)
For very early hires who want to early-exercise:
- Allow it. Make sure they understand 83(b) and the 30-day deadline. Document the conversation.
For refresh grants:
- ISOs again for employees. Plan to do them annually for strong performers — biennially at minimum.
Don’t offer RSUs at Series A/B unless there’s a specific structural reason. The options model serves employees better at your stage.
What founders and finance teams actually get wrong
A few patterns that show up repeatedly in our work:
Mistake 1: Granting ad-hoc without bands. Every new hire negotiates equity individually, and the resulting cap table is chaotic. Set bands by role and level. Publish them internally. Stick to them. Adhocracy in equity comp creates years of cleanup work.
Mistake 2: Not refreshing the 409A often enough. A stale 409A creates risk in two directions: if the real fair market value has gone up and you’re still granting at the old strike, you’re issuing options that the IRS may later classify as below-FMV (a Section 409A violation, with brutal tax consequences for the employee). If the real value has come down, you’re issuing options at strike prices nobody will exercise. Refresh annually at minimum, and after any material event (priced round, major hire, significant revenue change).
Mistake 3: Missing the 83(b) window for early-exercising employees. Every time. Set up a process: employee early-exercises → finance team sends them the 83(b) form within 24 hours → certified mail confirmation within 30 days → copy in employee file and company file.
Mistake 4: Surprising employees with AMT at exercise. Build AMT projections into the exercise process. When an employee asks to exercise, the finance team should be able to model the AMT impact for them based on current 409A and the size of the exercise. The information shouldn’t come from their accountant in March of the following year.
Mistake 5: Treating the option pool as the dilution that doesn’t count. Option pool refreshes at each round are real dilution. Investors expect them, but founders should manage them deliberately. A typical Series A round includes a 10–15% option pool refresh pre-money, which means founders are bearing most of that dilution.
When to bring in operator support
Equity compensation is one of the highest-leverage finance topics at a startup. A few hours of operator input here typically saves six-figure tax and recruiting problems down the line.
You probably don’t need outside help if you have a strong in-house head of people, an experienced general counsel handling equity admin, and a cap table provider managing the mechanics.
You likely do want operator support if you’re:
- Setting up your first equity plan from scratch
- Hitting a milestone (Series B, first executive hire, secondary tender offer, IPO prep) that materially changes your equity comp design
- Designing refresh grant policy for the first time
- Trying to figure out whether to offer extended post-termination exercise windows
- Dealing with an inherited equity mess from a prior CFO or admin
Pegacorn Group works with venture-backed startups on equity compensation design, cap table hygiene, and the operational mechanics that make sure none of this becomes a problem in your next audit or fundraise. If your equity plan needs a real review, let’s talk.
This post pairs with: Equity compensation 101, The Series A/B benefits stack, and Why your Series B isn’t worth 50x ARR.