Model Portfolio Management: Building, Maintaining, and Scaling Investment Models

Model portfolios are the foundation of scalable wealth management. Without them, every client account is a custom project — requiring individual attention for every rebalancing decision, every new contribution, and every market shift. With well-designed models, an advisor can manage hundreds of accounts with consistent investment discipline and significantly less operational overhead.

But model portfolios only deliver these benefits when they are built thoughtfully, maintained actively, and flexible enough to accommodate the real-world complexities that come with managing actual client money.

Designing Your Model Suite

Start with Risk Profiles, Not Products

The most effective model suite begins with clearly defined risk profiles that map to client objectives — not with a collection of funds you want to use. Start by answering: what are the distinct client profiles your firm serves?

Most RIAs find that five to seven models cover 90% of their client base:

• Conservative: 25-30% equities, 60-65% fixed income, 5-10% alternatives/cash. For retirees drawing income or clients with low risk tolerance.
• Moderate Conservative: 40% equities, 50% fixed income, 10% alternatives/cash. For near-retirees or cautious investors with moderate time horizons.
• Moderate: 60% equities, 30% fixed income, 10% alternatives/cash. The classic balanced portfolio for accumulators with moderate risk tolerance.
• Moderate Growth: 75% equities, 18% fixed income, 7% alternatives/cash. For younger clients or those with higher risk tolerance and longer time horizons.
• Growth: 90% equities, 5% fixed income, 5% alternatives/cash. For clients with long time horizons, high risk tolerance, and no near-term income needs.

Define Asset Class Targets and Ranges

Each model should specify not just target allocations but acceptable ranges. A Growth model might target 45% US Equity with an acceptable range of 40-50%. The range serves as the drift tolerance band for rebalancing — as long as the actual allocation stays within the range, no action is needed.

Setting ranges requires balancing two competing priorities: tight ranges maintain allocation discipline but generate more trades (and potentially more tax events). Wide ranges reduce trading but allow more drift from the intended risk profile. There is no universally correct answer — it depends on the asset class volatility, the account type (taxable vs. tax-deferred), and the client's sensitivity to allocation drift.

Select Securities Systematically

For each asset class in your model, select the specific securities (typically ETFs or mutual funds) that will represent that allocation. Document your selection criteria:

• Expense ratio: Lower is generally better, but not at the expense of tracking quality or liquidity.
• Tracking error: How closely does the fund match its benchmark index? Small tracking differences compound over time.
• Liquidity: Average daily volume and bid-ask spreads matter, especially for larger trades.
• Tax efficiency: ETFs generally create fewer capital gain distributions than mutual funds due to the creation/redemption mechanism.
• Availability: Is the security available through all your custodian relationships?

Handling Client-Specific Customizations

Models provide consistency, but every client has unique circumstances. The challenge is accommodating individual needs without undermining the scalability that models provide.

Account-Level Overrides

The most common customization is holding restrictions. A client who is a corporate executive may hold a concentrated stock position that cannot be sold (due to insider trading restrictions, tax consequences, or emotional attachment). Rather than creating a custom model for that client, apply an override that excludes that holding from rebalancing while adjusting the remaining allocation to account for the concentrated position.

A good model portfolio system lets you handle exceptions without creating a new model for every exception. Overrides should be the escape valve, not the norm.

Tax-Status Awareness

The same client may have both taxable and tax-deferred accounts. While both might be assigned to the same model, the security selection within each account type should differ. Tax-inefficient holdings (REITs, high-yield bonds, actively managed funds) belong in tax-deferred accounts. Tax-efficient holdings (index ETFs, municipal bonds) belong in taxable accounts. This is called asset location, and it is distinct from asset allocation.

Your model system should support account-type-specific security substitutions. The target allocation remains the same across all account types — 25% fixed income is 25% fixed income — but the specific bond fund used may differ between an IRA and a taxable brokerage account.

Household-Level Management

Many clients have multiple accounts — an individual brokerage account, a joint account, a traditional IRA, a Roth IRA, and perhaps a trust. Managing each account independently against its model ignores the opportunity to optimize across the household.

Household-level management treats all of a client's accounts as a single portfolio for allocation purposes, while optimizing the location of each holding across accounts for tax efficiency. This is more complex than per-account management, but it can add significant after-tax value — particularly for clients with large taxable positions and multiple account types.

Maintaining Models Over Time

Scheduled Model Reviews

Models should be reviewed at least annually — and more often if market conditions or your investment thesis change significantly. Common triggers for model updates include:

• Fund changes: A preferred ETF changes its index methodology, increases its expense ratio, or is acquired by another provider.
• Market regime shifts: Sustained changes in interest rates, inflation expectations, or equity valuations may warrant allocation adjustments.
• New asset classes: The availability of new, cost-effective ETFs may enable exposure to asset classes that were previously impractical (e.g., treasury inflation-protected securities, commodities).
• Regulatory changes: Tax law changes may affect the relative attractiveness of certain account types or investment strategies.

Communicating Model Changes

When you update a model, communicate the change to all affected clients before executing trades. Explain what is changing, why, and what the expected impact will be. This is not just good client service — it is a compliance obligation. Clients should understand and consent to material changes in their investment strategy.

Version Control

Maintain a history of every model version, including the date of change, the rationale, and the specific allocation adjustments. This version history serves as compliance documentation and allows you to reconstruct the investment process at any point in time — essential for regulatory examinations and client disputes.

Scaling from 50 Accounts to 500

Model portfolios become non-negotiable somewhere between 50 and 100 accounts. Below 50, a disciplined advisor can manage each account individually (though they probably should not). Above 100, manual management breaks down entirely — it is simply not possible to check drift, calculate trades, consider tax implications, and execute rebalancing for hundreds of accounts without systematic tools.

The transition from manual to model-based management typically follows this progression:

1. Define models: Formalize the investment strategies you are already using into named models with documented allocations and thresholds.
2. Assign accounts: Map every client account to a model. Identify clients who need overrides and document the reasons.
3. Automate monitoring: Use software to track drift across all accounts against their assigned models. Eliminate the spreadsheet review process.
4. Streamline execution: Generate bulk trade proposals for all accounts that need rebalancing, review them as a batch, and execute through your custodian in a single session.
5. Measure and report: Track model performance, account-level performance, and dispersion (the difference between model returns and actual client returns) to ensure consistency.

Each step reduces the time required per account and increases the consistency of your investment process. By step five, you have a practice that can scale to 1,000 accounts with the same team that managed 100.