Robo-Adviser
A robo-adviser is a digital service that builds and manages an investment portfolio automatically — a questionnaire sets the customer's risk level, an algorithm allocates the money across low-cost index funds, and software rebalances it — for a fee of a fraction of a per cent of assets. It brought portfolio management to accounts too small for a human adviser and forced the fees of the whole wealth industry down.
- Also known as
- Robo-advisor, Digital wealth manager, Automated investment service
- Used at
- retail investing, workplace pensions and ISAs, and as the wealth module inside neobanks and brokers
- Binding
- regulated as investment advice or discretionary management — FCA authorisation in the UK, SEC registration as an investment adviser in the US, MiFID in the EU — with suitability obligations on the questionnaire
- Typical fee
- 0.20–0.35% a year on top of fund costs of 0.05–0.20%, against 0.75–1.25% for a traditional adviser
- Typical economics
- average accounts of $20,000–50,000 at the standalone firms; customer acquisition costs of $200–400; payback measured in years, which is why most were bought or built inside larger firms — JPMorgan bought Nutmeg for about £700m in 2021, UBS agreed $1.4bn for Wealthfront in 2022 and then walked away
- Parties
- the customer, the adviser firm, the custodian holding the assets, and the fund managers whose ETFs the portfolio holds
How a robo-adviser works in practice
The product is a pipeline. Onboarding collects the customer's goal, horizon and tolerance for loss in a questionnaire that doubles as the regulatory suitability assessment. The answers map to one of a handful of model portfolios — a set of index funds in fixed proportions, from cautious to adventurous. Money is invested to the model, dividends are reinvested, and when market moves push the proportions off target, software sells the overweight funds and buys the underweight ones. Tax features — loss harvesting in the US, ISA and pension wrappers in the UK — sit on top. A human is available for questions but does not manage the money.
The economics are the difficulty. At 0.25% a year, a $30,000 account yields $75; if it cost $300 in marketing to acquire, the customer must stay four years before the firm has earned back the cost of finding them, before serving them at all. Standalone robo-advisers therefore either raised prices, added human advice for larger accounts, or were acquired by banks and brokers that already had the customers and needed the product. Inside a bank, the same service is cheap to distribute and turns deposits into managed assets; that is what the acquirers bought.
For a buyer the diligence questions are the account size distribution, the churn, the acquisition cost by channel, and whether the suitability process would survive a regulator's review of a customer who lost money.
Example
A standalone robo-adviser has $3.0bn of assets across 100,000 accounts and charges 0.25%: $7.5m of revenue. Its acquisition cost is $300 per customer and it adds 20,000 a year — $6m of marketing — with 8% annual churn. Servicing and technology cost $4m. It loses $2.5m a year and needs about $4bn of assets to break even at its current cost base. A bank with two million retail customers offers 2% of assets, $60m; inside the bank, acquisition cost falls to near zero because the customers are already there, and the same product on $5bn of the bank's customers' money earns $12.5m against $4m of cost. The price is a fraction of what the firm's last venture round implied, and the board takes it.
Common mistakes
- A questionnaire that satisfies the regulator on paper but maps every answer to the same three portfolios. - Pricing to win customers at an account size that can never pay back the acquisition cost. - Reporting AUM growth in a rising market as if it were customer growth. - Rebalancing rules that generate taxable events the customer did not need. - In an acquisition, paying for the technology when the value is in the acquirer's own distribution.