Aggregation is a reading problem first
Aggregating accounts across custodians sounds like an arithmetic problem — add up the balances, sum the holdings, done. In practice, almost all of the actual difficulty sits earlier than any arithmetic: getting the raw figures out of statements that were never designed to be combined with anything else in the first place.
A private bank statement, a brokerage statement, an alternative-investment custodian's report — each prints the same underlying kind of information in a completely different layout, using different words for the same concept. Aggregation that starts from a wrong or inconsistently read figure produces a wrong total, no matter how careful the summing step afterward is.
Why this isn't a simple lookup
Two custodians rarely describe the same kind of account the same way. One statement lists “cash and equivalents” as a single line; another breaks it into a checking sub-account and a money-market sweep sub-account that need to be summed to get the equivalent figure. A third might not separate cash from short-term fixed income at all.
Getting a genuinely comparable figure out of each statement means reading past the label each custodian happens to use and finding the actual underlying number the aggregation needs — a task that's manageable for one account and increasingly error-prone the more accounts and custodians are involved.
What consistent aggregation actually means
Consistent aggregation means the same kind of figure is found and categorized the same way across every account, every period, even when a custodian's exact statement layout shifts slightly from one period to the next — a line renamed, a sub-account added, a summary page reformatted.
That consistency is what makes a consolidated view trustworthy over time. An asset allocation that looks like it shifted because one custodian's cash sub-account quietly stopped being counted isn't a real allocation shift — it's a change in method disguised as a change in the portfolio, and it's exactly the kind of gap a careful reviewer eventually catches, usually at an inconvenient moment.
Why internal transfers are the riskiest figure
Money moving between two of a family's own accounts — a cash sweep from a private bank into a brokerage account, a distribution from one entity into another's operating account — is where an aggregation is most likely to overstate wealth if it isn't caught. The transfer shows up as new money coming into one account and doesn't automatically register as the same money leaving another.
Counted naively, a single $500,000 internal transfer looks like $500,000 of new wealth rather than the same $500,000 changing location — an error that compounds every time cash moves between accounts within the structure, which for an active family office can be often.
Why the same word means different things
“Fixed income” at one custodian might mean government and investment-grade corporate bonds only. At another, it might also include preferred shares or structured notes. Aggregating figures under the same category label without checking what each custodian actually means by it produces an allocation view that looks precise and isn't.
| Term | Where the variation usually shows up |
|---|---|
| Cash and equivalents | Whether money-market funds and short-term CDs are included |
| Fixed income | Whether preferred shares and structured notes are counted |
| Alternatives | How private equity, hedge funds and real assets are grouped |
| Net asset value | Whether accrued fees and pending trades are reflected |
This is exactly why reading and categorizing are kept as two visible steps here rather than one opaque one. Getting each statement's raw figures right is a document-reading problem this feature is built for. Deciding how your family office defines each category is a judgment call that stays with your team, applied consistently once made.
What gets read
| Field | Typical source |
|---|---|
| Account holder, account number, custodian | Statement header |
| Opening and closing balance | Summary section |
| Individual holdings and their values | Portfolio or positions detail |
| Deposits, withdrawals and internal transfers | Activity detail |
How it works
Upload each custodian's statement
For every account the aggregation needs to cover, however differently each is laid out.
Each account is read
Balances, holdings and transactions pulled with a confidence score.
Figures mapped into one structure
The same categories applied consistently, regardless of each custodian's own labels.
Internal transfers identified
Cash moving between the family's own accounts recognized rather than double-counted.
Exported
Excel, CSV or JSON, with every figure traceable to its source statement.
An account, aggregated
A private bank statement reports a closing balance of $3.2M across a mix of cash, bonds and a discretionary equity sleeve, presented as one blended relationship summary rather than clean separate line items.
| Component | Amount |
|---|---|
| Cash and money-market sweep | $410,000 |
| Fixed income | $1,180,000 |
| Discretionary equity sleeve | $1,610,000 |
| Total (matches statement's own closing balance) | $3,200,000 |
Because the three components are broken out and separately tagged, they aggregate correctly into the allocation view alongside every other custodian's holdings — rather than sitting as one opaque blended figure that would flatten a meaningful part of the picture.
Handling staggered statement dates
Custodians rarely issue statements as of the exact same date. One might close its statement period on the last calendar day of the month; another reports on a slightly different cycle. Aggregating figures that are technically as-of different dates without noting the gap produces a consolidated figure that looks precise and carries a hidden timing mismatch.
Each account's statement date is captured and carried alongside its figures, so a consolidated view can note exactly how current each component actually is — rather than presenting a blended total as though every account were measured at the same instant.
Manual vs. automatic
| Manual | Automatic |
|---|---|
| Categories mapped by hand, differently by whoever's doing it | The same category mapping applied consistently every period |
| Internal transfers tracked in memory or a loose note | Transfers between accounts identified and reconciled automatically |
| Statement dates assumed to line up | Each account's actual as-of date captured and carried through |
| Redone by hand as custodians are added | Same method applies regardless of custodian count |
From one family to many client families
A single family with a handful of accounts is a manageable manual exercise. A multi-family office with dozens of client relationships, each with its own custodian roster, multiplies the number of statements that need reading and aggregating every reporting period without multiplying the staff time available to do it.
Aggregating each family's accounts the same way regardless of scale keeps the per-account effort flat as the client roster grows, with each family's own category definitions still applied correctly to its own figures.
Who uses this
Single and multi-family offices
Custodian statements aggregated the same way every period, without a manual reassembly each time.
Private wealth controllers and bookkeepers
Each custodian's specific figures read and aggregated independently, then rolled up.
Outside accountants preparing family reports
A verified, traceable aggregation received rather than raw statements to work through.
Teams managing internal-transfer scrutiny
Every cash movement between accounts tracked, flagged if it looks like a genuine discrepancy rather than a transfer.
Edge cases worth knowing
An account that changed custodian mid-period — moved from one private bank to another, for instance — needs its balance history stitched across the transition rather than treated as two unrelated accounts, since the same underlying money simply changed hands.
A joint account with more than one beneficial owner, or an account held in a way that spans two entities in the family structure, is flagged for confirmation of how it should be attributed rather than assigned to one owner by default assumption.
Why consistency matters more than speed
An aggregation done quickly but inconsistently from period to period is worse than one done carefully but the same way every time — an inconsistent aggregation produces a wealth trend that doesn't actually reflect the family's position, which is precisely the kind of gap a careful principal or advisor eventually notices.
Reading the same categories the same way every period, with every deviation flagged rather than silently absorbed, is what keeps the reported trend trustworthy to the people relying on it.
What a confidence score actually tells you
Every aggregated figure carries a confidence score, and it's worth being specific about what that means. A high score means the figure was read from a clear, well positioned printed number with no ambiguity about which line it belongs to — not that it necessarily matches your specific category definition.
A lower score flags exactly where a human look is worth the time: a blended summary line, a figure split across a page break, a category label that doesn't match the expected pattern. Reviewing the handful of flagged fields each period, rather than re-checking every figure from scratch, is what makes the whole process fast without becoming careless.
What this doesn't do
Doesn't apply your own category definitions automatically
It reads and separates the raw figures; mapping them to your specific allocation categories stays your call.
Doesn't judge which custodian is 'right' when figures disagree
It flags the disagreement; resolving it stays a human decision.
Doesn't calculate performance or returns
It aggregates balances and transactions; performance calculation is a separate step your team or platform handles.
Doesn't replace your accounting or advisory team
It surfaces figures and inconsistencies; interpreting them stays with your team.
Getting your first statement aggregated
There's no setup step required before the first aggregation — no custodian list to configure in advance, no category taxonomy to define up front. Upload the current period's statements and the aggregation runs against them directly, surfacing every field a standard consolidated view typically needs.
The first period is also the natural point to note how your family office defines each category — since there's no prior period yet to compare against, this is the baseline every future period's consistency check measures against.
Most teams run their first aggregation against a recent, already-familiar statement before relying on it live — a useful way to confirm every figure lines up with what you'd expect from an account you already know well, before trusting it on a period you haven't reviewed by hand.
Why an audit trail matters here specifically
Aggregated wealth figures tend to get scrutinized more closely than most financial data — by the principal directly, by outside advisors preparing a tax filing, occasionally by an auditor. Each of those readers eventually asks the same question about at least one figure: where did this number actually come from. An aggregation that can answer that instantly, for any figure, is worth more than one that's merely fast to produce.
Every aggregated figure keeps a reference back to its source statement, the page it was read from, and the confidence the extraction assigned to it — not as an afterthought, but as the same output every figure carries, so answering “where did this come from” is a lookup, not a research project.
Onboarding a new custodian relationship
Adding a new custodian to an existing aggregation doesn't require reconfiguring anything for the accounts already in place. The new custodian's statements are read the same way as every other account's, using the same category definitions already established, so the first period with a new custodian slots into the existing consolidated view rather than starting a separate, parallel process that has to be merged in later.
The one thing worth doing deliberately when a new relationship starts is confirming, on that first statement, that the category mapping matches expectations — a private bank with an unusual way of presenting a specific asset class is easiest to catch and correct on day one, before several periods of history build up under a mapping that needs revisiting.
Security and privacy
Uploads are encrypted with TLS from end to end.
Processing runs on infrastructure with SOC 2-aligned controls.
Original documents are deleted shortly after processing.
Nothing you upload is ever used to train AI models.
For financial data that reveals the full shape of a family's wealth, that matters — details are on the security page.
Feeding this into an existing system
Aggregated figures rarely need to live only inside an exported spreadsheet. A family office that already maintains its own internal tracking sheet, or a reporting platform with an import function, typically pulls this output directly into whatever system already holds the rest of the family's records — the aggregation step becomes the reliable feed that system was previously missing, rather than a parallel process run alongside it.
For teams building this into a recurring internal workflow, the API exposes the same aggregation programmatically, so a scheduled job can pull newly processed statements into an internal system automatically rather than requiring someone to export and re-import a file by hand each period.