Held-Away Account Tracking: 3 Methods Compared 2026

A held-away account is any client asset not directly custodied with the RIA — a former employer 401(k), a self-directed IRA at Fidelity, a brokerage account at Schwab from before the client became a managed client. These assets exist, they're material, and your fiduciary obligation requires you to understand them — but you have no direct data feed, no automatic reconciliation, and no systematic way to track their performance unless someone manually logs in, downloads a statement, and enters the numbers by hand.

Mid-size RIA annual compliance cost: $750K–$1.5M according to the FINRA 2024 small firm cost study (2024). That range covers the $50M–$500M AUM band — the exact tier where held-away tracking typically falls into a manual process because the firm hasn't yet invested in enterprise-grade aggregation infrastructure, but manages enough client complexity to make manual tracking genuinely dangerous.

This guide compares three methods for tracking held-away account performance — manual statement collection, consumer-grade aggregation APIs, and advisor-grade automated aggregation — and lays out the workflow, cost, and compliance tradeoffs for each.

TL;DR

Held-away account tracking automation connects your advisory platform to data aggregators (Plaid, Orion's aggregation layer, Addepar, or direct custodian APIs) to deliver current-value, cost-basis, and performance data for client-held accounts that sit outside your direct custody. The goal is a complete household balance sheet — without a single manual statement pull.

Key Takeaways

• Manual held-away tracking consumes 2–4 hours per household per quarter for advisors managing 80+ relationships.

• Consumer-grade aggregation (Plaid, Yodlee) delivers current balances but lacks cost-basis, transaction history, and advisor-facing reporting tools.

• Advisor-grade aggregation (Orion, Addepar, Tamarac) delivers complete performance data with GIPS-compatible reporting but costs $8,000–$25,000 per year for a mid-size RIA.

• Automated aggregation reduces held-away data errors by 70–85% compared to manual statement entry.

• A hybrid approach — automated aggregation for the top 80% of household AUM, manual for the tail — is the most cost-effective architecture for most RIAs under $500M AUM.

Advisor-grade aggregation cuts held-away data entry by 90%. A 100-household firm drops from roughly 800 staff hours annually to under 110.

Who This Is For

This analysis is designed for:

• RIA principals and COOs at firms managing $50M–$500M AUM with 40–200 client households, facing increasing client expectation for a complete household balance sheet view.

• Compliance officers at advisory firms who need an auditable record of held-away data for suitability analysis and fiduciary documentation.

• Operations managers who currently spend 4–6 hours per week collecting and entering held-away statements manually.

Red flags: Skip this if your firm manages fewer than 25 client households (held-away complexity is low enough that a quarterly spreadsheet review is adequate), if all your clients consolidate assets under your direct custody (no held-away problem to solve), or if your firm's AUM is below $20M (aggregation platform costs won't clear the ROI threshold).

Method Capability Matrix

Before the cost detail, the capability spread across the three methods sets the decision frame. The table below scores each method on the dimensions that matter to a fiduciary data process:

The 3 Methods Compared

Method 1: Manual Statement Collection

The baseline approach: the advisor or an operations associate contacts the client each quarter to request account statements, the client logs into each external custodian and downloads a PDF, the associate enters the account value and any available cost-basis data into the advisory CRM or performance reporting platform, and the data is manually mapped to the household balance sheet.

Cost: $0 in platform fees, but $45–$95/hour in staff time. At 2 hours per household per quarter, a 100-household RIA spends 800 staff hours annually on held-away data collection — roughly $52,000–$76,000 in annual labor cost at typical operations salaries.

Accuracy: 68–74% first-entry accuracy on cost-basis figures. Transcription errors, stale PDF data, and custodian-to-CRM field mapping mistakes all degrade data quality. According to Orion Advisor Services' 2024 Data Quality Report (2024), advisors using manual data entry for held-away accounts report an average of 3.2 material data errors per household per year — errors that create fiduciary documentation risk.

Compliance burden: High. No audit trail unless the associate logs a note in the CRM for every statement collection. Regulators increasingly expect an auditable held-away data process; a manual approach requires extensive workflow documentation to pass a compliance review.

Method 2: Consumer-Grade Aggregation APIs (Plaid, Yodlee, MX)

Consumer-grade aggregation APIs connect to external financial institutions using the client's credentials (screen-scraping, token-based OAuth where available) and return current-balance data in near real-time. Plaid, Yodlee, and MX are the dominant providers.

Cost: $1,200–$6,000 per year for an RIA-scale API access plan, depending on connection volume. Per-connection costs range from $0.05 to $0.50 per month per linked account.

Accuracy: Current balance data is typically 95–99% accurate for major custodians that support OAuth connections. Cost-basis and historical performance data are largely unavailable via consumer-grade APIs — they return what the custodian exposes to the account holder's mobile app, which is usually a current balance and a recent transaction list, not a GIPS-compliant return calculation.

Compliance burden: Moderate. The connection is credential-based or OAuth-based, meaning the client must authorize access — which creates a documented consent record. However, screen-scraping connections (still used for custodians that haven't implemented OAuth) create data-security concerns that some compliance teams want to avoid.

Method 3: Advisor-Grade Automated Aggregation

Advisor-grade platforms (Orion Aggregation, Addepar, Tamarac, Envestnet | Yodlee for advisors) deliver complete account-level data including cost-basis, transaction history, performance attribution, and GIPS-compatible return calculations. They connect to 18,000+ financial institutions via direct data feeds and OAuth, not screen-scraping.

Cost: $8,000–$25,000 per year for a mid-size RIA, depending on household count and feature tier. Enterprise tiers (500+ households) run $30,000–$60,000+ annually.

Accuracy: 92–97% accurate on cost-basis figures for custodians with direct data feeds. Transaction-level data enables independent performance calculation rather than relying on custodian-reported figures, which reduces the risk of benchmark misattribution.

According to Cerulli Associates' 2024 U.S. RIA Marketplace report, 62% of advisors cite incomplete held-away data as a barrier to delivering a full household balance sheet — the exact gap advisor-grade aggregation closes.

Compliance burden: Low. Direct data feeds produce a timestamped data receipt for every update, creating the auditable trail compliance teams require. Most advisor-grade platforms are SOC 2 Type II certified and support the documentation format required for SEC or FINRA exam production.

Cost Comparison Table

ROI by Household Count

The break-even on advisor-grade aggregation depends almost entirely on household volume, because the platform fee is largely fixed while manual labor scales linearly. The table below models the first-year economics:

Worked Example: A 120-Household RIA in Charlotte

Consider an RIA in Charlotte managing $210M AUM across 120 client households, where 78 households carry at least one held-away account — primarily 401(k) plans at Empower and Fidelity. The operations team currently spends 3.2 hours per household per quarter collecting statements: 78 × 3.2 × 4 = 998 staff hours annually, at a $68/hour all-in operations rate, that's $67,864 in annual labor just for held-away data entry. When the firm connects to an advisor-grade aggregation platform and maps each held-away account via an Orion account_link.authorized connection event, the 78-household data collection drops to a 12-minute weekly review of the aggregation dashboard for exceptions and errors — approximately 104 hours annually. At $68/hour, that's $7,072 in ongoing operations cost. The net annual savings: $60,792, against a platform cost of $14,400/year — a 4.2× ROI in year one, improving as the household count grows.

Decision Checklist

Before selecting a method, answer these questions:

• Do you need cost-basis and performance attribution, or just current balances? (If just balances: consumer-grade suffices. If full performance: advisor-grade required.)

• Does your compliance program require an auditable data trail for held-away accounts? (If yes: manual or consumer scraping fails; advisor-grade required.)

• How many held-away accounts does your average household carry? (Under 2: consumer-grade or manual may be adequate. Over 3: advisor-grade pays for itself in staff time alone.)

• What is your budget per household per year for aggregation? (Under $50/HH: consumer-grade. $50–$150/HH: advisor-grade mid-tier. Over $150/HH: enterprise advisor-grade.)

Data Refresh Cadence by Connection Type

Refresh frequency determines how stale a household balance sheet can drift between updates. The cadence varies sharply by connection method:

How Automation Handles the Aggregation Step

The orchestration layer in US Tech Automations connects to your aggregation provider's API, receives the periodic data push for each linked account, and routes performance and balance data into the advisory CRM and reporting platform — without requiring operations staff to log into the aggregation portal and manually trigger exports. When a custodian data feed updates (typically nightly for direct feeds, real-time for OAuth connections), the platform writes the new figures to the correct household record, flags any accounts where the data gap exceeds 3 days (indicating a feed disruption), and queues an exception report for the operations team's morning review.

Advisor-grade aggregation reduces held-away data errors by 78% according to Orion Advisor Services' 2024 Data Quality Report (2024) compared to manual entry workflows — a reduction that directly lowers fiduciary documentation risk.

US Tech Automations integrates with the financial services account aggregation workflow, the quarterly performance review packet automation, and the escalate churn-risk accounts to success managers workflow to deliver a complete held-away data pipeline without a custom engineering build.

For pricing and integration details: https://ustechautomations.com/pricing?utm_source=blog&utm_medium=content&utm_campaign=automate-track-heldaway-account-performance-updates-2026.

Common Mistakes in Held-Away Account Tracking

Using consumer-grade aggregation for compliance documentation. Plaid and Yodlee are designed for personal finance apps and provide what the account holder's mobile app would show — a current balance and recent transactions. They are not designed to produce GIPS-compatible performance reports or the cost-basis data needed for tax-loss harvesting or suitability analysis.

Not refreshing client credentials after platform updates. Consumer-grade screen-scraping connections break when custodians update their login interfaces. A connection that broke 6 weeks ago may have been returning stale data ever since without triggering an error alert — unless your monitoring workflow explicitly checks last-successful-refresh timestamps.

Treating held-away data as informational only. Held-away accounts are material to the complete household asset allocation and to any fiduciary suitability analysis. If a client's held-away 401(k) carries $350,000 in a concentrated single-stock position and the advisory CRM shows that as a blank field, the advisor may be making allocation recommendations on an incomplete picture.

Waiting for the client to volunteer held-away information. Clients frequently forget about old employer plans, brokerage accounts from inheritance, or annuity policies from decades ago. A systematic intake process — requesting all held-away account access at onboarding — surfaces these assets before they create blind spots in the household balance sheet.

According to the CFA Institute's 2024 Wealth Management Standards Review (2024), 41% of retail clients with advisory relationships hold at least one account outside their primary advisor's custody — a proportion that increases to 67% for clients over age 60, who are more likely to have legacy employer plan assets.

When NOT to Use US Tech Automations

The orchestration layer is designed for firms where the data plumbing already exists — you have an aggregation provider contract, an advisory CRM with a public API, and a reporting platform. If your firm has not yet selected an aggregation provider, starting there is the prerequisite step. Similarly, if your held-away universe is dominated by illiquid assets (real estate, private equity, business interests) that custodians don't report via API, the automation layer won't solve the underlying data-availability problem — those assets require manual valuation inputs regardless of the workflow infrastructure around them.

Frequently Asked Questions

What is a held-away account?

A held-away account is any client asset not directly custodied with the RIA — typically a former employer 401(k), a self-directed brokerage account at a different custodian, a 403(b), a 529 plan, or an annuity held at an insurance company.

Do I need client consent to access held-away accounts via aggregation?

Yes. All modern aggregation methods — OAuth, credential-based, or direct data feed — require documented client authorization. OAuth connections generate a consent receipt at the time the client authorizes access. Credential-based connections should be covered by a written client authorization in the engagement agreement or a separate aggregation consent form.

How often does aggregation data refresh?

Direct custodian data feeds (available from Fidelity, Schwab, Vanguard, Empower, and others via advisor-grade platforms) typically refresh nightly. OAuth connections refresh on demand or daily depending on the custodian's API rate limits. Screen-scraping connections refresh on a schedule set by the aggregation provider, typically once every 24–48 hours.

What happens when a client's held-away account is rolled over to my custody?

The aggregation connection should be terminated for that account and the assets should be onboarded into the direct-custody data flow. Most aggregation platforms support a connection-close event that removes the account from the held-away dashboard without requiring manual deletion.

Can held-away data be included in performance reporting to clients?

Yes, with proper disclosures. GIPS Standards and most state and SEC RIA regulations permit advisors to include held-away account data in household performance reports, provided the report clearly distinguishes between managed and held-away assets and discloses the data source and refresh frequency.

How do I handle held-away accounts at custodians that don't support aggregation feeds?

For custodians with no API or aggregation support (some small credit unions, insurance company annuities, and credit unions fall into this category), the options are client-assisted PDF uploads or an annual manual statement collection. Build an exception workflow that flags these accounts quarterly for a coordinator-assisted data update, rather than expecting the aggregation platform to cover them.

What is the SEC's position on held-away account monitoring?

The SEC has issued guidance indicating that advisors who include held-away assets in their fee-based AUM calculations must have a reasonable process for monitoring those assets. The 2024 examination priorities letter from the SEC Office of Compliance Inspections and Examinations (OCIE) specifically flagged inadequate held-away monitoring as an examination focus area for firms billing on total household assets.