Robo-Advisor Pre-Launch Defensibility Assessment
The SEC's 2021 robo-advisor sweep produced a clear finding catalog: gaps in suitability evidence, weak conflict disclosure, missing portfolio-rebalancing rationale, and uneven oversight of automated processes. This assessment scores a robo-advisor program against each finding and routes gaps to the synthetic-data and checklist artifacts that close them.
What you walk away with
~10 min · 5 categories · 14 items- A finding-by-finding readiness score covering the seven major sweep finding categories.
- A radar chart showing where evidence is thinnest.
- A ranked remediation list mapped to the Robo-Advisor Pre-Launch Data Coverage checklist.
- Calibration anchored against the SEC 2021 robo-advisor sweep summary.
Answer every item (0 of 14 so far) to lock in a banded score and unlock the remediation roadmap. Live category scores update as you go.
Suitability evidence
0 / 3 answeredWhether the robo-advisor captures and uses the suitability evidence required to make defensible recommendations — the largest sweep finding category.
- KYC profile is structurally complete and current
Risk tolerance, time horizon, liquidity needs, investment objectives, experience, tax status. Updated on documented cadence and on material life events.
- Risk-tolerance questionnaire produces defensible scores
Questionnaire is calibrated, validated against actual behavior over time, and the score's mapping to allocation is documented.
- KYC is refreshed on a documented cadence
KYC refreshes annually plus on material change (life event, large deposit, market shock-driven behavioral change). Refresh events are logged.
Disclosure adequacy
0 / 3 answeredWhether disclosures clearly describe the algorithm, the conflicts, and the limitations — the second-largest sweep finding area.
- Algorithmic decision-making is disclosed in plain English
The disclosure describes how the algorithm produces allocations and rebalances, in language a retail client can understand.
- Conflicts of interest are disclosed clearly
Proprietary funds, revenue sharing, payment for order flow, and any preferential routing are disclosed at the relationship and recommendation level.
- Algorithm limitations are disclosed
Limitations (no tax-loss harvesting in non-taxable accounts, no consideration of held-away assets, etc.) are disclosed in the customer-facing materials.
Portfolio management defensibility
0 / 3 answeredWhether allocation decisions, drift management, and rebalancing actions are documented and reconcilable.
- Allocation rationale links to the suitability profile
For each portfolio, the allocation maps explicitly to the client's documented risk profile, time horizon, and tax situation.
- Drift thresholds and rebalance triggers are documented
Drift bands, rebalance triggers, and the resulting transaction set are documented per event.
- Tax-aware actions are documented per event
Tax-loss harvesting, asset-location decisions, and lot selection are documented with rationale.
Oversight of automated processes
0 / 3 answeredWhether humans supervise the algorithm — anomaly detection, exception handling, model risk management.
- Model risk management framework exists
Models have documented purpose, inputs, validation evidence, and review cadence. SR 11-7 framework or equivalent.
- Anomalous algorithm outputs are detected
Outputs that fall outside expected bounds (extreme allocation drift, unusual rebalance activity) trigger human review.
- Exception handling is supervised
When the algorithm produces an output flagged for review, a human reviews, decides, and the decision is logged.
Advertising and performance presentation
0 / 2 answeredWhether marketing claims are accurate, performance is presented per the Marketing Rule, and testimonials follow the post-2021 framework.
- Performance presentation complies with the Marketing Rule
Net-of-fee returns, time-period requirements, and disclosure of methodology all comply with Rule 206(4)-1.
- Testimonials follow the Marketing Rule framework
Testimonials disclose compensation, conflicts, and material conditions. The recordkeeping requirement is met.
Banded score reference
High Sweep Risk
0–35%Multiple sweep finding categories have structural gaps. A future SEC sweep-style examination would surface findings the firm cannot defend.
Next step: Tighten suitability evidence first; that's where the largest share of sweep findings concentrated.
Sweep-Findings Possible
35–60%Foundational compliance is in place but evidence is uneven. A sweep examination would likely produce findings.
Next step: Close the lowest-scoring category; the remediation list ranks them.
Pre-Launch Defensible
60–85%All major finding areas have structured evidence. The robo-advisor could open to public assets with manageable residual risk.
Next step: Tighten oversight of automated processes; rehearse the supervisory walkthrough.
Sweep-Hardened
85–100%Defensibility is operationalized end-to-end with audit-grade evidence packaging.
Next step: Operate steady-state; re-run after each major SEC sweep update.
Key takeaways
- Suitability evidence is where the most sweep findings concentrated. KYC completeness + questionnaire calibration + refresh cadence are the three pillars.
- Algorithm and conflict disclosure in plain English is a recurring finding. Rewriting Form ADV is necessary but not sufficient; customer-facing materials matter.
- Oversight of automated processes is a growing exam focus. Model risk management is the structural answer.
- Marketing Rule compliance is now mechanical. Testimonials without the disclosure framework are a quick finding.
FAQ
Is this scoped to digital-only platforms or hybrid?
Both. Hybrid (digital + advisor) platforms face additional supervision questions about how the human and algorithmic recommendations interact. The scorecard structure applies across both; some categories will score differently in hybrid models.
What about the SEC's predictive analytics rulemaking?
The proposed rules on predictive data analytics (proposed 2023, status as of 2026 varies) would tighten the conflict-disclosure category considerably. Firms preparing for the rule should re-run the scorecard with that in mind.
How does synthetic data help here?
Three places: stress-test the risk-tolerance questionnaire across a wide demographic; verify drift thresholds and rebalance triggers across long market trajectories; exercise edge-case suitability cases (senior clients, ITIN filers, multi-state).