Why ‘Why Now’ Is the Most Underrated Slide

Why now is the only slide that argues the opportunity has a clock on it. Every other slide argues the business is good. This one argues the business is good now and would have failed three years ago. That is the only claim explaining why a fund should own it this year rather than watch it.

Most founders fill it with trends. A trend gives you a direction. A window needs a level, a date, and a consequence.

Sequoia’s business-plan outline has carried the slide for years, and the instruction reads: set up the historical evolution of your category, define recent trends that make your solution possible.¹ Note the word. Trends. The template asks for a direction, and a direction is available to anyone with a browser.

The people on the other side of the table started reading it harder. Across 170 seed decks covering the second half of 2022 and the first half of 2023, DocSend measured a 65% rise in venture time spent on the “Why Now?” section while total time spent on decks fell about 20% year on year.² That slide gained scrutiny while the rest of the deck lost it. What sits on the slide did not improve.

Why these three models

This piece runs the Wire Model: score the features of the decision, route to a small ensemble of formal models, then force the ensemble to produce dated actions. The scores that mattered:

• Threshold and tipping (0.9). A why-now claim asserts that some quantity crossed a level, and that below the level the business could not exist.
• Strategic actors with asymmetric information (0.75). The founder knows more about the enabling condition than the investor does, and asserting urgency costs nothing.
• Regime-break risk (0.85). The slide claims the rules changed. Whether they did is an unobserved state that both parties infer late.
• Historical-analog density (0.7). Category-entry timing has a deep record, and most of that record is discouraging.
• Cognitive distortion (0.55). Real, and routed nowhere. See below.

That routes to three models: threshold and cascade (what makes now different), cheap talk (whether the claim transmits anything), and regime-switching (when either side can know). They span three outcome types, complex, equilibrium and cycle, so their errors point in different directions and partly cancel.

A behavioural fourth scored high enough to consider and was cut. Its lever, discounting a vivid trend story, already arrives through the cheap-talk constraint and through subtracting the counterfactual in CHAIN. A fourth card would return a move you already have.

The framework: a why-now claim is a dated crossing

1. The crossing: thresholds turn a slow variable into a sudden outcome

Granovetter gives the mechanic. Each actor moves once enough others have, and small differences in how thresholds sit across a population produce wildly different aggregate outcomes from almost identical starting conditions.³ Nothing happens for a long time. Then everything happens, and the input that changed was moving smoothly throughout.

Watts added the warning. On a sparse network of agents following a threshold rule, very large cascades occur rarely, and they are triggered by shocks indistinguishable in advance from shocks that produce nothing.⁴ The system looks robust for long stretches, then turns fragile without notice.

So the enabling condition never announces itself, and the honest form of the slide has four parts: the variable, the level, the month it crossed, and the thing that became possible on the other side.

Three crossings you can check.

Input cost. Inference priced at GPT-3.5-level performance on MMLU fell from US$20.00 per million tokens in November 2022 to US$0.07 per million tokens by October 2024, a reduction of more than 280-fold in roughly 18 months, with task-level declines running anywhere from 9-fold to 900-fold per year.⁵ Any product needing inference below a cent per interaction became buildable inside that window.

Permission. Kenya’s Digital Credit Providers Regulations came into force in March 2022. By April 2026 the Central Bank of Kenya had received more than 800 applications and licensed 227 providers, who had granted 7.5 million loans worth KSh 133.5 billion as of February 2026.⁶ That crossing changed which companies were allowed to exist, the most binding form a threshold takes.

Rails. Access to Kenya’s mobile money system lifted an estimated 194,000 households, about 2% of the country, out of poverty, through savings behaviour and occupational choice.⁷ That is the consequence one rail crossing carries, measured a decade later.

Name the variable, the level and the month it crossed, or you are describing a market and calling it a window.

Threshold and cascade, the crossing lens

Assumes: aggregate behaviour responds to a level being crossed, not to a direction of travel.

Fits because: threshold and tipping scored 0.9.

Breaks when: the variable you named was never the binding constraint. Cost collapses, adoption stays flat, and the real gate was trust or distribution.

Counteracts: treating a trend line as a reason to act.

May reinforce: single-variable thinking, and the belief that one crossing is sufficient.

2. The transmission: an unverifiable timing claim carries no information

Crawford and Sobel formalised what happens when a better-informed sender talks to a receiver who then acts for both of them. As their preferences diverge, equilibrium communication degrades into coarse partitions, and past a point the message carries no information at all.⁸

Urgency is where founder and investor preferences diverge most. You are paid to say the window is open now. They are paid to discount exactly that. So the default information content of a stated why-now is zero, and honesty has nothing to do with it. The sentence looks identical either way.

What restores transmission is a claim that could have come out the other way. Three properties, cheapest first:

• A number sourced outside your company, which a stranger can open in one click.
• A date the reader can check against the public record.
• A falsifier you write yourself: the observation that would prove the window is shut.

Founders under-price all three, for a measurable reason. In CB Insights’ analysis of 101 startup post-mortems, “Product Mis-Timed” was cited by 13% of failed founders and ranked tenth, behind poor marketing and lost focus.⁹ Bill Gross’ outside review of 200 companies put timing first, at 42% of the difference between success and failure.¹⁰ Take that second figure as directional, since it came from a ranking exercise rather than an identified study.

The gap between 13% and 42% is where the argument lives. Timing is the failure mode you cannot see from inside your own company. Founders who die of it file post-mortems blaming marketing. That is why the slide stays underrated on both sides of the table.

Cheap talk, the transmission lens

Assumes: the receiver knows your incentive and discounts your message accordingly.

Fits because: strategic actors with asymmetric information scored 0.75.

Breaks when: you have a costly, verified track record with this specific investor. Reputation substitutes for the signal being expensive in itself.

Counteracts: the belief that stating urgency creates urgency.

May reinforce: over-engineering the claim into a research report nobody asked to read.

3. The detection: everyone dates the regime break late

Regime-switching models treat the regime as an unobserved state, inferred from data with a probability attached, after the series has already broken.¹¹ Series shift abruptly around crises and policy changes, and the break gets its date retrospectively.

Both sides of the table run that filter. The investor runs it on public data, quarterly, across a portfolio. You run it on your own funnel, weekly, in one market, on raw observations rather than a summary.

Your sampling rate is higher. That is the founder’s whole structural advantage on this slide, and almost nobody spends it. Your own numbers straddling the crossing date are the only why-now an investor cannot buy in a research report. Cost per acquisition before the licence and after. Approval rate before the API and after. Same metric, same definition, twelve months either side of the month you named.

Then apply the base rate to yourself. Golder and Tellis ran a historical analysis of roughly 500 brands across 50 categories, including the non-survivors earlier databases had quietly excluded, and found 47% of market pioneers failed, while the eventual leaders entered on average 13 years after the pioneer.¹² Right about the crossing, early on the date, is the most common way to be wrong about timing.

Capital timing carries its own distribution. Nanda and Rhodes-Kropf found firms taking their first venture investment in hot markets are less likely to reach an IPO, yet conditional on going public are valued higher and hold more patents and citations.¹³ A funding window widens the outcome distribution at both ends.

Regime-switching, the detection lens

Assumes: the regime is unobserved and gets identified from data with a lag.

Fits because: regime-break risk scored 0.85.

Breaks when: the break is legislated. A licence regime arrives with a gazetted date, so there is nothing to infer and the advantage moves to whoever files first.

Counteracts: confidence that you can see the turn while standing in it.

May reinforce: paralysis, waiting for a confirmation that arrives after the window shuts.

GEER: the levers, cheapest first

Four channels carry this claim: the variable (what moved), the instrument (your own before-and-after data), the falsifier (what would kill it), the clock (who else gets in, and when). Pull the cheap reversible levers first.

1. Write the crossing in one sentence. “X crossed Y in [month, year], which makes Z viable for the first time.” Costs an hour.
2. Attach an external number. One source, one link, openable by a stranger. Costs an hour.
3. Pull your own series across the date. One metric, twelve months either side. Costs a day.
4. Write the falsifier into the deck. One line: if [observation], this window is shut. Costs an hour, and it is the highest-yield line on the slide.
5. State the clock. Who else can enter, what they need, how long it takes them. Costs half a day.
6. Acquire an artefact that only exists after the crossing. A licence under the new regime, an integration agreement, a rate card nobody could have signed 24 months ago. Costs weeks, and it is the only irreversible lever here.

No-lever flag: if you cannot name a variable, a level and a month, you do not have a why-now. Delete the slide, move the argument to execution, and stop paying rent on a claim you cannot support.

RADAR: the portfolio, dated

DO NOW, by T+3. Reversible, and dominant whether or not the window is real.

1. Write the one-sentence crossing. Variable, level, month, consequence.
2. Pull your own series across that date. If the series shows nothing, the crossing you named is not the binding constraint. Go find the one that is.
3. Delete every trend line you cannot connect to a line in your own accounts.
4. Write the falsifier and leave it in.

HEDGE, by T+14. Cheap insurance against the claim being contested.

1. Secure one third-party artefact obtainable only after the crossing.
2. Prepare the second question, which always arrives: why has nobody else done this. Answer with what a competitor would need, and how many months it takes them.
3. Build one alternative crossing on a different variable, in case the first is disputed.

DEFER AND TRIGGER. Irreversible, so wait, and pre-commit the trigger now.

1. Defer: capacity that only pays if the window is genuinely open. The field team, the warehouse lease, the capital held against a licence.
2. Trigger to commit: three of your first ten investor conversations repeat your crossing back to you, unprompted, in follow-up. That is evidence the claim transmitted. Commit the capacity in the two weeks after.
3. Counter-trigger: by T+28, zero conversations reference it. The claim is still cheap talk. Return to the variable rather than the wording.

If you are the one being pitched. DO NOW: ask for the variable, the level and the month, then ask what would falsify it. HEDGE: score the timing claim separately from the market-size claim, in writing, because the two get conflated and a large market is no clock. DEFER: revising your sector thesis until you have tracked dated crossings for two quarters and can compare claimed windows against realised ones.

CHAIN: what usually happens next

Match the reference class on structure. Entry into a category whose viability changed on a date runs the same machinery in retail, in telecoms after deregulation, in lending after a licence regime, in software after an input cost collapse.

The base rate is unkind. Pioneers fail 47% of the time, and the firms that end up leading arrive around 13 years later.¹² The modal outcome of being right about a crossing is that somebody else monetises it.

Present-state modifiers bite hardest for African founders. African tech ventures raised US$2.41 billion in equity across 462 rounds in 2025, with South Africa, Kenya, Nigeria and Egypt taking 81% of equity funding against 67% the year before, while unique equity investors fell 7% to 539.¹⁴ Fewer people can act on your window, and they sit in four markets.

Second-order consequence: a verifiable crossing becomes a filter, and founders who can produce one get a hearing that trend-deck founders stop getting. Third-order: the format saturates, everyone writes the dated sentence, and the separating evidence moves down to the instrument, the founder’s own series across the date. Build for that now and you are two years ahead of the format.

Subtract the counterfactual before crediting the slide. Teams that reason well about a crossing tend to reason well about pricing, hiring and cash, and would have outperformed without ever making the argument. Price the slide as correlation.

Matrix-break flag. A general-purpose enabling condition grants permission to everyone on the same morning. Inference costs falling more than 280-fold reached you and the incumbent whose distribution you hoped to outrun on the same day.⁵ Where that holds, the timing argument decays into a table stake within about two funding cycles, and the separating question becomes what you built inside the window that is expensive to copy.

What this ensemble cannot see

Four things, and each of them can be the whole answer.

The binding constraint. You can be right that a variable crossed and wrong that it was the one holding the market shut. Costs fall, adoption stays flat, and the real gate was trust or somebody’s settlement cycle.

The investor’s own clock. Fund vintage, deployment pace, the reserve decision taken last Tuesday. Their timing is invisible from where you sit, often decides the outcome, and gets misread as a verdict on the window.

Whether the window is already priced. By the time a crossing is legible enough to put on a slide, it may sit inside the entry valuation. Nothing here tells you whether you are early to the market and late to the round.

Silent crossings. Every case above is a threshold that produced something. Thresholds crossed that led nowhere generate no case studies, and they are the majority. Watts gives the formal version: cascades are rare, and the shocks that start them look, in advance, exactly like the shocks that fizzle.⁴

So run the ensemble on the two artefacts that survive all four blind spots. By T+14, produce one sentence naming the variable, the level and the month, and one chart of your own metric across that date. If the sentence holds and the chart shows nothing, you have a market and no window, so raise on execution and say so out loud. If both hold, move the slide to position three and watch how differently the rest of the deck gets read.

Sources and notes

1. Sequoia Capital, “Writing a Business Plan” pitch deck outline, listing Company Purpose, Problem, Solution, Why Now, Market Size, Competition, Product, Business Model, Team and Financials. The Why Now instruction reads: “Set-up the historical evolution of your category. Define recent trends that make your solution possible.” Reproduced copy hosted by the Gustavson School of Business, University of Victoria: pitch deck template PDF.
2. DocSend (Dropbox), annual seed report, announced via PR Newswire. 170 seed pitch decks analysed covering the second half of 2022 and the first half of 2023; a 65% increase in venture capital time spent reviewing the Why Now section; investors spent 20% less time reviewing pitch decks year on year. Press release.
3. Granovetter, M. “Threshold Models of Collective Behavior.” American Journal of Sociology 83(6), 1978, 1420-1443. Full text.
4. Watts, D. J. “A simple model of global cascades on random networks.” Proceedings of the National Academy of Sciences 99(9), 2002, 5766-5771. Global cascades occur rarely and may be triggered by shocks that are a priori indistinguishable from shocks that do not cascade. Author copy mirrored at the University of Vermont: full text PDF.
5. Stanford Institute for Human-Centered AI, Artificial Intelligence Index Report 2025, Chapter 1: Research and Development. Inference cost for a system scoring at GPT-3.5 level (64.8 on MMLU) fell from US$20.00 per million tokens in November 2022 to US$0.07 per million tokens by October 2024. Task-level inference price declines of 9-fold to 900-fold per year, analysis conducted with Epoch AI. Chapter PDF.
6. Central Bank of Kenya, press release “Licensing of Digital Credit Providers,” 14 April 2026. More than 800 applications received since March 2022; 227 DCPs licensed; licensed DCPs had granted 7.5 million loans valued at KSh 133.5 billion as of February 2026. Press release PDF.
7. Suri, T., and Jack, W. “The long-run poverty and gender impacts of mobile money.” Science 354(6317), 2016, 1288-1292. Access to the Kenyan mobile money system increased per capita consumption levels and lifted 194,000 households, or 2% of Kenyan households, out of poverty. Record and abstract.
8. Crawford, V. P., and Sobel, J. “Strategic Information Transmission.” Econometrica 50(6), 1982, 1431-1451. Author copy, full text.
9. CB Insights, “The Top 20 Reasons Startups Fail,” analysis of 101 startup failure post-mortems. Product Mis-Timed cited in 13% of post-mortems, ranked tenth; No Market Need first at 42%. Post-mortems allowed multiple reasons, so the percentages sum above 100. Report PDF.
10. Gross, B. “The single biggest reason why start-ups succeed.” TED 2015. Across 200 companies, 100 from Idealab and 100 outside it, timing accounted for 42% of the difference between success and failure, ahead of team and execution and ahead of the idea. Ranking exercise by the author rather than an identified econometric study; treat as directional. Talk page.
11. Hamilton, J. D. “Regime-Switching Models,” prepared for the Palgrave Dictionary of Economics, 2005. Economic series exhibit abrupt breaks associated with crises and policy changes; the regime is modelled as an unobserved state inferred probabilistically from the data. Author copy, full text.
12. Golder, P. N., and Tellis, G. J. “Pioneer Advantage: Marketing Logic or Marketing Legend?” Journal of Marketing Research 30(2), 1993, 158-170. Historical analysis of approximately 500 brands in 50 product categories, including non-survivors: pioneer failure rate 47%; early market leaders entered on average 13 years after pioneers. Author copy, full text.
13. Nanda, R., and Rhodes-Kropf, M. “Investment Cycles and Startup Innovation.” Journal of Financial Economics 110(2), 2013, 403-418. Working paper version, January 2012: firms receiving their initial venture investment in hot markets are less likely to IPO, but conditional on going public are valued higher, hold more patents and receive more citations. Working paper PDF.
14. Partech, 2025 Africa Tech Venture Capital Report. US$2.41 billion in equity funding across 462 equity rounds in 2025; South Africa, Kenya, Nigeria and Egypt captured 81% of equity funding against 67% in 2024 and 69% of equity deal activity; 539 unique equity investors participated, down 7% year on year. Report page.