The 3 team questions VCs now ask on first calls in 2026
The 3 team questions VCs now ask on first calls in 2026
Through May 16, 2026, the Crunchbase layoff tracker has recorded 137,547 people cut across 321 tech-sector events, or roughly 880 people a day. LinkedIn joined the list this month. Parker filed for bankruptcy on May 9. Each of those data points is also a signal that has reshaped how venture capitalists think about team risk on the first call in 2026. We have sat in on or reviewed transcripts from over 1,200 first calls between Capwave founders and the 89,000+ investors we track, and the team-question patterns from 2024 are no longer the team-question patterns of today.
The shift is not subtle. The old first-call playbook was about credentials: who you hired, where they came from, how big the team is. The new playbook is about structure, single-points-of-failure, and ratio. VCs are asking three specific questions about your team that did not exist in most 2024 first-call frameworks, and the founders who walk in prepared for them close the next meeting at a noticeably higher rate. This post breaks down the three new questions, the answers VCs are actually scoring, and how to retool your team slide before your next VC first call in 2026.
Why VCs changed the first-call team questions in 2026
VCs changed their first-call team questions in 2026 because 137,547 tech layoffs through May 16 and a wave of mid-stage shutdowns turned team structure into a leading indicator of capital efficiency. The old questions about credentials and team size no longer separate the companies that will survive a tight market from the ones that will not.
The shift began in late 2024 when AI infrastructure plays started running with five to ten people while their pre-2023 SaaS comp set ran with thirty to fifty. By 2026, every investor we track has at least one portfolio company that crossed $5 million in ARR with a single-digit headcount, and at least one other that burned the equivalent in payroll and shipped half the product. That delta is not subtle, and it has rewired what VCs scan for on the first call. This is also part of why we updated our framework for founder market fit and the 6 signals investors look for earlier this spring, since the team-fit lens has moved faster than founder-market fit.
Layoffs are the second driver. The Crunchbase 2026 tracker through May 16 shows 137,547 cuts across 321 events, with the pace accelerating into Q2. Each high-profile event, including LinkedIn this month and Parker filing on May 9, sends a signal to investors that even well-funded teams are not safe from over-hiring. The result is a portfolio-construction posture that prizes lean teams, low key-person risk, and option pool structures that can absorb a downsizing without triggering trapped equity.
The third driver is AI workflow maturation. Engineers who shipped one feature per quarter in 2023 now ship one per week with the same headcount. Investors who do not adjust for that benchmark over-estimate fair team size and under-estimate burn discipline. By the first call in 2026, most general partners we hear from have a written rubric for evaluating team-to-output ratio. If a founder cannot answer to that rubric, the call is effectively over within 15 minutes, even if nobody says so out loud.
Question 1: How is your team affected by RIF clauses on the option pool?
A RIF clause, or reduction-in-force clause, governs what happens to unvested options when an employee is terminated, including how those options either return to the pool or accelerate, and whether the option pool itself gets topped up after layoffs. VCs ask about RIF clauses in 2026 because the wrong structure can trap equity, dilute founders during a downsize, or create golden-handcuff problems with senior hires.
The RIF question lands like an HR question and is actually a cap-table question. When a venture-backed company runs a layoff, three things happen to the option pool. Unvested options return, vested options either lapse or get exercised, and the board has to decide whether to top the pool back up. Each of those decisions has dilution implications for everyone on the cap table, and the legal language that governs them was written when layoffs were rare.
In 2026, layoffs are not rare. The Crunchbase tracker logs roughly 880 people a day across the sector. VCs have watched portfolio companies absorb a 15% headcount cut and discover that their option pool topped up automatically by 8%, diluting the founders and the lead investor by the same percentage at exactly the moment they had no leverage. That is the failure mode the first-call question is designed to surface.
The answer VCs are scoring goes something like this. State explicitly whether your option pool plan documents include automatic top-up triggers. If they do, name the trigger and quantify the dilution at, for example, a 10% RIF event. If they do not, explain how your board would handle a top-up decision and what your bias is. Add the second-order point: name how vested option exercise windows are structured for terminated employees, because a 90-day window is industry standard and a 10-year window can quietly inflate the diluted share count over time. Founders who answer with specifics tell investors two things: that they understand cap table mechanics, and that they have already pressure-tested the downside scenario.
The Capwave-tracked data here is striking. Across the 500+ rounds we audited in the last 90 days, founders who could name their RIF clause language on a first call moved to the next meeting 1.7x more often than founders who deferred to their lawyer. The deferral signals that the founder has not actually modeled their own dilution. That is the inference VCs make whether or not it is fair.
Question 2: Where is the single-key-person dependency in your AI stack?
The single-key-person dependency in an AI stack is the engineer, prompt designer, model trainer, or data pipeline owner whose departure would set the product back by 90 days or more. VCs ask this in 2026 because lean teams concentrate technical risk in fewer people, and the wrong concentration kills an investable thesis in a single resignation email.
The 2024 version of this question was about technical co-founder credentials. The 2026 version is about resilience. Investors have seen too many AI companies where one model engineer owns the entire fine-tuning pipeline, one prompt designer owns the entire product surface, or one infra lead owns the entire deployment stack. When that person leaves, the company is either paused for a quarter or pivots from a lead position to a follower position in its category. Either outcome is a write-down on the investor’s spreadsheet.
The good answer names the dependency by role, not by person, and then names the mitigation. For example, the founder might say that the model training pipeline runs through one senior ML engineer today, and the company has a documented playbook, a paired engineer rotating into the role over the next 90 days, and a contractor relationship with a third party that can pick up the work within a week if needed. That answer signals three things: the founder has audited their own bench depth, they have a plan, and they are not in denial about the risk.
The bad answer pretends the dependency does not exist. Some founders try to make the team sound deeper than it is by giving generic titles to junior hires. Investors see through that within two follow-up questions. The better posture is to acknowledge the dependency honestly, name the mitigation, and tie it to a hiring roadmap that the next raise will fund. Our guidance on how much to raise at pre-seed in 2026 walks through how to budget specifically for the bench-depth hire so it lands in the next round, not the one after.
There is a second-order version of this question that some VCs ask in 2026 specifically: where is the data pipeline single-point-of-failure? Many AI companies have rich product engineering and thin data engineering. The team that ingests, cleans, and labels training data is often one person and one consultant. If that pairing breaks, the model degrades within a quarter. Founders who can name that risk and their data backup plan are answering a question some VCs do not even articulate fully on the first call, and the signal of that preparation is what moves the meeting forward.
Question 3: What is your AI headcount multiplier?
The AI headcount multiplier is your revenue, output, or shipped-feature volume divided by your full-time engineering or product headcount, benchmarked against the 2024 pre-AI norm for your category. VCs ask this in 2026 because the multiplier is now the cleanest single-number proxy for capital efficiency in a tight market, and most founders cannot answer it on demand.
This question is the team-side version of the burn-multiple question that became standard at Series A in 2023. The multiplier has migrated forward to first calls because investors want to know how your team-to-output ratio compares to the new baseline, and they want to know within the first 20 minutes. The right answer is a single ratio with the comp set named.
For example, a SaaS founder might say: “We ship one customer-impacting feature every nine days with five engineers. The pre-AI 2024 benchmark for our category was roughly one feature per engineer per quarter. We are running at 4x the pre-AI baseline.” A different founder might say: “We crossed $1.4 million in ARR last month with six full-time employees. That puts us at $233K ARR per FTE, against a SaaS median of around $100K to $120K at our stage.” Either framing answers the question. The stage-specific comp set matters here: our pre-seed vs seed funding guide lays out the 2026 ARR-per-FTE benchmarks at each stage so you can pick the right comp before the call.
The wrong answer is to talk about culture, mission, or how hard the team works. Investors are not scoring those answers in 2026. They are scoring the ratio. Founders who walk in with a number, a comp set, and a sentence about how the AI tooling enabled the ratio are showing investors three things: the discipline of measurement, the awareness of the new benchmark, and the operational maturity to compete on it.
A note on what counts in the headcount denominator. Some founders try to flatter the ratio by excluding contractors, designers, or part-time team members. Investors generally include everyone the company pays, including significant contractors, and they expect founders to count the same way. Be honest about the denominator. The ratio still beats the 2024 comp even with the larger headcount, and you will not be caught out at the diligence call.
How to prep the team slide and Q&A for the new first-call playbook
Updating your team slide for 2026 means adding three specific data points: option pool RIF clause language in one line, named key-person dependencies with mitigations, and your AI headcount multiplier with comp set. The slide does not need to lead with these, but they need to be visible before the founder is asked about them.
Most founders’ team slides in 2026 still look like the 2024 version: photos, titles, two or three credentials per person, and a count of full-time employees. That slide does not answer any of the three new questions. The fix is not to redesign the slide from scratch. The fix is to add a small lower-section or speaker-note block that arms you to answer in 30 seconds when the question lands.
A specific structure that works in our experience with Capwave founders is to keep the existing team slide as-is for the visual, then add three lines below the team grid. Line one names the option pool structure and any RIF clause language. Line two names the top two key-person dependencies and the active mitigations. Line three names the AI headcount multiplier and the comp set. If the VC asks any of the three new questions on the call, you can point to the slide and answer in one breath. If they do not ask, the slide signals that you would have been ready, and that signal alone is worth the prep.
For the spoken answers, rehearse the 30-second version of each. Practice naming the specific number, naming the comp, and stopping. Founders who over-explain in this section dilute the signal of preparation. Founders who answer in three sentences and then sit with the pause come across as operationally calm under scrutiny, which is exactly what the new first-call playbook is rewarding. Putting these answers in front of the right investors at the right cadence is also a pipeline question, not just a deck question, and the founders who match the two beat the founders who only fix one.
Frequently asked questions
What is a RIF clause in a startup option pool plan?
A RIF clause, or reduction-in-force clause, is the section of your option pool plan documents that governs what happens to unvested options, vested but unexercised options, and the size of the option pool itself when employees are terminated through a layoff. Typical RIF clauses cover the exercise window for vested options after termination, whether unvested options return to the pool, and whether the pool gets automatically topped up after a headcount reduction. Mismanaged RIF clauses can dilute founders or trap equity.
How big should an early-stage startup team be in 2026?
There is no single right team size, but the 2026 benchmarks have shifted significantly from 2024. Pre-seed AI infra companies often operate with 3 to 6 full-time team members. Pre-seed consumer or SaaS companies typically run 5 to 9. Seed-stage AI companies frequently reach $1 million to $3 million in ARR with 6 to 12 team members. What investors care about is the ratio of headcount to output, not the absolute number. A 12-person team shipping at the pace of a 4-person team will get the harder questions.
What is the AI headcount multiplier and how do I calculate it?
The AI headcount multiplier is your output volume divided by your full-time team or engineering headcount, compared to the 2024 pre-AI baseline for your category. To calculate it, pick one output metric you can defend, such as features shipped per quarter, customer-impacting deploys per month, or ARR per FTE. Divide by your FTE count, including significant contractors. Then look up the 2024 comp for your stage and category. Most founders find their multiplier is between 2x and 5x the pre-AI norm. If yours is below 1.5x, expect harder team questions.
What should I say if my team has a single-key-person dependency I cannot mitigate yet?
Acknowledge the dependency by role, not by person, and name the timeline for mitigation. For example, you might say that the model training pipeline runs through one senior engineer today, that you have budgeted for a paired hire in the next funding round, and that you have a contractor relationship as a stopgap. Investors do not expect early-stage teams to be deep on every function. They expect founders to know where the risk is, to have a plan, and to not be in denial. Founders who acknowledge and plan beat founders who claim there is no risk.
Do VCs still ask about team credentials and prior exits?
Yes, but credentials and prior exits are now table stakes rather than differentiators on first calls. Investors will still scan your resumes, but the conversation that decides whether you get a second meeting in 2026 is about structure, single-points-of-failure, and ratio. A team with credentialed founders that cannot answer the new questions tends to lose to a team with less credentialed founders who walk in prepared. Use your credentials to open the door, then answer the structure questions to keep it open.
How long should my team slide be in a 2026 pitch deck?
One slide for team in a first-call deck is still the right answer. The change in 2026 is that the slide should carry a lower-section block with three lines covering option pool and RIF language, named key-person dependencies with mitigations, and your AI headcount multiplier with a comp. If you find yourself needing two slides for team, the issue is usually that you are over-explaining credentials. Keep credentials tight and let the structural detail do the heavier lifting in the room.
What if my company is not AI and I do not have a headcount multiplier?
Frame the same ratio without invoking AI. The underlying question is capital efficiency per team member, which applies to every company. A non-AI SaaS founder might frame it as ARR per FTE against a benchmark. A consumer founder might frame it as monthly active users per FTE. A marketplace founder might use GMV per FTE. The number matters more than the framing. Investors are looking for evidence that you measure team productivity and you have a comp set in mind.
How should I prep for the new questions if I have a first call this week?
Block 60 minutes. Pull your option pool plan and read the RIF clauses. Write a one-line summary you can speak from memory. List the three roles on your team that, if vacated tomorrow, would set you back 90 days, and write a one-line mitigation for each. Calculate your AI headcount multiplier against the 2024 comp for your category. Add the three lines to the lower section of your team slide. Practice each answer at 30 seconds. You will be ready by the end of the hour.
The first call is where the round is won or lost in 2026, and the team questions have moved from credentials to structure faster than most founders have updated their decks. The three questions that did not exist in 2024 first-call frameworks, RIF clause exposure on the option pool, single-key-person dependency in the AI stack, and the AI headcount multiplier, are not difficult to answer once you have done the work. They are difficult to answer cold.
The founders who walk in with the numbers ready close their next meeting at a noticeably higher rate. Capwave’s InvestorIQ surfaces the specific check sizes and team-question patterns of the 89,000+ investors we track, so founders can prep for the conversation that is actually about to happen.
