Skip to content
Meridian

Investment diligence for angels and small funds

First-pass diligence,
in an hour.

Drop in a startup’s pitch deck and data room. Meridian reads them and produces a structured investment memo: the business, the market, the numbers, the risks, and the sharp questions to ask the founder.

Two memos a month, free. No card required.

fernpath-seed-deck.pdf

reading

FERNPATH

AI copilot for independent insurance agencies

TRACTION

Launched October 2024. As of June 2025 we are at

$412,000 ARR across 48 agencies, up from $90,000 in

January. Net revenue retention is 104%. Logo churn

is 3.1% monthly. Our largest agency, Brookline

Insurance Group, is 22% of ARR.

BUSINESS MODEL

$690 per seat per month, four seats average. Gross

margin 61% after LLM and integration costs. Blended

CAC $6,400, fourteen month payback.

The problem

The weekend data-room slog

A small investor sees ten or twenty decks in a busy month. Reading each one carefully enough to form a view takes most of a weekend. The ones worth a second look are buried in the stack.

A weekend
what a careful first-pass read of a few decks still takes by hand
10 to 40
decks an active syndicate lead can see in a single busy month
1 in 20
early checks that return the fund, so the first-pass filter has to be fast

Illustrative figures used to describe the problem Meridian is built for, not results from a survey we ran. The ranges reflect what active angels and syndicate leads report about their own deal flow.

How it works

Three steps, about a minute

01

Drop the deck and data room

Upload the pitch deck as a PDF or paste the text. Add any data-room documents you have: financials, cap table, customer data. A ten to fifteen slide deck with a handful of exhibits is the typical input. Meridian works out what it is reading without a template to pick.

PDFs where the text is real text, which covers almost every export from Keynote, PowerPoint or Google Slides. Image-only files come back with a message that explains why, rather than a guess.

Uploading files

  • fernpath-seed-deck.pdf
    2.4 MBready
  • q1-2025-financials.xlsx
    380 KBready
  • cap-table.pdf
    120 KBreading

Add data-room documents or paste text

02

Meridian reads and structures the memo

Every section of the deck is read against the whole document, because a revenue figure means something different next to a burn rate and a churn number. Each evidence quote is verified to appear in your document before you see it. A number the model produced that is not in your deck is dropped, not shown.

Long or self-contradicting documents escalate to a frontier model. If anything was dropped for failing verification, the memo says how many and why.

Building memo

  • Business summarydone
  • Market and comparablesdone
  • The numbers that matterdone
  • Risks and red flags...
  • Founder questions...
03

Review the risks and questions

The memo comes back with the business in plain terms, the market and comparables, the numbers that matter, the risks ranked by severity, and the sharp questions to ask the founder. Each risk links to the exact line in the deck that flagged it. Copy the question list and send it.

Most investors read the risk section first, then the numbers, then the summary. The memo is built in that order. Export to PDF or Word on Angel and Fund plans.

Fernpath seed memo

2 high risks

  • highOne customer is 22% of revenue
  • highMonthly logo churn at 3.1%
  • mediumThin gross margin for software

See a memo

A real deck, read in an hour

Fernpath is a seed-stage insurtech. Click any risk in the Risks tab to see its exact evidence highlighted in the source deck. Nothing here is a mockup: this is the product running on a real sample.

FernpathSeed, raising $2.5M
2 high risks

Brookline Insurance Group is nearly a quarter of ARR. If it leaves at renewal, Fernpath drops from $412,000 to about $321,000 and the growth story resets. At seed this is survivable, but it means the revenue is less diversified than the topline suggests, and the round should be sized to survive losing it.

In the deck

Our largest agency, Brookline Insurance Group, is 22% of ARR.

What you get

Six things, all of them the memo

The business in plain terms

What the company does, who it sells to, how the business model works, and what the founder claims the edge is. One paragraph, no buzzwords lifted from the deck.

Sample summary

Fernpath sells software that automates the paperwork independent insurance agencies run on: it reads carrier emails and policy PDFs and drafts the quotes, renewals and compliance documents. It charges per seat, about $33,000 a year per agency, and is early but growing fast off a small base.

Market and comparables

Where the company sits in the market, who the real competitors are, and whether the TAM claim was built top-down or from the ground up.

The numbers that matter

ARR, growth, retention, margin, CAC and the round, pulled verbatim from the deck. A number that is missing becomes a founder question.

Logo churn
3.1% / mo
Gross margin
61%
ARR
$412,000

Risks and red flags

Ranked high, medium, low, with the exact line from the deck that flagged it and why it matters for this stage and sector.

Sharp founder questions

The questions the memo ends with, each with a one-line reason it changes your view. Ranked so the sharpest one is first.

Works on any stage or sector

Pre-seed consumer, seed SaaS, Series A deeptech. The framework applies the right lens: what healthy retention looks like at seed is not what it looks like at Series B.

Model-agnostic routing

Each step routes to the best model for it. The structured memo and the risk pass go to a balanced model. Long data rooms or self-contradicting financials escalate to a frontier model. A lab changing its price or policy is a config change here, not a rebuild.

See the routing table

Built AI-native

The model is the engine, not a feature we added

Meridian was not a spreadsheet that later added AI. There is no version of this product without modern language models, because the work is reading. No regular expression finds concentration risk phrased as “our largest agency is 22% of ARR”, and no template notices that a top-down TAM and a high monthly churn rate together make the growth chart harder to trust.

Meridian is model-agnostic by design. Every step asks for a task, not a model, and one file decides which lab answers. A lab changing its price, its policy or its quality is a config change here, not a rebuild. Long or self-contradicting data rooms escalate to a frontier model automatically.

Routing table, generated from the code

  • Read the deck and data room and structure the memogpt-4.1-mini/balanced
  • Surface the risks and red flagsgpt-4.1-mini/balanced
  • Write the plain-terms business summarygpt-4.1-mini/balanced
  • Work out the sector and stagegpt-4.1-nano/fast

Candidates from other labs are declared behind the same interface and switch on when their key is present. The escalation rule and the fallback chain are on the How the AI works page.

Our stance

Analysis, not investment advice

Meridian is analysis and organization support, not investment advice or a recommendation to invest. Every memo is a draft to inform a human decision. Meridian never says buy or pass, never assigns a score, and never recommends a decision. A memo tells you the shape of the deal, the risks to investigate, and the questions to bring to the founder. The judgement stays yours. For anything with real money on it, a memo is the start of diligence, not the end.

Investors

What it catches

“I got a deck for a seed SaaS on a Tuesday afternoon. The memo came back in about fifty seconds: ARR of $340,000, but logo churn at 4.2% a month, which annualizes to nearly half the customer base. The founder had not flagged that. I asked about it on the call and we had a much better conversation than I would have had cold.”
James ParkAngel investor. 40 checks across 6 years, 3 funds.
“We run a rolling deal pipeline with four partners, and we were losing a weekend a month just building the same memo structure for every deck. Meridian does the first pass and we spend the time arguing about the numbers rather than finding them. The founder questions at the end of every memo are the part I steal for my own calls.”
Sandra OduyaSyndicate lead. Series A syndicate, 200 LPs, Chicago.

Illustrative investor stories, written to show what Meridian catches. They are composites, not quotes from named customers.

Pricing

Two memos a month, free

Prices are in US dollars and that is the only price list there is. No setup fee, no minimum term, cancel from the dashboard.

Free

Free

Two memos a month, done properly. Enough to run a real first pass on the deals actually in front of you this month.

  • 2 investment memos a month
  • The business in plain terms
  • The numbers that matter, pulled from the deck
  • Risks and red flags, ranked by severity
  • Sharp questions to ask the founder
  • Upload a deck or paste the text

Most popular

Angel

$39/ mo

or $33/mo billed annually

For the angel who sees real deal flow. Unlimited memos, export to send to a syndicate, and a searchable history of everything you have looked at.

  • Everything in Free
  • Unlimited memos
  • Full risk analysis and founder question sets
  • Export a memo to PDF or Word to share with a syndicate
  • Searchable history of every deal you have run
  • Priority model routing on long data rooms
  • Email support

Fund

$149/ mo

or $124/mo billed annually

For the micro-fund where more than one person reads the deck. Five seats, shared deal notes, and answers the same day.

  • Everything in Angel
  • 5 seats
  • A shared deal room the whole partnership can read
  • Notes and a decision on each memo, kept with the deal
  • A house diligence checklist applied to every memo
  • Priority support with a same-day first reply

Questions

The ones people ask first

If yours is not here, send it to us and a person will answer.

Run your next deal before the weekend

Two memos a month, free, with the full risk analysis and founder questions included. No card required.