A term sheet is two or three pages long, arrives as a PDF, and is usually read once, quickly, with all the attention going to one line: the valuation. That is the line least likely to determine what you walk away with.
Everything underneath it — preference, pool, ratchet, consents, drag — is where the money and the control actually get allocated. Founders who understand those six things negotiate a materially better deal than founders who only understand the headline number, and the difference tends to show up years later at the only moment that matters, which is the sale.
Most of it is not binding — and that matters both ways
A term sheet is generally a statement of intent rather than a contract. A handful of clauses usually are binding: confidentiality, who pays whose legal costs, and exclusivity — the no-shop period during which you agree not to talk to other investors. Exclusivity of four to eight weeks is normal. Longer than that, with no deposit and no deadline, means you have taken your fundraise off the market while somebody decides at their leisure.
The rest is not legally binding but is practically binding. Once a term is on the term sheet, arguing it back out during the long-form documents costs goodwill and legal fees. Negotiate at term sheet stage or accept it.
1. Liquidation preference
The single most important clause and the one most often skimmed. It sets who gets paid first when the company is sold, and how much they take before anyone else sees anything.
The market standard in the UK is a 1x non-participating preference: on an exit the investor takes the greater of their money back or their percentage share, not both. Participating preference means they take their money back and then share in what is left.
Put numbers on it. An investor puts in £1m at a £4m pre-money valuation, so £5m post-money and 20% of the company. Sell the business for £6m and, on a 1x non-participating preference, the investor takes the higher of £1m or 20% of £6m, which is £1.2m; founders and everyone else split £4.8m. On a participating preference they take their £1m off the top and then 20% of the remaining £5m, which is £2m in total, leaving £4m. Same valuation, same percentage, £800,000 of difference in the clause nobody read.
Now sell for £3m. Non-participating: the investor takes £1m, the rest of you split £2m. The preference does its real work on the downside, which is exactly why investors care about it more than they care about arguing over the valuation.
Ask the question this way: at a sale price of £3m, £6m and £15m, how much do I personally receive? If the term sheet cannot answer that in five minutes, you do not yet understand your own deal.
2. The option pool, and where it comes from
Investors will want an employee option pool, typically 10% to 15%, to hire the team the plan depends on. Reasonable. The question is whether it comes out of the pre-money or the post-money.
Standard practice is pre-money, meaning existing shareholders — you — are diluted by the whole pool before the investor's shares are counted. Take that same £4m pre-money and £1m raise. Insert a 10% post-deal pool from the pre-money and the effective valuation of your existing shares drops to roughly £3.5m, because £500,000 of the £4m you thought you were being valued at is really being set aside for future employees. Nothing about that is dishonest, and it is entirely normal. But it means a £4m pre-money with a 15% pre-money pool is a worse deal than a £3.8m pre-money with no pool, and you should price the two against each other rather than reading only the bigger number.
3. Anti-dilution
This protects the investor if you later raise at a lower valuation. Broad-based weighted average is the customary and defensible version: the investor's conversion price adjusts partially, in proportion to how much new cheap stock was issued. A full ratchet reprices all their shares as if they had invested at the new lower price, which in a serious down round can transfer a startling amount of the company from founders to one investor. A full ratchet in a seed term sheet is worth pushing back on.
4. Control: board seats and consent matters
Ownership and control are different things, and founders conflate them constantly. You can hold 80% of the shares and still be unable to do a long list of things without investor consent — borrow money, hire above a salary threshold, change the business plan, issue shares, sell the company, or in some drafts approve the annual budget.
Read the consent matters list as an operating document, not a legal formality. Ask yourself which decisions you would realistically need to take at short notice in the next two years, then check whether each of them now requires a phone call to someone who may be on holiday. Board composition matters for the same reason: two founders, one investor director and an agreed independent is a very different company from two founders and two investor directors.
5. Drag-along and tag-along
Drag-along lets a defined majority force everyone else to sell on the same terms. It exists for a good reason — one small shareholder should not be able to block a sale everyone else wants — but the threshold is the negotiation. A drag that can be triggered without any founder support means the company can be sold out from under you. Tag-along is the mirror image and protects the minority: if the majority sells, you can join on the same terms. You want a sensible drag threshold and a solid tag.
6. Warranties, and your personal exposure
Founders are usually asked to give warranties personally about the state of the business — the accounts, the contracts, who owns the intellectual property, whether there is any litigation. Two numbers matter: the cap on your liability, which should be a multiple of what you actually receive rather than the whole investment, and the time limit. Full and accurate disclosure against the warranties is your protection, so the disclosure letter deserves more of your attention than it usually gets.
The tax relief trap sitting under all of it
This one is specific to UK deals and it catches people. SEIS and EIS shares must be full-risk ordinary shares that are not redeemable and carry no special rights to the company's assets — only limited, non-cumulative dividend rights are permitted. A conventional preference share structure with a preference on a winding up is therefore incompatible with the relief.
That matters because the relief is most of why UK angels invest at all: SEIS gives an investor 50% income tax relief on up to £200,000 in a tax year, and EIS gives 30% on up to £1m (or £2m where at least £1m goes into knowledge-intensive companies), with a three-year minimum holding period in both. On the company side SEIS is capped at £250,000 in total, with gross assets under £350,000 and a qualifying trade under three years old. EIS limits rose from 6 April 2026: a company can now raise up to £10m in any 12 months and £24m over its lifetime across the venture capital schemes, with lower limits of £5m and £12m for certain specified companies.
So if your term sheet says preference shares and your investors are relying on EIS, somebody has made a mistake, and it is better found now than after completion. Get advance assurance from HMRC before you issue anything. SEIS and EIS explained for founders raising money covers the qualifying conditions in more detail.
Before you sign
Model your own outcome at three exit prices, using the actual preference and pool in front of you, not the percentages. Get a corporate solicitor who does this weekly — a general commercial firm is not the same thing, and the fee is small against what a full ratchet costs. Ask the investor which clauses they consider negotiable; a good one will tell you honestly. And take up references on them by speaking to founders they have backed, including one whose company did not go well.
The other half of this is knowing what they are assessing before any of it is drafted, which what investors actually look for before they write a cheque covers, and what the relationship feels like afterwards, which is what happens after you say yes to an investor.
Common questions
Is a term sheet legally binding?
Mostly not, but partly yes. The commercial terms — valuation, preference, board seats — are normally expressed as a statement of intent, subject to due diligence and long-form documentation. A few clauses are usually drafted to bind immediately: confidentiality, an allocation of legal costs, and exclusivity or no-shop, which stops you talking to other investors for an agreed window. Treat the non-binding parts as practically binding anyway, because renegotiating an agreed term during the long-form documents costs goodwill and legal fees and rarely succeeds. The time to argue a clause is before you sign the term sheet, when you still have alternatives and the investor has not yet spent money on diligence.
What does a 1x non-participating liquidation preference actually mean for me?
It means that on a sale the investor takes the greater of their money back or their percentage share of the proceeds, not both. If an investor put in £1m for 20% and you sell for £6m, they take 20% — £1.2m — because that beats £1m, and the rest is split among everyone else. If you sell for £3m they take their £1m first and the remaining £2m is split. The participating version is the one to watch: there the investor takes £1m off the top and then their percentage of what is left, which on a £6m sale means £2m rather than £1.2m. Same headline valuation, very different outcome.
Why does the option pool come out of the pre-money valuation?
Because that way the dilution falls on existing shareholders rather than on the new investor, which is the market convention and is not in itself unfair — the pool exists to hire the team that delivers the plan the investor is funding. The practical effect is that your real pre-money valuation is lower than the number on the page. On a £4m pre-money with a 10% post-deal pool created beforehand, roughly £500,000 of that £4m is set aside for future employees, so your existing shares are effectively valued nearer £3.5m. Compare offers on that adjusted basis, and negotiate the size of the pool against a genuine hiring plan rather than accepting a round number.
Can preference shares cost my investors their EIS or SEIS relief?
Yes, and this is a common and expensive mistake in UK deals. SEIS and EIS require full-risk ordinary shares that are not redeemable and carry no special rights to the company's assets, with only limited non-cumulative dividend rights permitted. A conventional preference share carrying a preference on a winding up does not qualify. Since the relief — 50% income tax relief on up to £200,000 a year under SEIS, 30% on up to £1m under EIS — is often the main reason a UK angel is investing at all, discovering the incompatibility after completion is painful for everyone. Apply for advance assurance from HMRC before any shares are issued.
