What Investors Actually Think When They See Your Market-Size Slide

30 July 2026

What Investors Actually Think When They See Your Market-Size Slide

Most founders walk into deck preparation believing the market-size slide is where they make their case: name a big enough total addressable market, and the size of the opportunity does some of the persuading for you. It is an understandable instinct. It is also not how the person on the other side of the table is reading that slide.

An investor looking at your market-size numbers is not really asking whether this market is big enough. They are asking something closer to: does this founder understand where this number came from, and has anyone besides them actually behaved like a paying customer. Analysis of early-stage pitch decks (DocSend, US data, 2022) found investors spend somewhere around 30 to 40 seconds on the market-size section specifically, with those surveyed saying the number itself mattered far less than how it had been derived. The slide is not a size contest. It is a short, unforgiving test of your reasoning.

That reframing matters more now than it has for a while. British Business Bank data shows equity investment into UK smaller businesses fell 4 per cent to £12.3 billion in 2025, with seed-stage deal numbers down 27 per cent year on year even as AI companies alone took a record 44 per cent of total equity investment. Fewer cheques are being written at the stage most Foundry founders raise at, so each one has to survive more scrutiny of exactly the kind this piece covers: not whether your market is large, but whether your evidence for it, and for the demand sitting inside it, holds up at pre-seed and beyond.

The market-size slide is a reasoning test, not a size contest

The instinct to reach for the biggest defensible market figure is not irrational - a bigger number implies a bigger prize. The problem is what investors call the 1 per cent fallacy: the founder who says ‘we only need 1 per cent of a £50 billion market’ is, in an investor’s eyes, revealing that they have not done the harder work of counting real customers. What actually earns credibility is the relationship between the top-down figure and a bottom-up one built by counting reachable customers and multiplying by a realistic price. One practitioner rule of thumb, worth treating as exactly that rather than as research, holds that if the two estimates land within roughly double one another the underlying assumptions are probably sound, and if they are ten times apart something in the model is broken.

This is where TAM, SAM and SOM earn their reputation as the most consistently weak slide in a deck: one review of over a thousand decks by a deck-review firm found market sizing came up short more often than any other section, largely because founders present the top-down number as the argument itself rather than as a ceiling to check against something smaller and provable. Free UK sources make the bottom-up half of that exercise more achievable than founders often assume: ONS population and business-count data by industry code, Companies House filings on comparable businesses, and HMRC statistics can all be cross-referenced with trade body reports to build a number an investor can actually interrogate, rather than one borrowed wholesale from a market-research report neither of you can fully stand behind.

What investors are actually buying is evidence, not enthusiasm

There is a rough hierarchy to how much weight different kinds of validation carry, and it is worth being honest with yourself about where your own evidence sits on it. Cash on the table - pre-orders, paid pilots, deposits - sits at the top, because a customer risking money is hard to fake. Reputation risk comes next: a customer prepared to stake their professional standing on your product by, say, making an introduction or agreeing to a public case study. Time and data commitments follow, and compliments sit at the bottom - which is exactly what most founders at the idea-validation stage mistake for real traction.

The waitlist is the clearest example of this confusion in practice. A landing page collecting email addresses is not proof of a market; it is proof that a headline and a form worked. Practitioner benchmarks suggesting anything above roughly 25 per cent waitlist-to-engagement is a strong signal, and anything under 10 per cent a red flag, are useful as instinct-checks but have no real research base and should never be presented to an investor as findings. What does travel well is a story like Monzo’s: the company launched an Alpha in November 2015 to 3,000 prepaid cardholders and built a 20,000-person waitlist behind it, then ran a £1 million crowdfunding round with 6,500 pre-registrations before opening the product to that full waitlist. The waitlist was never the evidence on its own - it sat behind a live product already in real people’s hands, a detail founders retelling the Monzo story tend to leave out.

The say-do gap: why interviews alone will not convince anyone, including you

Rob Fitzpatrick’s Mom Test remains one of the clearest frameworks for customer conversations, and its central discipline is simple to state and hard to practise: talk about the customer’s life instead of your idea, ask about specific past behaviour rather than hypothetical futures, and listen considerably more than you talk. Fitzpatrick’s warning about three types of bad data - compliments, hypothetical enthusiasm, and wish-list feature requests - describes almost every set of customer discovery notes founders bring to an investor meeting expecting them to count as proof.

The reason this matters more than it sounds is a documented gap between what people say and what they do. Research into purchase-intention surveys (Chandon et al., cited via MarketingProfs) found the correlation between stated purchase intent and actual buying behaviour was 58 per cent stronger among people who had been surveyed than among comparable people who had not, meaning the act of asking someone whether they would use something inflates the very signal you are trying to measure. A run of enthusiastic interviews, without a customer ever giving up money, time, or reputation, is evaluation. It is not validation, and investors who have sat through enough pitches tend to spot the difference quickly, even when the founder in front of them cannot yet.

The UK’s own gatekeeper is quietly raising the evidence bar

It is worth remembering that HMRC is, in effect, a second audience for your market story before most UK investors ever see it. A large share of early-stage UK investment activity runs through SEIS or EIS-eligible structures, which makes Advance Assurance close to a functional gatekeeper for most angels and dedicated funds. SEIS currently allows qualifying companies to raise up to £250,000, a limit expanded in April 2023, with investors receiving 50 per cent income tax relief; EIS carries 30 per cent relief and considerably higher raise limits. HMRC statistics show 2,430 companies raised £276 million through SEIS in 2024-25, up 14 per cent year on year, while 3,735 companies raised £1.575 billion through EIS.

The direction of travel on scrutiny is the more interesting detail. HMRC received 4,085 SEIS Advance Assurance applications in 2025-26, of which 76 per cent had been approved at the time of reporting, down from an 85 per cent approval rate the year before, though some of that gap is likely a processing-lag artefact rather than a genuine tightening, and the final figure will probably move once outstanding applications clear. Either way, HMRC’s risk-to-capital condition means a regulator, not just an investor, is reading your business plan and asking whether the growth story stands up, which is exactly the kind of due diligence most founders assume only happens once a term sheet is on the table. Worth flagging too: London and the South East accounted for 66 per cent of SEIS investment in 2024-25, a regional concentration that founders elsewhere in the UK should factor into how they plan a raise. All of these figures move with policy, so treat them as a snapshot rather than a permanent rule.

Premature validation is its own trap

Founders who take validation seriously sometimes swing too far the other way, building before the underlying problem is properly understood. Startup Genome’s widely cited, if now dated, 2011 analysis of roughly 3,200 high-growth internet startups found about 70 per cent had scaled prematurely, and that companies which scaled at the right time grew around 20 times faster than those that jumped ahead of themselves. The same research found 80 per cent of ‘consistent’ companies spent their discovery phase understanding the problem space, against 60 per cent of inconsistent ones who moved straight to validating a product, and that the inconsistent group wrote 3.4 times more code during discovery than they needed to.

The practical lesson is not to distrust building altogether - it is to be honest about which phase you are actually in. An MVP built to test a specific, falsifiable assumption about customer behaviour is validation. An MVP built because interviews felt encouraging and momentum felt good is usually premature scaling wearing validation’s clothes. The Startup Genome data is old enough that it should be read as a caution rather than a rulebook, but the underlying pattern, discovery work skipped in favour of visible progress, shows up in founder behaviour just as often now as it did then.

Why ‘no market need’ is messier than it sounds, and that is useful to know

CB Insights’ frequently repeated claim that around 42 per cent of startups fail due to ‘no market need’ is treated by a lot of pitch advice as settled fact. It is not. Secondary sources quote the figure at 35 per cent, 42 per cent, and, following a 2024 reframing of the same underlying post-mortem data, 43 per cent citing poor product-market fit specifically, with the 70 per cent of failures that involve running out of cash treated as a downstream symptom rather than the root cause. The inconsistency is not a reason to distrust the general point, failing to find a market is a real and common cause of failure, but it is a reason to distrust anyone who states the number with more precision than the underlying research supports.

UK data offers a useful corrective to the more apocalyptic version of this story. ONS figures show 93.4 per cent of new UK businesses registered in 2019 survived their first year, 55.9 per cent reached three years, and 38.4 per cent reached five, undercutting the commonly repeated but untraceable claim that 20 per cent of UK businesses fail within twelve months. The ONS also recorded 280,375 business deaths in 2024, the large majority of which were retirements, mergers, or voluntary closures rather than failures in any meaningful sense, with only around 23,900 companies actually entering insolvency that year. None of this is an argument that failure is rare or that market research does not matter. It is an argument for reading statistics about why startups fail with the same scepticism you would want an investor to apply to your own market slide, rather than repeating whichever number is most dramatic.

It is also worth knowing that experienced investors do not fully agree on what predicts success in the first place. Marc Andreessen’s well-known argument is that a great market matters more than anything else, including team; Idealab founder Bill Gross, presenting an analysis of around 200 companies at TED in 2015, argued instead that timing explained roughly 42 per cent of the difference between success and failure, ahead of team and idea. Both rest on relatively small, non-random samples, and the genuine disagreement between them is more useful to a founder than picking a side would be: it suggests a market-size slide alone was never going to carry the argument, and that investors reading your deck are weighing market, timing, and team evidence together, not scoring one slide in isolation.

What good evidence actually looks like in the UK market right now

Beauhurst’s State of UK Investment data for the first half of 2025 found seed-stage companies led every other stage on deal volume, with 1,242 deals, while ranking bottom of all stages for total amount raised, at £1.16 billion - a large number of comparatively small cheques, competing on the strength of their evidence rather than the size of their market claims. That squeeze is compounded by the concentration of capital in AI, though Beauhurst’s separate first-time-deal data for 2025 found a genuine bright spot too: first-time deals rose 23.6 per cent year on year, with the total amount raised in first-time deals up 74.3 per cent to £6.27 billion, a record since 2020. Beauhurst’s Head of Research, Henry Whorwood, has described the resulting environment as a two-speed market, in which early-stage funding stays active but lean while later-stage rounds get fewer and considerably larger. Read together, the honest framing is not that UK early-stage funding has dried up, but that it has become more selective, which raises rather than lowers the bar for the evidence behind your numbers.

It is worth closing this section with a caution rather than a template. Paid commitment is not the same as proven demand, and two well-known examples make the point without needing UK equivalents: Juicero’s high-tech juice press failed in part because customers realised they could squeeze the packs by hand, and Quibi sold out $150 million of first-year advertising before launch, only to discover that pandemic-era viewers wanted long-form content rather than short mobile video. Investor money, advertiser commitments, and enthusiastic interviews can all point the same direction and still be wrong, which is exactly why investors keep asking founders to show their working rather than their conclusions.

Before your next investor conversation: what to fix first

Before you next sit across from an investor, get precise about what your market-size slide is actually resting on. Build the bottom-up number first, from real reachable customers and a defensible price, and let the top-down figure serve only as a ceiling check rather than the headline. Know exactly which pieces of your traction are cash, which are reputation risk, and which are simply compliments dressed up as interest, because an investor will make that distinction whether or not you have made it yourself first. As Graeme Donnelly of Quality Company Formations has put it, the founders who stand out are usually the ones who can describe the problem vividly enough that you feel it too, and that vividness almost always comes from real customer conversations rather than a market report.

If you are raising under SEIS or EIS, get your Advance Assurance position checked early and against current guidance rather than a template built for a different scheme year, since thresholds and approval patterns move. None of this guarantees a term sheet. What it changes is your investability: the perception of risk an investor forms before terms are even discussed. A founder who can show a bottom-up number, a named piece of paid or reputational validation, and an honest account of what they still do not know is not asking an investor to trust a feeling. They are asking them to check a set of numbers that already point the same way the founder does, which is a considerably easier, and more investable, conversation to have.

This article is for general information only and does not constitute financial, investment, tax, or legal advice. The Startup Foundry is not a regulated financial adviser. You should seek independent professional advice before making any financial, investment, or legal decisions.