If pillar #8 was about what HELM stores, this post is about what HELM produces. Specifically the artifact most operators we talk to said they wanted before we'd built anything else: a banker-grade quarterly PDF they could hand to their CPA, their CFP, their estate attorney, or — in the bleakest version of the use case — a surviving spouse with no spreadsheet skills who needs to know what the household actually owns.
This is the document side of HELM's whole thesis. The product database, which we publish in full, is upstream. The quarterly PDF is downstream — it's the artifact that gets emailed, printed, taped to a refrigerator, scanned by a new advisor, and dropped into an estate folder. It has to read like something a private bank would generate, because the operator is going to forward it to professionals who deal with private-bank documents every day. If our PDF looks like a SaaS app dashboard exported as a PDF, it gets ignored. If it looks like the document the CPA was already asking for, it gets opened.
The rest of this post is the spec — page by page — for what's in HELM's quarterly review document, what's deliberately not in it, and the design reasoning behind both lists.
The use case the document is shaped around
The PDF is generated quarterly (or on-demand) and is meant to be reviewed in three minutes by someone who is not the operator. Three minutes is the realistic attention budget for a CPA running 40 client reviews in a week, a CFP coming back from a vacation, or an estate attorney prepping for a meeting. Anything that takes longer than three minutes to extract the headline numbers from gets put in the "I'll come back to this" pile and never gets opened again.
The whole document is built around two design constraints derived from that use case:
- Top-of-page typography is non-negotiable. The first thing the reader sees is the name on the cover, the date of the snapshot, and the headline net worth. Those three numbers settle whose document it is, when it was generated, and whether anything material has changed. Everything else is secondary.
- Every page must work standalone. Pages get printed individually, screenshotted, forwarded as attachments. We don't assume the reader is paginating through. Each page restates the operator name, snapshot date, and a two-line "context" caption so any one page can stand alone in an email thread.
This is the part of the design that took the longest. Most "export to PDF" features in finance apps produce a multi-page screenshot of the dashboard. That's not a usable artifact for any of the readers in the real use case. The HELM quarterly is built for the readers, not for the operator's love of nice charts.
Page by page
Identity, date, headline.
Operator name (the display name they set in HELM — never a legal name unless they explicitly typed one). Snapshot date in long form. The headline net worth at the snapshot date and the change since the previous quarter, in dollars and as a percentage. A small "prepared by HELM · Vantage Digital studio" mark in the footer for provenance. No marketing copy. No company logo crowding the cover.
Decisions encoded here: the cover is what survives the social-share / forwarding gauntlet. If it looks like a brokerage statement at a glance, it gets opened. If it looks like a marketing PDF, it gets archived unread.
Where the change came from.
A waterfall chart breaking down the prior-quarter-to-current-quarter delta into four buckets: contributions (cash in), distributions (cash out), market movement (price change on holdings you didn't transact), and category shift (assets that moved between buckets — e.g., 401(k) rolled to IRA). Every bucket reconciles to the headline change. No black-box "other."
Decisions encoded here: the waterfall is the page the CPA actually looks at. It answers "did the client save more this quarter, did the market do the work, or is something off." That single page tells the story 80% of quarterly reviews are about.
By asset class. By account type. By location.
Three pie charts side by side, one row each: by asset class (US equity / international equity / fixed income / cash equivalents / real assets / private), by account type (taxable brokerage / Roth / traditional IRA / 401(k) / HSA / other tax-advantaged / cash), and by tax-location category (taxable / tax-deferred / tax-exempt). Numbers in dollars and percentages.
Decisions encoded here: the asset-location chart is the page that catches "you have $200K of high-yield bonds in your taxable account and your Roth is full of S&P index" — a problem we wrote a whole pillar about. Many operators don't realize they have a location problem until they see the three-pie comparison.
Every position, in one table.
One row per holding: account label, ticker or asset name, asset class, quantity (where applicable), current value, percentage of total net worth, and (for taxable lots) cost basis and unrealized gain/loss. Sorted by value, descending. Page breaks across the table cleanly so the rows aren't orphaned.
Decisions encoded here: this is the page the CPA cross-references against the prior-year tax return. It's also the page the CFP uses to identify rebalancing targets. It looks boring. Boring is correct here.
Wash-sale flags · harvest opportunities · QSBS clock · RMD calendar.
This is the page that's distinctive to HELM and the page most operators forward to their CPA first. Four sections:
- Wash-sale flags. Any sales in the last 30 days that triggered or are at risk of triggering wash-sale disallowance under IRC §1091, including across accounts. Empty in most quarters, which itself is informative — confirms no surprise disallowances.
- Harvest opportunities. Positions currently at a loss with their unrealized loss amount and the 30-day-prior-and-after substantially-identical lookout list pre-computed. Saves the CPA 20 minutes of manual hunting.
- QSBS clock. Any positions you've flagged as §1202 eligible, with the days-to-five-year-mark counter. Our pillar #3 is on this; this section is its dashboard form.
- RMD calendar. If you have an inherited IRA or are post-RMD age, the schedule for the year and your year-to-date status. Year-end-tax-loss-harvesting deadlines noted.
What changed in the projection since last quarter.
If you've been running scenarios in HELM, this page shows the delta on your primary scenario since the previous quarter — change in starting value, change in projected final value at your target year, change in success probability if you have a Monte Carlo run attached. If you haven't been running scenarios, this page is omitted (we don't pad).
Decisions encoded here: this page exists for operators in equity-comp situations — vesting schedules, ISO exercise windows, secondary tender offers — where the "what would happen if I exercised in October vs January" question is the actual decision driving the quarterly review.
Five numbers that predict the next 12 months.
The leading-indicator dashboard from pillar #7: savings rate × 12, marginal-dollar deployment, cash cushion in months of expenses, debt-paydown velocity, and vesting-pipeline NPV. Each shown with the prior-quarter comparison and a small green/amber/red status pill.
Decisions encoded here: this is the page that tells the operator (or the CFP reviewing on their behalf) whether the curve is accelerating or decelerating. Net worth is lagging; these five are leading. The page exists so the quarterly conversation isn't only about what already happened.
What's filed, what's expiring, what's missing.
If you use the document vault, this page lists every file you've uploaded by category (estate planning, tax letters, insurance riders, account-opening confirmations, beneficiary forms), with the upload date and any "renewal due" flags. Items older than 12 months in the "estate planning" or "insurance" categories get a soft amber flag suggesting a review.
Decisions encoded here: this is the page the surviving-spouse use case pivots on. If something happens to the operator, this page is a literal table of contents pointing the family to "the estate plan is in HELM, dated 2024, the will is in HELM, dated 2023, the life insurance policy is in HELM, dated 2022 and overdue for review." It's the document version of "where the keys are."
The receipt for every number on every page.
Plain-language methodology: how each page's numbers are computed, which inputs they're derived from, whether market values are operator-entered or pulled from public-API price data, what's deterministic and what's probabilistic. Plus the standing disclaimer that HELM is software, not a registered investment advisor, not a CPA, not a tax attorney, and that all figures are for informational and planning purposes only.
Decisions encoded here: when a CPA or attorney reviews someone else's analytical document, the first question is "where did these numbers come from." Page 9 answers that question explicitly so the rest of the document gets the benefit of the doubt. The disclaimer also exists for legal-compliance reasons; we are not your lawyers and we won't pretend otherwise.
What we deliberately leave off
The PDF is shaped as much by what's missing as by what's there. The omissions are intentional — every one of them was a candidate that we made the call to leave out:
- No account numbers, ever. The PDF references "Vanguard taxable" or whatever account label the operator typed in, never the actual account number. If the document is forwarded, mislaid, or photographed, no actionable account identifier leaks. (And: HELM doesn't store account numbers in the first place — see pillar #8.)
- No legal name unless explicitly typed. The display name field defaults to whatever the operator entered; many use a first-name-only or initials-only label. The PDF respects whatever was entered. We never auto-fill a "real name" from anywhere because we don't have it.
- No marketing copy. No "powered by HELM" footers on every page, no upsell to a higher tier, no "share this report" CTAs. The provenance line on the cover is the only branding. This is not the place to grow our marketing surface.
- No vague "estimated future returns" without methodology. If a number depends on an assumed return rate, the rate is shown explicitly on that page, not buried in a footnote. We refuse to print "you'll have $4M at retirement" as a single number — it always shows the assumption stack underneath.
- No prediction or recommendation. The PDF reports state and shows derived calculations. It does not say "you should sell" or "you should rebalance" or "you should switch to a different IRA." The operator decides; the PDF is a substrate for that decision, not a substitute for it.
- No confidence intervals where the math doesn't support them. The Monte Carlo page shows P10/P50/P90 because that's an explicit probability surface. The waterfall page does not have confidence intervals on contributions and distributions because those are observed values, not estimates. False precision is worse than no precision.
Why a PDF in 2026?
Reasonable question. PDFs feel anachronistic next to live dashboards and link-shareable views. The reasons we ship a PDF anyway:
First, the PDF is the lowest-common-denominator format for the readers in the use case. Every CPA, CFP, estate attorney, and surviving family member can open a PDF without an account, without an internet connection, without a subscription. Web links require trust and access; PDFs require neither. The forwarding cost approaches zero.
Second, the PDF is an artifact, not a state. A live dashboard always reflects "right now," which is the wrong reference frame for a quarterly review — the whole point is to mark a specific snapshot date and reason about what changed between then and now. A PDF is a snapshot by construction.
Third, the PDF is exportable to physical paper. Several operators we've talked to print the quarterly PDF and physically file it. We'd never have predicted that workflow when we started — but for some operators (especially older operators, or operators with surviving-spouse use cases at top of mind) the act of printing and filing is part of how the discipline holds. We support what works.
What "banker-grade" actually means here
Plenty of fintech companies use "banker-grade" as a marketing word. We use it as a constraint on the design choices. Specifically:
- Source Serif Pro for titles and table headers — same family register as a Goldman Sachs quarterly or a Morgan Stanley annual. Sans-serif feels SaaS-y; serif feels documentary.
- Inter for body copy at slightly larger leading than is typical for screen reading. Reads well in print and pleasantly on screen at 100% zoom.
- JetBrains Mono for any tabular numeric column. Tabular figures align across rows. Aligned columns are the smallest possible signal that says "this was made by someone who cares about how numbers look on paper."
- One typeface family per kind of content. Display sans-serif headlines for marketing pages, serif everywhere money is being discussed. This is the same instinct that makes the Wall Street Journal's print edition feel different from a tech blog.
- Absurd attention to vertical rhythm. The leading on every page resolves to the same baseline grid. The reader doesn't notice; the document just feels less janky than the average export-to-PDF.
None of this is what makes the document useful — the underlying numbers do that. But the typography is what makes the document get opened. And an unopened document is worth zero, no matter how good the numbers are.
How the PDF gets used in practice
Three patterns we've watched emerge across operators in the beta cohort:
The CPA prep pattern. The operator generates the PDF the night before their quarterly meeting, attaches it to a calendar invite, and shows up to the meeting with the CPA having already pre-read it. The first 15 minutes that used to be "let me pull up your accounts and look at things" become "I read the wash-sale flag on page 5 and want to talk about it." The meeting becomes substantive faster.
The estate-prep pattern. The operator generates the PDF, prints it, files it in the same physical folder as the will and the medical directives. Every quarter the prior PDF gets shredded and the new one replaces it. The folder is referenced in the estate plan ("see HELM quarterly review for current asset detail"). If something happens, the most recent PDF is the financial map.
The household-CFO pattern. One member of the household runs HELM; the other doesn't engage with finance day-to-day. The quarterly PDF is the deliberate ritual where the second person reviews what's actually happening, asks questions, and stays current enough to take over if needed. The PDF makes the conversation possible because both people are looking at the same artifact at the same time.
None of these workflows are exotic. They're what operators have always wanted from their finance tools and rarely actually got.
The wealth OS that produces a document worth forwarding.
HELM is built so the quarterly PDF is the kind of artifact a CPA or attorney would generate themselves — minus the $500/hour. Manual-first, non-custodial, and aimed at operators who want to think in quarters, not in dashboards.
See HELM →