Insights · 12 min read
A practical look at what goes into a robo-advisor: onboarding, rebalancing, compliance, and the real cost to build one. Written for founders and wealth managers.
By GGP Editorial
When a wealth manager or a fintech founder talks about building a robo-advisor, they usually describe the part customers see: a clean app that asks a few questions and then invests the money on its own. The hard part is everything behind that screen. This guide covers what a robo-advisor actually is, the components you have to build, the rules you have to satisfy, and what it costs.
I write from the vendor side. GlobeSoft is a software company founded in 2018 with 40-plus engineers and more than 300 delivered projects. We have built trading and financial systems for brokers across Hong Kong, US, and mainland China markets, so I have seen where these projects actually spend their budget. I will be direct about it.
A note on naming before we start. A robo-advisor is not the same thing as financial management or accounting software, which tracks income, expenses, and ledgers (we covered that in financial management system development). A robo-advisor manages investment portfolios. The two touch money in very different ways, and confusing them in a spec is how projects get scoped wrong from the first call.
A robo-advisor is a service that builds and manages an investment portfolio for a client with little or no human involvement. The client answers a short questionnaire, the system assigns a risk profile, and the software then constructs, monitors, and rebalances a portfolio made up mostly of low-cost funds.
It is not a trading platform. A trading platform gives the user the tools to place their own orders and make their own calls; a robo-advisor makes the calls for them. That distinction drives the entire build. If the user picks the stocks, you are building something closer to what we describe in trading platform development. If the software picks and manages the portfolio, you are building a robo-advisor.
There are two ways to sell one. Direct to consumers, where the robo-advisor is the product. Or white-label, where you license the engine to banks, wealth managers, and advisors who put their own brand on it. White-label deals often close faster and carry lower customer-acquisition cost, but the integration requirements are heavier because you are fitting into someone else's systems.
A robo-advisor is a chain of connected systems, not one app. Here are the pieces, roughly in the order the money touches them.
| Component | What it does | Notes |
|---|---|---|
| Onboarding and KYC | Collects identity, verifies the client, runs sanctions and PEP checks | Usually outsourced to a provider; see KYC integration |
| Risk profiling | A questionnaire that maps answers to a risk tolerance and an investment objective | The output constrains everything downstream |
| Portfolio construction | Maps a risk profile to a target allocation, usually built from ETFs or model portfolios | These rules are your investment methodology |
| Order management | Turns rebalance decisions into real trades with a broker or custodian | The messy integration point |
| Rebalancing engine | Compares actual holdings to targets, generates trade lists, applies thresholds | The technical core |
| Reporting | Performance, fees, tax documents, statements | A surprising amount of work |
| Client portal | Web and mobile views of balances, activity, and projections | What most founders picture as "the app" |
| Billing | Management fees, tiered pricing, invoicing | Easier to get wrong than it looks |
| Compliance and audit | Records every decision and trade for regulators and your own trail | Cannot be bolted on later |
Most of these are not glamorous. The questionnaire and the dashboard are the easy parts. The rebalancing engine, the custodian integration, and the audit trail are where the project lives or dies.
Rebalancing sounds simple: when a portfolio drifts from its target, sell the overweight assets and buy the underweight ones. In practice it is a long list of edge cases.
The system has to handle fractional shares, minimum trade sizes, and cash that is not large enough to buy a full unit. It has to decide whether a drift is big enough to justify trading at all, because trading too often erodes returns through costs and taxes. It has to respect tax lots where they apply and avoid wash sales in markets that penalize them. And it has to do all of this with an audit trail that shows, trade by trade, why a decision was made.
The market data feeding the engine matters more than most founders expect. If your prices are stale or your corporate actions are missing, the engine rebalances against wrong numbers. That is how you generate trades that lose a client money and attract a regulator's attention. Splits, dividends, and mergers all have to be handled correctly, or the target weights quietly stop matching reality.
None of this needs a large team to start, but it does need someone who has built it before. A team that treats rebalancing as a generic "rules engine" rather than a financial subsystem learns the hard way, usually after launch.
You have three honest paths.
Build the engine yourself when the investment methodology is your competitive edge, or when you plan to license it to others. This is the most expensive path and the one that most justifies working with a team that has shipped financial systems before.
License a third-party robo engine when you want to get to market fast and the methodology is not your differentiator. Several providers sell white-label robo infrastructure; you handle the brand, the onboarding, and the client experience, and they run the engine.
Buy a turnkey platform when you are a wealth manager who wants to offer automated portfolios to existing clients without building anything. You give up control and margin, but time-to-market is measured in weeks.
The decision has the same shape as build vs buy software, with one addition: in wealth management, the compliance burden of building means you should only build when the methodology or the distribution justifies it.
Wealth management is a regulated activity in most markets, and a robo-advisor inherits the same rules as a human advisor, sometimes with extra scrutiny.
In the US, a robo-adviser is generally a registered investment adviser under the Securities and Exchange Commission, with disclosure and suitability obligations. The SEC has issued specific guidance for automated advisers, focused on whether the questionnaire actually gathers enough information to give suitable advice.
In Europe, MiFID II applies to firms providing investment services, including suitability and appropriateness assessments and reporting. In the UK the FCA regulates advice and discretionary management. Brazil has CVM, Singapore has MAS, and each has its own registration and conduct rules.
The point for a founder is simple: the rules depend on your product and your markets, and they shape the build. Suitability means the questionnaire and the portfolio mapping need to be auditable. Record-keeping means every recommendation and rebalance needs a trail. Marketing rules mean what you can promise in the app is constrained. Work these out with a compliance advisor before you write code, because retrofitting them is expensive.
This is not legal advice, and the details differ by jurisdiction. The data-protection and security work that underpins all of it is covered in security requirements for fintech platforms.
A robo-advisor sits at the expensive end of fintech builds because it combines a regulated activity, live market data, a custodian integration, and a consumer product all at once. Two things drive the price: the scope of the integrations and the compliance requirements, in that order.
The table below shows the monthly rate of a dedicated engineer through a China-based team like ours. These are the same published rates we quote international clients, and they include project management and QA.
| Level | Monthly rate (China-based, dedicated) |
|---|---|
| Junior (1 to 3 years) | $2,000 to $3,500 |
| Mid (3 to 6 years) | $3,000 to $5,500 |
| Senior (6+ years) | $5,000 to $8,000 |
| Architect / tech lead | $7,000 to $10,000 |
Expect a focused MVP to land between $60,000 and $120,000. That covers onboarding with KYC, a risk questionnaire, a small set of model portfolios, automated rebalancing against one custodian, and a basic client portal. A full production platform runs $200,000 to $500,000 or more, once you add white-label support, tax-aware rebalancing, multi-market data, deeper integrations, and the compliance and audit tooling a regulated firm needs.
Those are ranges, not quotes. The only number that matters is the one from a scoped proposal. We wrote about getting to that number in how to estimate a custom software project. For the wider picture on fintech budgets, see how much does it cost to build a fintech app.
The drivers, roughly in order of impact:
A focused MVP can go from idea to a working product in four to six months with a small team. A production platform with the compliance, integrations, and polish of a licensed service commonly runs nine to eighteen months. The schedule is set by the custodian integration and the compliance review, not by the questionnaire or the dashboard.
Plan to spend the first weeks on the methodology and the regulatory scope, not on the app. If you do not know exactly what "suitable" means for your product and your market, the software cannot enforce it.
Building the dashboard first. The client portal is the easy part and the part founders over-invest in. The rebalancing engine and the custodian integration are what make the product real.
Underestimating the custodian integration. Sandbox environments, settlement timing, and edge cases around corporate actions eat months. Start this integration early.
Treating compliance as a checklist. Suitability and record-keeping are not features you add at the end. They shape the questionnaire, the portfolio rules, and the data model from day one.
Ignoring cash handling. Deposits, withdrawals, fractional cash, and the money that sits uninvested are where a surprising number of bugs and client complaints come from.
Skipping the audit trail. When a client disputes a trade or a regulator asks why the system sold a position, you need the answer in writing, down to the data that drove the decision.
It is a service that builds and manages an investment portfolio for a client automatically, using a risk questionnaire and a set of rules, with little or no human involvement.
A focused MVP typically lands between $60,000 and $120,000. A full production platform with compliance and integrations runs $200,000 to $500,000 or more. These are ranges, not quotes.
In most markets, yes. In the US a robo-adviser is generally an RIA under the SEC, and equivalents apply in Europe under MiFID II, in the UK under the FCA, and elsewhere. The exact requirements depend on your product and markets, so confirm with a compliance advisor.
Build when the methodology or the distribution is your edge. License a white-label engine when you want speed and the methodology is not the differentiator.
Four to six months for a focused MVP. Nine to eighteen months for a licensed production platform, depending on integrations and compliance.
The rebalancing engine and the custodian or broker integration. They look simple on a whiteboard and generate most of the real complexity and risk.
GlobeSoft is a software development company founded in 2018, with 40-plus engineers and more than 300 delivered projects for over 100 clients across the US, Brazil, South Africa, and Singapore. We build custom software, AI products, SaaS platforms, FinTech systems, ERP and CRM, and mobile apps on a stack that includes Java, Spring Boot, Go, Python, React, Vue, and Node.
We have built a multi-market trading system that serves brokers across Hong Kong, US, and mainland China markets, along with financial management and payment products. Wealth management sits naturally on top of that work. You can read more about how we approach financial software in our fintech development services.
If you are planning a robo-advisor or a broader wealth platform, talk to us about your project. Tell us the market, the custodian you plan to use, and whether the methodology is yours or licensed. We will help you work through the build-versus-license call, the scope, and what to ship first.
Tell us what you are building and where you are today. We typically reply within 24 hours.
