Custodian Data Aggregation

The Multi-Custodian Reality

Most RIAs Use Two or Three Custodians

The average registered investment advisor does not work with a single custodian. Firms grow through advisor recruitment, acquisitions, and client referrals—each of which can introduce a new custodian relationship. A firm that started on Schwab may have added Fidelity to accommodate advisors who brought their books with them. A subsequent acquisition added Pershing accounts. The result: the same firm now manages client assets across three separate institutions, each with its own portal, its own data model, and its own reporting tools.

This is not a niche situation. Industry surveys consistently show that the majority of RIAs with more than $500 million in AUM maintain relationships with two or more custodians. For firms managing $1 billion and above, operating across three or more custodians is the norm, not the exception.

The Siloed View Problem

Each custodian provides a portal that is genuinely excellent—for that custodian's accounts. You can see every Schwab client in Schwab Advisor Services, run performance reports, generate statements, and monitor positions in real time. The same is true in Fidelity's institutional platform and Pershing's NetX360. The problem is that none of these portals knows anything about the other custodians.

A client with a retirement account at Schwab and a taxable account at Fidelity appears as two separate, unconnected individuals across two separate systems. You cannot see that household's total assets, run a consolidated performance report, or make asset allocation decisions that account for both accounts without manually combining data from two portals. Multiply this across hundreds or thousands of clients and the operational burden becomes enormous.

What Custodian Aggregation Solves

Firm-Wide AUM

The most immediate benefit of custodian aggregation is a single, accurate number for firm-wide assets under management. When assets are spread across three custodians, firm-wide AUM is not a number you can look up—it is a number you calculate manually, usually by exporting reports from each custodian and summing columns in a spreadsheet. Aggregation replaces that process with an automated daily calculation that draws from every custodian feed simultaneously. The number is always current, always complete, and never requires a staff member to compile it.

Household Views Across Custodians

A household typically includes multiple individuals with multiple accounts at potentially multiple custodians. A spouse holds a Roth IRA at Schwab; the other spouse holds a 403(b) that rolled to Fidelity; and they have a jointly-held trust account at Pershing. From any single custodian's perspective, those three accounts belong to three separate client records. Custodian aggregation combines them into a single household entity—one AUM figure, one performance history, one consolidated view of asset allocation across all accounts.

This matters not just for reporting but for investment decision-making. Advisors managing asset allocation across a household cannot optimize unless they can see all assets in one place. Aggregation makes cross-custodian household management operationally viable.

Unified Performance Reporting

Performance reporting that covers only some of a client's accounts is not useful for communicating total relationship performance. When a client's assets span Schwab and Fidelity, performance reports that cover only one custodian understate or misrepresent the client's actual investment results. Aggregation enables performance reporting that captures every account, every custodian, and the complete investment picture—without requiring advisors to manually reconcile reports from multiple sources.

Cash Management Across Custodians

Cash management and sweep analysis require visibility into cash positions across all custodians simultaneously. If a client is sitting on excess cash at Fidelity while being invested aggressively at Schwab, that imbalance is invisible without aggregation. Firms that aggregate custodian data can monitor cash positions firm-wide, identify sweep opportunities, and ensure consistent cash management across the entire client base regardless of custodian.

Account Transition Tracking

When a client moves assets from one custodian to another—a common occurrence during firm transitions, mergers, or custodian consolidations—tracking the in-flight assets requires visibility into both the sending and receiving custodian simultaneously. Aggregation provides a single view that shows assets departing one custodian and arriving at another, with position reconciliation that flags discrepancies before they become problems.

How Custodian Data Feeds Work

File Formats

Custodians deliver data in several formats, and the format varies by custodian and data type. The most common formats are:

• Flat files (fixed-width): Legacy format still used by many large custodians. Fields occupy fixed character positions, requiring a specification document to parse correctly. Robust but inflexible.
• CSV (comma-separated values): Widely used for position, transaction, and account files. Simpler to parse than fixed-width, but field definitions vary significantly across custodians.
• XML: More common in newer feeds and API-adjacent delivery. Self-describing format that includes field names, though custodian XML schemas differ from one another.
• JSON: Used in modern API-based custodian connections. Increasingly common as custodians modernize their data delivery infrastructure.

A single custodian may use multiple formats for different data types—flat files for positions, CSV for transactions, and an API for account information. An aggregation platform must handle all of them.

Delivery Mechanisms

Custodians deliver data through several channels:

• SFTP (Secure File Transfer Protocol): The dominant delivery mechanism for daily custodian feeds. Custodians deposit files to a secure server on a defined schedule; consuming systems pull those files and process them. Reliable and widely supported.
• API connections: Newer custodians and updated infrastructure from established custodians increasingly support REST APIs for data retrieval. API connections can support more frequent data refresh than daily file drops.
• Direct feeds: Some custodians support dedicated data connections for large institutional clients, typically with higher data volumes, lower latency, and direct database-style access.

Data Categories

Custodian data feeds cover several distinct categories of financial data:

• Positions: Current holdings, quantities, prices, and market values for every security in every account. The core of any custodian data feed.
• Transactions: Buys, sells, dividends, interest payments, transfers, fees, and corporate actions. Transaction history is critical for performance calculation and tax reporting.
• Cash: Cash balances, pending settlements, and money market holdings. Often delivered as a separate file or data category from positions.
• Account information: Account type, registration, owner information, beneficiaries, and account status. Links financial data to client identity.
• Corporate actions: Splits, mergers, spin-offs, dividends, and other events that affect position quantities or values. Critical for accurate performance calculation.

Timing and Frequency

Most custodian data feeds operate on an end-of-day cycle: files are generated after market close, deposited to SFTP servers overnight, and available for consumption by early morning. This means aggregated data typically reflects the prior day's closing positions and values.

Some custodians offer intraday updates for specific data types, and API-based connections can support more frequent refreshes. Real-time data remains relatively rare in traditional custodian feeds, though newer custodian platforms are beginning to offer it for certain data categories. For most operational reporting purposes, end-of-day data is sufficient; for active trading oversight or cash management, intraday updates add meaningful value.

Custodian-Specific Considerations

Each major custodian has distinct characteristics that affect how aggregation works in practice.

The Normalization Challenge

Why Raw Aggregation Fails

Simply collecting data files from multiple custodians and dumping them into a single database does not create a usable dataset. It creates a larger mess. The core problem is that each custodian represents the same real-world financial concepts using different conventions, formats, and identifiers. Without normalization, the combined dataset is internally inconsistent and cannot be reliably queried or reported against.

Account Numbering

Every custodian uses its own account numbering scheme. Schwab account numbers follow one format, Fidelity uses another, and Pershing uses a third. A client with accounts at all three custodians appears in three separate records with three different account identifiers, with no native connection between them. Normalization assigns each real-world account a canonical internal identifier and maps the custodian-specific account numbers to it. This is the foundation of cross-custodian household views.

Security Identifiers

Securities can be referenced by CUSIP, ISIN, ticker symbol, or custodian-specific internal identifiers. Custodians do not consistently use the same identifier for the same security. Apple Inc. common stock may appear as AAPL in one feed, as CUSIP 037833100 in another, and as an internal custodian code in a third. Normalization maps all representations to a canonical security master, ensuring that position data for the same security from different custodians can be combined and compared.

Transaction Codes

Transaction type codes vary across custodians. A dividend reinvestment transaction might be coded "DRIP" at Schwab, "DIV" at Fidelity, and something entirely different at Pershing. Corporate action codes, transfer codes, and fee codes each have their own custodian-specific vocabularies. Normalization maps all transaction codes to a standardized transaction taxonomy, enabling accurate performance calculation and transaction analysis across all custodians.

Date and Value Formats

Seemingly minor formatting differences create significant processing challenges. Settlement dates, transaction dates, and as-of dates may be formatted as MM/DD/YYYY at one custodian and YYYYMMDD at another. Numeric values may use different decimal conventions. Currency fields may or may not include currency codes. Normalization standardizes all date formats, numeric representations, and field conventions so that data from different custodians can be combined arithmetically without errors.

How a Data Platform Handles Custodian Aggregation

Pre-Built Custodian Connectors

A purpose-built data platform for wealth management arrives with pre-built connectors for major custodians. These connectors handle the custodian-specific details of data retrieval: SFTP authentication, file format parsing, field mapping, and error handling for common custodian feed issues like delayed files or partial data loads. Pre-built connectors eliminate the months of custom engineering work that building custodian feeds from scratch requires, and they encode institutional knowledge about each custodian's quirks and edge cases.

Automatic Normalization

Normalization logic runs automatically as data is ingested from each custodian. Account numbers are mapped to canonical identifiers, security identifiers are resolved against a master security reference, transaction codes are translated to standardized types, and date formats are converted to a single internal representation. The resulting normalized dataset behaves as if all data came from a single source with a consistent schema.

Position Reconciliation

Custodian data is not always clean. Files arrive with missing fields, duplicate records, mismatched settlement dates, and position discrepancies that require investigation. A data platform includes reconciliation logic that compares incoming custodian data against prior day positions, flags unexpected changes, and surfaces discrepancies for review. Automated reconciliation replaces the manual daily reconciliation process that many back-office teams spend hours on every morning.

Unified Household Mapping

Once accounts from all custodians are normalized, household mapping links them together. The mapping logic connects accounts to individuals (using name matching, tax ID matching, or direct configuration), and individuals to households. An account at Schwab and an account at Fidelity belonging to the same person become linked records in a single household entity. Household AUM, household performance, and household asset allocation are then calculated automatically across all linked accounts regardless of custodian.

Data Freshness and Monitoring

Custodian feeds occasionally fail to arrive on time, arrive incomplete, or contain data quality issues. A production-grade platform monitors feed arrival, validates completeness, and alerts operations teams when feeds are late or contain anomalies. Data freshness timestamps communicate clearly which data is current and which is stale, preventing downstream users from making decisions on outdated information without realizing it.

What You Can Do With Aggregated Custodian Data

Firm-Wide Dashboards

With all custodian data normalized into a single dataset, firm leadership gains a real-time view of total AUM, AUM by advisor, AUM by custodian, and AUM by client segment. Dashboard metrics that once required manual compilation—sometimes taking days to produce—update automatically whenever new custodian files arrive. For firms reporting to leadership, boards, or parent companies, the accuracy and timeliness of AUM reporting improves dramatically.

Multi-Custodian Performance Reporting

Performance reports that cover a client's complete asset picture—not just the accounts at one custodian—require aggregated data as their foundation. With normalized positions, transactions, and corporate actions from all custodians in a single dataset, performance calculation engines can compute time-weighted returns, money-weighted returns, and benchmark comparisons that accurately reflect total client wealth. Client-facing performance reports become more meaningful; advisor-facing analytics become more accurate.

Cash Sweep Analysis

Cash sweep programs vary across custodians. Understanding which clients have excess cash, in which accounts, at which custodians—and whether that cash is earning competitive yields—requires visibility across all custodians simultaneously. Aggregated custodian data enables analysis of cash positions firm-wide, identification of uninvested cash opportunities, and consistent cash management that accounts for all accounts regardless of where they are held.

Fee Billing Reconciliation

Advisory fee billing is calculated against AUM, which means accurate billing requires accurate position data from every custodian. With aggregated data, billing platforms can pull verified AUM from a single source rather than attempting to combine exports from multiple custodian portals. Billing discrepancies—where fees were calculated on incorrect AUM—become easier to identify and correct. Compliance documentation of fee calculations becomes more defensible.

AUM Tracking for Compensation and Reporting

Advisor compensation is often tied to AUM managed, new assets gathered, or revenue generated—all of which require accurate, complete AUM data across all custodians. Aggregation enables compensation calculations that reflect the complete scope of each advisor's client relationships, not just the accounts at one custodian. It also enables the regulatory AUM reporting required on Form ADV to be calculated consistently and accurately across all custodian relationships.

Transition Monitoring

Account transitions—moving assets from one custodian to another—are a routine part of managing a multi-custodian firm through advisor changes, acquisitions, and strategic custodian consolidations. Monitoring transitions requires visibility into both the sending and receiving custodian: tracking assets as they depart one institution and confirming their arrival at another. Aggregated data from both custodians provides the complete picture needed to manage transitions confidently and identify delays or discrepancies before they become problems.

Frequently Asked Questions

Conclusion

Multi-custodian operations are a fact of life for most RIAs of scale. Schwab, Fidelity, Pershing, and newer platforms like Altruist each serve different segments of a firm's client base, and no single custodian portal provides visibility across all of them simultaneously. The result is fragmented data that makes firm-wide reporting, household management, and operational oversight genuinely difficult without deliberate aggregation infrastructure.

Custodian data aggregation solves this by collecting data from every custodian, normalizing it to a consistent schema, and delivering a unified dataset that reflects the firm's complete asset base. The normalization step is what separates functional aggregation from raw data dumps: standardized account identifiers, security references, transaction codes, and date formats make the combined dataset analytically useful rather than merely large.

Firms that invest in proper custodian aggregation infrastructure gain capabilities that are simply impossible with fragmented data: accurate firm-wide AUM, true household views across custodians, consolidated performance reporting, and the foundation for AI and analytics applications that require complete data to deliver reliable results. The complexity of the problem is real—but a purpose-built platform with pre-built custodian connectors and normalization logic makes it tractable.