Every engagement begins with an assessment and plan.

Arrange an assessment

Accelerator founders

The 90 days after Demo Day: a go-to-market plan for YC founders

Demo Day ends the batch and starts the clock. Here is a 90-day go-to-market plan, in three phases, that turns founder-led traction into a system without taking the founder out of it.

Dineth Ratnayake

Founder of Codax · 11 March 2026 · 9 min read

A quarterly roadmap drawn on a whiteboard

The short answer

After Demo Day, close your seed round in a few weeks, then spend the next 90 days building a go-to-market system around the founder. Days 1 to 30 assess what drives demand, days 31 to 60 fix the foundation and build proof assets, and days 61 to 90 test channels against an agreed account list, with the founder kept as the voice throughout.

Key takeaways

  • Demo Day is built to start investor conversations, and YC's own advice is to keep the raise short.
  • Carta data puts the median time from seed to Series A at 2.2 years in 2025, so the post-Demo Day period is when the Series A story starts.
  • Run the 90 days as three phases: Assess, then Fix and Build, then Build and Test.
  • Keep the founder as the sender, the voice and the closer, and build the system around their time.
  • Day 90 should end with a repaired foundation, an agreed account list, written proof and live controlled tests.

Demo Day feels like a finish line. For three months the batch has pointed at one day, and then it arrives, the pitch goes well and the inbox fills with investors. The week after, the structure that held everything together is gone.

What you do in the next 90 days decides whether the round you are raising turns into a Series A. This article sets out a go-to-market plan for that period, in three 30-day phases, built so the founder stays the voice of the company.

What happens after YC Demo Day?

After Demo Day, the batch ends and the fundraise begins. YC describes Demo Day as the day its latest batch presents to an invite-only audience of approximately 1,500 investors and media, and that audience generates the meetings that follow.

YC's own guidance is clear that the pitch is a door, not a deal. Geoff Ralston's guide to Demo Day presentations says a short presentation will seldom convince investors to invest on the spot, and that the goal is to intrigue them enough that they want to meet you and learn more.

YC is a three-month programme that runs four times a year, and it notes that office hours do not stop when the programme does. The advice is still there after Demo Day. The batch structure and its deadlines are not. From here, the founders set their own pace, and the plan has to come from inside the company.

How long should fundraising take after Demo Day?

Fundraising after Demo Day should take weeks, not months. Sam Altman's fundraising advice for YC companies says a short process is enough for a seed round.

“a few weeks and 3 meetings per investor for a seed round is enough”
Sam Altman, Fundraising Advice for YC Companies, Y Combinator blog

In the same post he advises closing the first, say, $200k from the first reasonably good investors that offer it on reasonable terms, and reminds founders that the most important thing investors want to figure out at this stage is how much your users love you. The raise is a means to building the company, not a replacement for it.

That is why the go-to-market plan below starts on day 1. It runs alongside the raise, and it takes a few hours of founder time a week, not the whole calendar.

What mistakes do founders make after Demo Day?

The most common mistake is letting the raise take over the quarter. The second is buying reach before the company can convert it. Both feel productive and both spend runway that the Series A depends on.

  • Treating the round as the goal and pausing selling until it closes.
  • Hiring a first marketer or SDR before anyone knows what works.
  • Buying ads, outbound and content from separate providers with no shared account list.
  • Taking the founder out of outreach, so buyers stop hearing from the person they trust.
  • Measuring activity, such as posts and emails sent, instead of qualified pipeline.

Each of these trades long-term pipeline for short-term motion. The plan below avoids them by putting assessment before spend, and by keeping the founder at the centre of every channel.

How long do you have to get from seed to Series A?

You have about two years, and the odds are tighter than the headlines suggest. Carta's data puts the median time from seed to Series A at 2.2 years in 2025, and 1.9 years for AI companies.

2.2 years

Median time from seed to Series A in 2025, Carta

1.9 years

Median time from seed to Series A for AI companies in 2025, Carta

32.6%

Share of Q4 2019 seed cohort that reached Series A within 2 years, Carta

Carta's cohort data, covering 12,249 seed rounds from Q1 2018 to Q3 2025, shows how hard the step is. In the Q4 2019 cohort, 32.6% had reached Series A after two years, and no cohort has reached a 30% graduation rate since Q3 2021.

The companies that make it show repeatable pipeline beyond the founder's network. The 90 days after Demo Day are the cheapest time to start building it, because the team is small, the round is fresh and the founder still has the relationships that produced the first customers. Our guide to how to spend the YC $500K shows how to budget for it.

What does a 90-day go-to-market plan look like?

A 90-day go-to-market plan runs in three phases: assess what is driving demand, fix and build what blocks it, then test channels against an agreed account list. Each phase maps to Codax's method of Assess, Fix, Build and Test, set out in full on how we work.

The 90 days after Demo Day, phase by phase

  1. Days 1 to 30Assess. Read every signal in the business and the market before sending anything new. Close the seed round.
  2. Days 31 to 60Fix and Build. Repair the site, CRM, domains and tracking. Turn design partner proof into case studies and founder content.
  3. Days 61 to 90Build and Test. Start controlled four to six week channel tests against the account list, with the founder as the sender.
  4. Day 90Review. Read the first results account by account and set the plan for the next quarter.

The phases overlap in practice. In Codax's method, Fix runs from month one to month four, Build from month two to month six and Test from month two to month eight. The 90-day plan is the opening stretch of that sequence, not the whole of it. Scale, where budget and volume move to what converted, starts from month five, after the first tests have been read. In the agentic AI healthcare engagement, Assess, Fix and Build were done in two months.

Days 1 to 30: what should you assess first?

Assess where your first customers really came from and what is stopping the next ones. Spend the month reading signals before you send a single new campaign.

Start inside the business. Look at revenue deal by deal, what the founder hears on calls, how design partners use the product and why each customer bought. Then look outside it at who is researching your category, which accounts engage on which channels and how you appear in search and AI answers.

  • Every closed deal and design partner, with how it started and who championed it.
  • The words customers use to describe the problem, taken from call recordings.
  • Site, CRM, sending domains and tracking, checked for anything that breaks measurement.
  • A first account list of companies that look like your best customers.
  • Competitor positioning and where buyers already compare you.

The founder's job in this phase is to give access and context: call recordings, CRM, the story of each early deal and an hour or two of interview time. Everything else is reading and analysis.

The output is a written report, a prioritised repair list and a first read on the account list. This is the same assessment every Codax engagement begins with, and it stops the next 60 days being spent on guesses.

Days 31 to 60: what needs fixing and building?

Fix whatever blocks demand from converting, and build written proof from the customers you already have. Do both before you buy any reach.

Fixing covers the site, the brand, the founder's executive presence, CRM and lead definitions, sending domains, tracking and collateral. None of it is glamorous. All of it decides whether the tests in the next phase produce a result you can trust.

  • A site that explains the problem in the customer's words and loads without errors.
  • One written definition of a qualified lead, used by everyone who touches pipeline.
  • Warmed sending domains, so outbound reaches inboxes and protects the main domain.
  • Tracking that ties every account's engagement back to its source.
  • A founder profile that a CIO or head of operations would trust before a first call.

Building turns design partner wins into case studies, the founder's thinking into executive content and newsletters, and the first webinar or partner programme into an inbound path. In the cybersecurity services engagement, three months of repair came before the first outbound sequence, and yearly qualified pipeline grew from $548K to $2.2M in twelve months.

The founder's job in this phase is to review and approve. Case studies need their sign-off and design partners' consent. Founder content comes from the recorded interviews, written up by the team and checked by the founder before it goes out in their name.

Days 61 to 90: how do you test channels without wasting runway?

Test each channel as a short, controlled experiment against an agreed account list. Change one variable at a time, read the result with whoever sells and move money only to what converted.

How to run a controlled channel test

  1. Agree the account list

    Fix the accounts and roles the test will reach, and agree them with whoever sells.

  2. Choose one variable

    Test one thing at a time: the sender, the role you lead with, the topic or the offer.

  3. Run for four to six weeks

    Long enough for buying groups to respond, short enough to protect runway.

  4. Send replies to sales the same day

    Every positive reply and signal reaches the founder while the account is still warm.

  5. Read results account by account

    Judge each test on qualified pipeline, not opens or clicks.

  6. Move budget to what converted

    Scale the winning variant and retire the rest at the next monthly review.

Small changes produce large differences. In the healthcare engagement, leading with the CMIO produced 2.1x the meetings, and a health system speaker on webinars produced 2.6x the registrations. In the cybersecurity engagement, a SOC operations webinar topic drew 2x the registrations of a compliance topic.

Pick the first tests from what the assessment found. If early customers all came through one buyer role, test leading with that role against another. If the founder's posts already draw replies, test the founder as the sender against the company. Keep the number of live tests small enough that each one can be read cleanly.

How do you keep the founder as the voice while the system gets built?

Keep the founder as the sender, the speaker and the closer, and build everything else around a small, fixed amount of their time. Founder-led sales is right at this stage. It just cannot run on the founder's calendar alone.

Codax asks a founder for 1 to 2 hours a week, captured from recorded interviews. That time becomes posts, newsletters, invitations and outreach in the founder's name, so buyers keep hearing from the person they trust while a team does the writing, targeting and measurement.

The results show why this matters. In the healthcare engagement, the CEO as the sender produced 3x the replies, and the company closed $1.02M ARR in seven months from LinkedIn and email, channels that had produced none, with 0 hires added. In the cybersecurity engagement, founder-signed invitations produced 3x the acceptances and founder ads beat company ads by 2.4x on click-through.

A growth department is one senior team that owns qualified pipeline end to end, from strategy to execution, under a single accountable lead. That is the system that keeps the founder's voice while removing the founder from every task behind it. Our article on scaling the founder rather than founder-led sales goes further on this.

What should you have at the end of 90 days?

At day 90 you should have a closed seed round, a repaired foundation, an agreed account list, written proof and live tests producing qualified pipeline from accounts outside the founder's network.

  1. A written assessment and a repair list with most critical items closed.
  2. An account list agreed with whoever sells, with buying groups mapped.
  3. Case studies from design partners and a steady flow of founder content.
  4. Two or three controlled tests running, each with one variable.
  5. A monthly pipeline review, account by account, with leadership.

This is also what a Series A investor will ask to see later: where pipeline comes from, how much of it sits outside the founder's network and which channels convert. Starting the record at day 1 means that, by the time you raise, you have many months of it rather than a few weeks.

From there, budget and volume move to what converted, and the Series A story builds month by month. Our guide from design partners to 100 customers covers the next stage, and the full series is on the accelerator founders hub.

Questions and answers

What should YC founders do after Demo Day?

Close the seed round quickly, then build a go-to-market system around the founder. Sam Altman's advice is that a few weeks and three meetings per investor is enough for a seed round. Use the following 90 days to assess, fix, build and test.

How many investors attend YC Demo Day?

YC says its latest batch presents to an invite-only audience of approximately 1,500 investors and media. The aim of the pitch, in YC's own guidance, is to get investors interested enough to meet you.

How long does it take to go from seed to Series A?

Carta's data puts the median at 2.2 years in 2025, and 1.9 years for AI companies. In Carta's Q4 2019 seed cohort, 32.6% had reached Series A after two years.

Should founders keep doing sales after YC?

Yes. The founder's voice still converts best at seed, so the founder should stay the sender, the speaker and the closer. What changes is that a system is built around a fixed amount of their time instead of relying on their whole calendar.

When should a startup start channel testing after Demo Day?

Start once the foundation is repaired and the account list is agreed, which in this plan is around day 61. Run each channel as a four to six week controlled test and move budget only to what produced qualified pipeline.

Sources

  1. Demo Day, Y Combinator
  2. About Y Combinator, Y Combinator
  3. A Guide to Demo Day Presentations, Y Combinator
  4. Fundraising Advice for YC Companies, Y Combinator
  5. Planning to Raise VC Every 18 Months Is Planning to Fail, Carta
  6. Most Seed Startups Never Reach Series A, Data Shows, Carta

Keep reading

More from the Newsroom

Accelerator founders
5 Aug 2026 · 9 min readHow to spend the YC $500K: a growth budget for your first 12 monthsThe YC deal gives you $500,000 and no rules on how to spend it. Here is a framework for splitting it across the first 12 months so it buys the evidence your next round needs.Read
Accelerator founders
17 Dec 2025 · 9 min readFounder-led sales does not scale. The founder can.Founder-led sales is the right way to start, and accelerators are right to insist on it. What breaks is the founder's calendar, not the founder's voice. Here is how to keep the voice and remove the ceiling.Read
Accelerator founders
4 Mar 2026 · 9 min readFrom 10 design partners to 100 customers: growing past your networkYour network gets you the first ten customers. The next ninety come from proof, a sharp account list and channels that make you known to strangers before they start looking.Read

Start here

Every engagement begins with an assessment of what is already running

Findings shared in full, with a prioritised repair list, before anything is agreed.

What you receive

  • A written report of everything found
  • A prioritised repair list
  • A first read on the account list
  • A recommended plan across the five phases
Start with an assessment