Your next investor will ask about gross margin. Better to know the answer first.

In a growth round the AI capability is often the entire case for the multiple. In a large share of companies described as AI-powered, that capability is a prompt against a commercial model — replicable by any competitor in a week. The company can still be a good investment, but the premium belongs on the distribution and the workflow, not on the API call.

The second problem compounds with growth. Inference cost scales with usage while most pricing scales with seats, so a company can look healthy at current volume and be structurally unprofitable at three times the volume. Meanwhile venture funding masks infrastructure waste: with a recent round in the bank, nobody scrutinizes a cloud bill that is growing alongside everything else.

The work is the same as the private equity practice: quantify it, attach a confidence level, and translate it into the measure the reader is held to. For a venture-backed company that measure is months of runway and points of gross margin rather than EBITDA. And because many venture-backed companies are eventually sold to a private equity buyer, it is also a preview of what that buyer's diligence team will look for.

Four Ways In

One is the investor's question — technical and AI due diligence before the term sheet. The other three are the startup's, asked before the investor gets to them. Each starts with a fixed-scope, fixed-fee assessment.

Growth-Round AI Diligence

2-3 weeks

For the lead investor in a Series B or later round, where the AI thesis is carrying the valuation.

At growth stage the AI capability is often the whole case for the multiple, which makes the question of whether it is a capability or a wrapper sharper than it is in a buyout. This is technical due diligence centered on the AI: it runs the five diligence tests, models gross margin at plan volume rather than current volume, and maps the dependency on model vendors — written for the partnership, not the engineering team.

What you get

  • Capability versus vendor passthrough, stated plainly with the evidence behind it
  • Gross margin modeled at the volume the round is underwriting, not at today’s
  • Model and vendor dependency, with switching cost estimated
  • Data rights and provenance, including what survives an acquisition
  • A written view on whether the revenue multiple’s AI premium is supportable
Book a Discovery Call

Pre-Raise Readiness Review

2 weeks

For founders six to twelve months from a raise: the investor’s technical diligence, run on your side of the table first.

The sell-side counterpart to growth-round diligence. A lead investor will ask whether the AI is defensible, why gross margin is where it is, and what happens to it at three times the volume. This review finds the answers that will not survive that conversation while there is still time to change them, and writes up the ones that will.

What you get

  • The five AI diligence tests, applied as an investor would apply them
  • Gross margin trend and its drivers, explained in the terms a partner uses
  • Cloud and inference spend against benchmarks, with waste quantified by confidence
  • A prioritized fix list sequenced against the raise timeline
  • Written answers to the findings a diligence team will raise
Book a Discovery Call

Runway Extension Review

2 weeks

Cloud and inference spend reviewed for what it is worth in months of runway and points of gross margin.

Venture funding masks infrastructure waste: with a recent round in the bank, nobody is scrutinizing the cloud bill while it grows alongside everything else. This review quantifies the waste with confidence levels attached and expresses every finding as runway and margin, so the board can weigh it against hiring and roadmap rather than against a dashboard.

What you get

  • Waste quantified by category, with dollar figures and confidence levels
  • Savings expressed as months of runway at current burn
  • Gross margin impact, separated from operating expense
  • A prioritized plan with cost, timeline, and confidence for each item
Book a Discovery Call

Fractional CTO for Founders

1-2 days/week

Senior technology leadership after a raise, without spending the option pool or the runway on a full-time executive.

The company has outgrown the architecture and the team that got it here, and the next eighteen months decide whether the round is spent on growth or on rework. A fractional engagement owns the roadmap, the hiring plan, the build-vs-buy and model decisions, and the infrastructure spend — and reports on them to the board in financial terms.

What you get

  • Technology roadmap ownership tied to the board plan
  • Engineering hiring plan and org design for the next stage
  • Build-vs-buy and model selection decisions with cost modeling
  • Cloud and inference spend governance tied to unit economics
  • Board and investor reporting in gross margin and runway terms
Explore Fractional CTO

Same Method, Different Measure

Every finding is quantified the same way. What changes is the number it is expressed in.

PE-backedVenture-backed
What a saving is worthEBITDA, at the exit multipleMonths of runway, and points of gross margin
Where the AI premium shows upTurns of EBITDAThe revenue multiple on the round
The moment it mattersPre-LOI, diligence, and 12-24 months before exitThe next raise, and the eventual sale
Who reads the reportThe investment committee and the operating partnerThe partnership, the board, and the founder
Who usually engagesThe fund, or the portfolio company at the sponsor’s directionThe lead investor for diligence; the company for everything else

To see the venture-backed view on your own numbers, both the cloud savings calculator and the AI cost & margin calculator have a venture-backed mode that reports runway and gross margin instead of EBITDA.

Who This Is For

Growth-Stage Investors

You are leading a round in a company whose deck says AI on every page, and the comparable rounds support a higher multiple. You need to know before the term sheet whether the capability is proprietary or a vendor passthrough, and whether gross margin holds at the volume you are underwriting.

VC Platform Teams

Your portfolio founders keep asking the same questions about cloud bills, inference cost, and when to hire a CTO. You need someone to send them to who will scope honestly, charge a fixed fee, and say so when the answer is that they do not need help yet.

Founders and CEOs

You have a raise coming, an inference bill growing faster than revenue, or an architecture that will not carry the next stage. You need a peer who can tell you what your next investor will find, and what it costs to fix before they do.

This is a fit when

  • Post-Series A, with a product in market and revenue to model against
  • Cloud and inference spend is a meaningful share of burn or of cost of revenue
  • The AI capability is central to the valuation, not a feature on the roadmap
  • A raise, an acquisition, or a board decision is coming that will turn on these numbers

This is not a fit when

  • Pre-product-market-fit, where the architecture will be rebuilt before anyone looks at it
  • Seed-stage diligence at a check size where a two-week assessment is out of proportion
  • What you need is an implementation team to build the AI. That is a hiring or vendor decision
  • You want a number to put in a deck rather than a finding you can defend

If the discovery call establishes that this is the wrong engagement, I will say so and point you at the right one — including when the right one is nothing at all.

Being Acquired by a PE Firm?

When a venture-backed company is sold to a private equity buyer, the buyer's diligence team reads the cloud and AI line the way Cloud Economics for Private Equity describes. The pre-exit cloud cost audit finds what they will find, early enough to fix it rather than have it priced in.

How an Engagement Works

01

Discovery call

Thirty minutes to understand the round, the timeline, and whether this is a diligence question, a readiness question, or an ownership question. If it is none of them, I will say so.

02

Assessment

A fixed-scope, fixed-fee assessment, usually two weeks. It is a complete piece of work on its own — plenty of companies take it, act on it themselves, and need nothing further.

03

Engagement

Where ownership is the right answer, the agreed cadence begins against the board plan, with standing time with the CEO and engineering leadership and a regular reporting rhythm with the board.

04

Handover

Every engagement is built to end. Documented architecture, model dependencies, vendor relationships, and roadmap — handed to the permanent leader with onboarding support.

Common Questions

What does technical due diligence for a Series B round cover?

At Series B and later, technical due diligence is less about whether the product works and more about whether the economics hold at the scale the round is paying for. For an AI-positioned company that means five things: whether the AI capability is proprietary or a vendor passthrough, what the company owns that the model does not, its dependency on model vendors and what switching would cost, whether output quality is measured or asserted, and gross margin modeled at the volume the round underwrites rather than today’s. Alongside those sit the conventional questions — architecture, the engineering organization, delivery record, security, and cloud spend against benchmarks. The written result is a view on whether the AI premium in the valuation is supportable, in the terms the partnership uses.

Why runway and gross margin instead of EBITDA?

Because a venture-backed company is usually EBITDA-negative by design, so a finding expressed as EBITDA impact at an exit multiple describes a business it is not. The method is the same one used for private equity — quantify the waste, attach a confidence level, translate it into the measure the reader is held to. For a venture-backed company those measures are months of runway at current burn and points of gross margin, because gross margin is what a revenue multiple is quietly adjusted for.

Who pays — the fund or the company?

It depends on whose question it is. Growth-round diligence answers the investor’s question, so the fund engages it. Readiness reviews, runway reviews, and fractional CTO work answer the company’s question, so the company engages them — often after an introduction from the investor’s platform team. Either way the fee is fixed and agreed before work starts.

Can you run diligence for an investor on a company where you are the CTO?

No. The same analysis runs on either side of the table, which is exactly why it cannot run on both sides of the same one. I do not run diligence on a company for an investor while I hold its CTO seat, and I do not work both sides of the same round or the same sale.

Your book is about private equity. Why does it apply here?

Two reasons. The method transfers directly: waste quantified with confidence levels, and findings translated into the financial measure the reader is accountable for. And many venture-backed companies end up sold to a private equity buyer, whose diligence team will read the cloud and AI line exactly the way the book describes. Knowing what that buyer looks for is useful years before the sale.

Do you work with seed-stage companies?

Rarely for diligence or reviews, because at that stage the spend is usually too small for the findings to pay for the work, and the architecture is likely to change anyway. Fractional CTO work is the exception, when a seed-stage company has an architecture or hiring decision coming that will be expensive to get wrong. The discovery call is the place to establish which case you are in.

Do I have to commit to a full engagement up front?

No. Every engagement on this page begins with, or is, a fixed-scope, fixed-fee assessment that produces a written report and a prioritized plan. It is a complete piece of work on its own, there is no obligation to continue, and if the assessment concludes nothing further is warranted, it will say so.

Raising, leading a round, or running out of runway?

Book a free 15-minute discovery call. We'll discuss the situation and I'll tell you honestly whether any of this is worth doing yet.

Schedule a Call