State of M&A Data Rooms — Q1 2026 Read the report →
Peony LogoPeony

How to Share an Interactive Financial Model With an Investor (Securely)

Co-founder at Peony. Former M&A at Nomura, early-stage VC at Backed VC, and growth-equity / secondaries investor at Target Global. I write about investors, fundraising, and deal advisors from the deal-side perspective I spent years in.

I'm Sean Yu, co-founder of Peony. I've sat on both sides of this: the founder hitting send on a financial model at 11pm, and the investor who opens it the next morning. The single most common mistake I see in a raise isn't the model itself — it's the way it gets sent.

You spend three weeks making the numbers defensible, then you attach the .xlsx to an email and lose control of it in one click. The formulas, the hidden tabs, the assumption cells you meant to scrub, the comment from your last raise — all of it now lives on a stranger's laptop, forwardable to anyone, with no way to know who opened it. And the moment that investor edits a single cell, your shared source of truth quietly forks into two.

This post is about the opposite outcome: the model that survives the send. A live, interactive model that stays interactive for the investor — they flex scenarios, toggle assumptions, watch outputs recompute — while you keep per-viewer control: an identity-bound watermark, view analytics, and instant revoke. The file leaves your hands, but not your control.

I run Peony, a data room company, and we serve 5,900+ customers, a lot of them founders and CFOs in exactly this moment. So I have a point of view. I'll also tell you, honestly, where a free public link beats us — because for some sends, it does.

Quick answer: Don't email the workbook. Render the model as an interactive view in a secure viewer and send each investor an identity-bound link. They still flex the model live; you keep a per-viewer watermark, page-level analytics, and instant revoke. The file effectively leaves your hands but not your control.

Emailing an .xlsx versus sharing a live model in a secure viewer — a row-by-row comparison of what you lose with email and what you keep with an identity-bound viewer.

What goes wrong the moment you email an .xlsx to an investor?

You lose the three things that matter most: secrecy, control, and a single source of truth.

An emailed workbook is the most leaky way to share a model that exists. Here's everything that walks out the door with it:

  • Your live formulas and logic. The investor doesn't just see your outputs — they see exactly how you built them. Every formula, every circular reference you fought, every fudge factor.
  • Your hidden tabs and assumption cells. Most models carry a graveyard of hidden sheets: an old scenario, a different cap table, notes to yourself, sometimes another investor's terms. "Hidden" in Excel means "right-click, unhide."
  • Forwardability with no trail. A file has no link to expire and no identity attached. It can be forwarded to a competitor, an associate, or a friend "for a sanity check," and you'll never know.
  • The single source of truth. This is the subtle one. The instant the investor changes a growth rate in their copy, their model and yours diverge. On the follow-up call you're each looking at different numbers and don't realize it.

A dead .xlsx snapshot is also just a worse experience for the investor. Half of them will open it, see a wall of tabs, and bounce. The other half will start "improving" it — and now you're defending a version of your own model you never built.

If you instead want to know how to build the model itself — the tabs, the waterfall math, the Excel mistakes that get a model rejected — that's a different post. See Independent Sponsor Financial Model. This post is strictly about sharing one you've already built.

What does it actually mean to keep a model "interactive" when you share it?

It means the recipient can change inputs and watch the outputs recompute — live, in their browser — without you handing over the underlying file.

This is the heart of "the model that survives the send." A great model isn't a number; it's a machine. The investor's whole job is to turn the dials: bump CAC, cut the growth rate, push the exit out two years, and see whether the thing still holds. A PDF can't do that. A screenshot can't do that. A flattened export of your model is just a photograph of a machine that no longer runs.

In 2026, more and more of these models are built or rendered as interactive HTML — whether you exported a scenario tool from your spreadsheet, or you built the whole thing in Claude or GPT as a live calculator or dashboard. That's a genuinely better artifact to put in front of an investor: it invites them to interrogate the model instead of just reading it.

The problem is sharing it. Here's the trap most people hit: most secure-sharing tools either reject .html outright or convert it to a PDF on upload. The app dies. The toggles stop toggling. The investor opens your live model and finds a static image of it.

This is where I'll be specific about what Peony does, because it's the core reason this post exists. Peony renders .html/.htm natively in the in-browser viewer with JavaScript executing — so a live, AI-built or spreadsheet-derived interactive model actually runs for the recipient. They change an assumption; the outputs recompute, right there in a viewer you control. Most tools can't do this. We can. (The product detail lives on the HTML viewer page if you want the mechanics.)

How do you share a financial model with an investor while keeping control?

You stop sharing a file and start sharing an identity-bound view. The view carries your controls with it.

Here's the practical sequence I'd run if I were sending one model to investors tomorrow:

  1. Render the model as an interactive view, not a workbook. Either export your scenario layer to HTML, or, if you built it in an AI tool, share the live artifact directly. The goal is that the investor can flex it without ever holding the .xlsx.
  2. Issue each investor their own link, tied to their name and email. Not one link for everyone — one per person. This is the single decision that makes everything downstream possible: attribution, analytics, and selective revoke all depend on it.
  3. Turn on the controls that match the sensitivity of the model. A per-viewer dynamic watermark burns each investor's name, email, and a timestamp over the live render. Download prevention and screenshot protection raise the cost of exfiltration. NDA gating puts a click-through wall in front of access. Granular per-file and per-link permissions let you show one investor the full model and another a trimmed version.
  4. Watch the analytics, then follow up precisely. You'll see who opened the model, which scenarios they stressed, and for how long. A cold "just following up" becomes "I saw you spent time on the downside case — want to walk the churn assumption?"
  5. Revoke the instant anything changes. Round oversubscribed, a term sheet signed, an investor passes — kill their link without touching anyone else's.

That's the whole argument in five steps: the file leaves your hands but not your control. A public link can't do step 2 onward, which is why it can never run a real diligence process.

If you're not sending a single model but standing up the whole fundraising room — cap table, incorporation docs, financials, customer contracts, the full diligence inventory — that's a bigger job. See Data Room for Investors and, for a priced round specifically, Series A Data Room. This post is the single act of sharing one live model; those are the room.

It depends on how confidential the model is and whether you need to know who looked at it. Let me be genuinely fair here, because the honest answer makes the choice clearer.

ShareDuo and Stacktree are real, good tools, and for some sends they win. Both let you publish an AI-built artifact to a fast, public-or-unguessable link — free or cheap, no login for the viewer — with a password and an expiry date. Stacktree even has an MCP publish flow with update-in-place, so you can revise the model and the same link reflects it. If your goal is "get a live, interactive model in front of someone in 30 seconds with zero friction," they are excellent, and I'd genuinely use them for a non-confidential teaser.

What a public or unguessable link fundamentally cannot do, no matter how good the tool:

  • Bind the view to a named investor. A password is shared the moment it's useful; it identifies no one.
  • Stamp a per-viewer watermark. Everyone sees the same anonymous render, so a leak has no signature.
  • Gate on an NDA before the model loads.
  • Show you analytics on who opened it and what they looked at.
  • Revoke one person after the fact without breaking the link for everyone.

That's the whole distinction: a public link is sharing a file; an identity-bound view is running a diligence process. Google Sheets "anyone with the link" and a raw DocSend send sit on the file side of that line too — convenient, but anonymous.

Peony's edge is deliberately upmarket and deal-grade. We're not trying to win the free-public-link race; we lose that one on purpose. We win when the model is confidential, the investors are named, and you need to be able to prove who saw what and shut it off. That's the trade, stated plainly.

Where does Peony lose — and where does it win?

Peony loses on free, frictionless, public sharing, and it's not the tool for building the model. It wins on identity-bound, deal-grade control over a live model.

I think conceding this clearly is more useful than a feature dump, so:

Where Peony is the wrong tool:

  • You want a free, no-login, public link in 30 seconds. Use ShareDuo or Stacktree. We require a more deliberate, identity-bound setup by design.
  • You haven't built the model yet. Peony shares models; it doesn't build them. For the architecture and formulas, start with the financial model build guide.
  • You want clause-level citations, a public REST API and CLI, or to self-host the whole thing on your own servers. Those are explicitly not us — I'd rather tell you than have you discover it later.

Where Peony is the right tool:

  • The model is confidential and the investors are named people you can attribute a view to.
  • You need the model to stay live — JavaScript executing in the viewer — not flattened to a dead PDF.
  • You need a per-viewer watermark, page-level analytics, NDA gating, and instant revoke in one place.
  • You'd like to build the model in Claude and push it straight into a watermarked, revocable room via the MCP server without re-uploading.

With 5,900+ customers, a large share of what we see in fundraising is exactly this single, high-stakes send. It's a small action with an outsized downside if you get it wrong.

What does it cost, and is the live-render feature locked behind Enterprise?

It's affordable for a single raise, and no — keeping your model interactive is not an enterprise-only feature. That's the question I get most, so I'll answer it directly.

Native HTML rendering — the thing that keeps your model actually running for the investor — works across tiers. You do not need an enterprise contract to keep your model interactive. Here's the canon:

PlanPriceWhat you get for sharing a model
Free$0Get started; native HTML rendering included
Business$30/admin/monthScreenshot protection, download prevention, NDA gating, page-level analytics
Data Room$52/admin/monthAdds per-viewer dynamic watermark and mobile screenshot blocking (Screenshield)
Deal TeamHigherFor larger deal teams
EnterpriseCustomConnect your own GPT + Claude + Gemini under zero-retention, permission-scoped, audited access

For most founders sharing a model in a raise, the Data Room plan at $52/admin/month is the landing spot, because the per-viewer watermark is the control that makes a confidential model safe to send to multiple investors at once. The full breakdown is on pricing.

One more capability worth knowing about, because it changes the workflow: the Peony MCP server is publicly available now, and it's bidirectional — an AI agent can read room contents and push an artifact into the room. So you can build the interactive model in Claude, install the MCP, and push it directly into a watermarked, NDA-gated, revocable room without leaving the chat. And at the Enterprise tier, you can connect your own GPT, Claude, and Gemini to the room under zero-retention and permission-scoped access — the model sees only what the viewer is allowed to see — with every AI query and document access logged.

The bottom line: send the model, not the file

The mistake isn't your model — it's the .xlsx attachment. Emailing the workbook leaks your formulas, exposes your hidden tabs, hands over a forwardable file, and forks your source of truth the instant someone edits a cell. None of that has to happen.

Share the model as a live, interactive view bound to a named investor instead. They get to flex it; you keep a per-viewer watermark, page-level analytics, and instant revoke. If the model is confidential and the investors are named, that identity-bound control is the difference between sharing a file and running a process. If it's a casual, non-confidential share, a public link from ShareDuo or Stacktree is genuinely fine — I'd use one myself.

For the secure, deal-grade version — the model that survives the send — that's what we built Peony for. 5,900+ customers, and counting.


Disclaimer: This article is for general informational purposes and is not legal, financial, or investment advice. Plan names, prices, and features reflect Peony's offering as of June 2026 and may change — confirm current details on the pricing page. Always consult qualified counsel before sharing confidential financial materials in a fundraising or M&A process.

Sources