For thirty years the unit economics of advice held a quiet constant: a senior advisor could service roughly 80 to 120 households before quality broke, and the firm priced around that ceiling — usually 1.00% AUM on the first dollar, sliding to 0.50%-0.75% at the upper tiers, or an annual flat fee anchored to the same households-per-advisor capacity assumption. The AI-leveraged practice in 2026 quietly broke that constant. Jump, Zocks, FinMate, Sybill, Pulse360, Holistiplan, FP Alpha, Wealth.com, RightCapital, eMoney, MoneyGuidePro, Salesforce FSC + Einstein, Wealthbox, Redtail Engage, Orion Eclipse, 55ip, BlackRock Aladdin Wealth, Microsoft Copilot, Smarsh, and Global Relay together produce a documented 8-12+ hours/week of recovered advisor capacity per the Schwab 2026 RIA study, the Zocks productivity claim, and the L4 Ch5 L2 ROI dashboard most firms in the program now run. Capacity is up. Cost-to-serve is down. Marginal price to add another household is lower. None of which automatically translates to a new fee schedule — because every fee change passes through ADV Part 2A, the engagement letter, the Marketing Rule 206(4)-1 disclosure standard, and the Reg BI Conflict Obligation under §240.15l-1(a)(2)(iii). This lesson installs the framework for re-pricing under AI leverage without becoming the next AI-washing enforcement headline.

Why the Unit Economics Shifted

The 2026 advisor's hour mix is not what it was in 2023. Kitces' time-and-task split — roughly 20% in client meetings, 36% in meeting prep / planning / servicing, 15% in prospecting, 20%+ in admin — used to be load-bearing because there was no leverage on the prep / planning / servicing / admin buckets. Jump and Zocks ate meeting notes, follow-up drafts, and CRM updates. Holistiplan ate 1040 OCR + bracket scans. FP Alpha and Wealth.com ate trust / will / POA extraction. RightCapital, eMoney, MoneyGuidePro ate scenario regeneration. Pulse360 and Wealthbox ate post-meeting recap + activity logging. Salesforce FSC + Einstein and Redtail Engage ate next-best-action prompts. Orion Eclipse + 55ip + BlackRock Aladdin Wealth ate tax-loss harvesting + rebalance + IPS-aligned trading. Smarsh and Global Relay ate the FINRA Rule 4511 + SEC Rule 204-2 retention overhead.

The aggregate effect — documented in the L4 Ch5 L2 ROI dashboard that most program firms run by Q3 2026 — is that a senior advisor's true work-product hour can now cover noticeably more households at equivalent or higher service quality. The exact number varies (90 → 120, or 120 → 160, depending on segment, pod structure per L5 Ch4 L1, and the proprietary-workflow depth from L5 Ch2 L1) but the direction is unambiguous. The cost-to-serve curve moved. The pricing did not — yet. That gap is the strategic opening this lesson addresses.

Four Pricing Models Under AI Leverage

Each of the four canonical advisor pricing models behaves differently under AI cost-to-serve compression. None is universally right. Each carries its own ADV Part 2A disclosure profile and its own Reg BI Conflict Obligation surface.

AUM Pricing

The dominant model. AI leverage under AUM pricing tends to flow first to advisor compensation and firm margin rather than to client fee reduction — because the marginal-revenue / marginal-cost wedge widens with no automatic mechanism to compress it back into the schedule. The honest 2026 question for an AUM-priced practice is whether AI savings should flow (a) to higher advisor comp, (b) to firm margin and reinvestment in proprietary workflows (per L5 Ch2 L1), (c) to fee reduction across the schedule, (d) to expanded scope at the same fee — adding a financial-planning-included tier that previously cost extra (estate audit per L4 Ch7 implications, tax-equivalent-yield workflow per L5 Ch7 L2, dynamic-spending policy per L5 Ch7 L3). The fourth option is the most defensible under Reg BI Conflict Obligation §240.15l-1(a)(2)(iii) and the Marketing Rule because it converts cost-to-serve savings into client-observable value without triggering a fee-cut competitive cascade. The ADV Part 2A fee-schedule item-5 typically does not require amendment in option (d); options (a)-(c) require ADV amendment per L5 Ch7 L6 off-cycle filing framework if fee schedule changes materially.

Flat-Fee Pricing

The fastest-growing model in 2024-2026 (mass-affluent + young-accumulator segments). Flat-fee pricing under AI leverage compresses naturally — the practice's cost-to-serve drop is more visible to the client because the price is not buried in AUM math. Two patterns: (1) hold the flat fee constant but expand scope (the option-(d) move above) — add estate gap audits, tax-equivalent-yield workflows, dynamic spending policy, U4 DRP drafting capabilities, ADV amendment support, etc.; (2) introduce a lower-cost flat-fee tier for younger / less-complex households that previously fell outside the practice's served band — a "$1,500/year AI-leveraged planning subscription" that uses Holistiplan, RightCapital, and Jump-driven prep without consuming the senior advisor's hour budget. Either pattern requires ADV Part 2A item-5 fee-schedule disclosure, engagement-letter clarity on scope, and Marketing Rule 206(4)-1 compliance on any marketing of the new tier.

Subscription Pricing

Common at the planning-first practices (XY Planning Network alumni, mass-affluent CFPs, dual-income professionals). Monthly or quarterly subscription decouples fee entirely from AUM. AI leverage's most natural home — because the subscription model already prices time-and-scope, and AI shifts the time per unit of scope. The 2026 pattern at strong subscription practices: introduce tiered subscriptions (e.g., Foundation $200/mo, Growth $400/mo, Wealth Builder $700/mo, Family Office Lite $1,500/mo) where each tier names the included AI-leveraged deliverables (annual tax review using Holistiplan, estate audit using FP Alpha + Wealth.com, RSU/equity-comp modeling, dynamic-spending policy, Roth-conversion screen, 529-funding plan, business-owner-exit framework per L5 Ch2 L1 proprietary workflows). The "included deliverables" framing is critical for Marketing Rule 206(4)-1 + L4 Ch7 L1 substantiation file. Each deliverable should match a named output the firm has actually produced and can show on demand.

Hybrid Pricing (Project + AUM, or Retainer + Project)

Increasingly common at firms serving complex households (business owners, equity-comp executives, multi-generational families). The hybrid splits a base retainer or AUM line from project-priced specialty work (Roth-conversion ladder design, ISO/QSBS exercise modeling per L3 Ch6, SLAT / ILIT / CRT / CLAT / DAF decision tree, business-owner-exit workflow). AI leverage on hybrid pricing typically expands the menu of project-priced specialty deliverables (because the marginal cost to add another specialty output drops) without compressing the base retainer or AUM line. The ADV Part 2A disclosure must enumerate the project-price menu with sufficient specificity to satisfy item-5 fee-schedule transparency.

ADV Fee-Schedule Update Mechanics

Any change to the fee schedule passes through ADV Part 2A item 5 (Fees and Compensation). The annual updating amendment (the CCO's mid-March workflow per L5 Ch7 L6) catches non-material year-end changes; material changes trigger an off-cycle (prompt) amendment under IA-1992 within 90 days under Form ADV General Instruction 4. Three patterns common in 2026 AI-leveraged practices.

Material Fee Cut (Compress Schedule)

An AUM-priced firm cuts the top tier from 1.00% to 0.85% on first $1M, slides change accordingly down the schedule. Material → ADV Part 2A item-5 amendment, engagement-letter update, client notification (typically 30-60 days advance per state-law fiduciary practice), Marketing Rule 206(4)-1 compliance on any marketing of the cut, L4 Ch7 L1 substantiation file documents the cost-to-serve analysis that justified the cut, Reg BI Conflict Obligation §240.15l-1(a)(2)(iii) implications since the fee is the comp-conflict surface. The firm should retain the cost-to-serve workbook (Jump time-saved, Zocks meeting-time saved, Holistiplan / FP Alpha extraction time saved, Orion Eclipse / 55ip rebalance time saved) per FINRA Rule 4511 + SEC Rule 204-2 retention via Smarsh / Global Relay — five-year retention with two-year accessible per Rule 204-2(e)(1).

Expanded Scope at Same Fee

Same fee schedule. The engagement letter and ADV Part 2A item 4 (Advisory Business) gain a new bulleted list of included deliverables — estate gap audit using FP Alpha + Wealth.com, dynamic-spending policy using RightCapital / eMoney + Guyton-Klinger logic per L5 Ch7 L3, tax-equivalent-yield workflow per L5 Ch7 L2, AI-augmented client portal access per L5 Ch5 L1. Generally non-material in the fee-schedule sense (item 5 unchanged), but item 4 changes require amendment. Marketing Rule 206(4)-1 substantiation file documents each new deliverable with a named template, named tool, named output. This is the most defensible AI-savings disposition because the client observes expanded value without a fee cut that signals commodity pricing.

New Tier Introduction

The practice adds a new pricing tier — typically a lower-cost flat-fee or subscription tier for previously-out-of-band households. Material → ADV Part 2A item 5 amendment for the new fee schedule, item 4 amendment for the new advisory service, engagement-letter template for the new tier, Marketing Rule 206(4)-1 compliance for the marketing of the new tier, and Reg BI Conflict Obligation analysis because the new tier creates a new comp surface that may differ from the firm's traditional schedule. The firm's L4 Ch7 L1 AI Compliance Specialist (or equivalent role per L5 Ch4 L1) maintains the substantiation file for any external claim the new tier makes ("AI-leveraged planning," "comprehensive plan for $1,500/year," etc.) — the 2024-2025 Delphia and Global Predictions AI-washing settlements set the floor on what a credible substantiation file looks like, reinforced by the January 2026 SEC staff FAQs.

Conflict Disclosure When AI Lowers Cost-to-Serve

The Reg BI Conflict Obligation under §240.15l-1(a)(2)(iii) requires identification and disclosure of material conflicts of interest associated with recommendations. AI cost-to-serve compression introduces a subtle but real conflict surface that the 2026 ADV Part 2A and Form CRS need to acknowledge.

The Asymmetric-Savings Conflict

If AI savings flow to firm margin and advisor comp rather than to client fee, the firm should disclose — at minimum in ADV Part 2A item 5 supplement — that the firm uses AI tools that reduce its cost-to-serve, and that this reduction may flow to firm and advisor compensation rather than to client fee. The disclosure language matters: it must be "clear and prominent" per Marketing Rule 206(4)-1 reasoning even though the rule technically governs marketing rather than ADV. The 2024-2025 enforcement pattern — particularly the Delphia / Global Predictions settlements — has made it clear that the SEC and state regulators read AI disclosures aggressively. Generic "we use AI" language fails. Specific "we use AI in meeting notes, document extraction, planning scenario generation, rebalance, and compliance archiving, which reduces our cost-to-serve" language passes.

Recommendation-Influence Conflict

If the firm uses AI-driven next-best-action prompts (Salesforce FSC + Einstein, Wealthbox AI, Redtail Engage, Pulse360) that may surface specific products or strategies, the Conflict Obligation requires identification of any compensation differential that biases the AI's prompt — for example, if the firm receives higher comp on annuity recommendations than on advisory-only recommendations, and the AI surfaces annuity prompts at frequency disproportionate to the advisory-only book. The L4 Ch5 L2 audit logs (which AI tool surfaced which prompt at which household at which time) become Reg BI evidence under §240.15l-1(a)(2)(iii). Smarsh / Global Relay archive these audit logs under FINRA Rule 4511 + SEC Rule 204-2.

Pricing-Cascade Conflict

If the firm introduces a new lower-cost tier (above), there is a structural conflict between the lower-cost tier and the traditional tier — the firm has an incentive to keep complex households on the traditional schedule even when their needs would be adequately served by the new tier. ADV Part 2A item 11 (Disciplinary Information adjacent) and item 14 (Client Referrals and Other Compensation) reasoning extend by analogy — the firm should disclose the tier-placement methodology and the criteria that determine which tier a household sits in. The L1 Ch2.3 Cardinal Rule and L1 Ch1 framing on fiduciary duty apply: the recommendation memo for tier placement should be documented per L2 Ch7 L2 Reg BI rollover-pattern documentation logic, even though tier placement is not technically a rollover recommendation.

Cost-to-Serve Quantification — The Substantiation File

Any external claim of AI-driven cost-to-serve compression — in marketing, in M&A diligence per L4 Ch8, in regulatory inquiry — needs a substantiation file. The L4 Ch7 L1 Marketing Rule audit framework + the January 2026 SEC staff FAQs require contemporaneous documentation. The 2026 best-practice substantiation file contains:

• Baseline hour-mix audit: pre-AI advisor hour mix by category (meeting time, prep, planning, servicing, prospecting, admin) measured against Kitces' time-and-task buckets at a named date (e.g., Q4 2023 baseline).
• Tool-by-tool time recovery: Jump / Zocks / FinMate / Sybill (meeting notes + follow-up drafts), Holistiplan (1040 OCR + bracket scan), FP Alpha + Wealth.com (estate doc extraction), RightCapital / eMoney / MoneyGuidePro (scenario regeneration), Pulse360 + Wealthbox + Salesforce FSC + Redtail Engage (CRM activity logging + next-best-action), Orion Eclipse + 55ip + BlackRock Aladdin Wealth (rebalance + tax-loss harvesting), Microsoft Copilot (doc drafting), Smarsh + Global Relay (compliance archiving) — each with a measured hours/week recovery range, a sampling methodology, and a named period.
• Aggregate hour-mix delta: post-AI hour mix vs. baseline. Per-advisor average. Per-pod average per L5 Ch4 L1 pod structure.
• Cost-to-serve conversion: hours recovered × fully-loaded advisor cost = cost-to-serve dollar delta. Per household. Per segment. Per tier.
• Disposition narrative: where the recovered cost flowed (advisor comp, firm margin, fee reduction, scope expansion). Linked to ADV Part 2A item 5 amendments + engagement-letter updates + Marketing Rule 206(4)-1 substantiation for any external claim.
• Retention: Smarsh / Global Relay archive per FINRA Rule 4511 + SEC Rule 204-2 — five-year retention with two-year accessible. Outside counsel review at least annually per L5 Ch4 L1 AI Compliance Specialist workflow.

Case Study — $1.4B RIA Pricing Restructure

A $1.4B Midwest RIA with 18 advisors and 1,650 households ran an AI-driven pricing restructure across Q3 2025 - Q2 2026. Pre-restructure: AUM 1.00% on first $1M, 0.80% $1-3M, 0.60% $3-10M, negotiated above; 280 mass-affluent households below $500K paying min fees of $4,000/year often above their proportional AUM cost-to-serve. Q3 2025 baseline hour-mix audit per Kitces buckets: 19% meetings, 38% prep/planning/servicing, 14% prospecting, 22% admin, 7% advice judgment. Q1-Q2 2026 post-AI hour-mix (after Jump, Zocks, Holistiplan, FP Alpha, Wealth.com, RightCapital, Wealthbox, Salesforce FSC + Einstein, Orion Eclipse, 55ip, Microsoft Copilot, Smarsh deployment with L5 Ch4 L1 pod restructure): 24% meetings, 28% prep/planning/servicing, 16% prospecting, 17% admin, 15% advice judgment.

Hours recovered per senior advisor: ~9.5 hours/week (Schwab 2026 study range + L4 Ch5 L2 ROI dashboard observed). Cost-to-serve compression: $1,650/household/year median across the book. Disposition decision after a 90-day deliberation through the L5 Ch4 L1 executive interlock (CEO, COO, CCO, CTO, Head of Wealth): (a) AUM schedule held constant — no fee cut; (b) included scope expanded — estate gap audit per FP Alpha + Wealth.com, dynamic-spending policy per RightCapital + Guyton-Klinger per L5 Ch7 L3, tax-equivalent-yield workflow per L5 Ch7 L2, AI-augmented client portal access per L5 Ch5 L1, ADV Part 2A item 4 updated to reflect expanded service; (c) new mass-affluent flat-fee tier launched — $2,000/year for households below $500K, ADV item 5 amended, separate engagement letter, separate Marketing Rule 206(4)-1 substantiation file, separate Reg BI Conflict Obligation analysis; (d) margin and advisor-comp distributed the residual.

Q3 2026 outcomes: zero AUM-tier household departures attributable to pricing (vs. industry baseline ~3-5% annual retention loss); 180 households moved to the new mass-affluent tier (net of 100 onboarded from prior-out-of-band relationships); ADV amendment filed off-cycle in May 2026 per Form ADV General Instruction 4; Marketing Rule 206(4)-1 substantiation file passed outside counsel review with no findings; L4 Ch7 L1 audit framework rated 9/10; firm AI maturity attribute per L4 Ch8 valuation model improved from 0.7x to +1.2x EBITDA multiple impact (Mercer Capital / ECHELON Q3-Q4 2025 framing — top-quartile RIAs at ~8x-10x adjusted EBITDA, premium-top ~11.6x, with AI maturity contributing +0.5x to +1.5x of that spread). The pricing restructure became a documented competitive-differentiation narrative at the firm's two acquisition conversations in H2 2026.

Key Takeaways

• The 2026 AI-leveraged advisor's unit economics shifted: 8-12+ hours/week recovered per the Schwab 2026 RIA study + Zocks productivity claim + L4 Ch5 L2 ROI dashboard. Cost-to-serve curve moved. Pricing did not — yet. The gap is the strategic opening.
• Four canonical pricing models behave differently under AI leverage: AUM (savings flow first to advisor comp / firm margin / reinvestment unless deliberately redirected), flat-fee (compresses naturally, two patterns: hold + expand scope, or add lower-cost tier), subscription (tiered subscriptions with named AI-leveraged deliverables — Foundation / Growth / Wealth Builder / Family Office Lite), hybrid (project + retainer / AUM, expand specialty-deliverable menu without compressing base).
• Every fee change passes through ADV Part 2A item 5 (Fees and Compensation) — annual updating amendment for non-material year-end changes per L5 Ch7 L6 workflow; off-cycle prompt amendment under IA-1992 + Form ADV General Instruction 4 within 90 days for material changes. Item 4 (Advisory Business) amends for scope expansion. Engagement-letter update. Marketing Rule 206(4)-1 substantiation file per L4 Ch7 L1 + January 2026 SEC staff FAQs + 2024-2025 Delphia / Global Predictions precedent.
• Reg BI Conflict Obligation §240.15l-1(a)(2)(iii) creates three AI-specific conflict surfaces: asymmetric-savings conflict (disclose if AI savings flow to firm / advisor comp rather than client fee), recommendation-influence conflict (next-best-action prompts that may surface higher-comp products), pricing-cascade conflict (incentive to keep complex households on traditional tier when new lower-cost tier could serve them). L1 Ch2.3 Cardinal Rule and L1 Ch1 fiduciary framing apply to tier-placement memos.
• The substantiation file is the load-bearing artifact: baseline hour-mix audit per Kitces buckets, tool-by-tool time recovery (Jump / Zocks / Holistiplan / FP Alpha / Wealth.com / RightCapital / eMoney / MoneyGuidePro / Pulse360 / Wealthbox / Salesforce FSC + Einstein / Redtail Engage / Orion Eclipse / 55ip / BlackRock Aladdin Wealth / Microsoft Copilot / Smarsh / Global Relay), aggregate hour-mix delta, cost-to-serve dollar conversion, disposition narrative, FINRA Rule 4511 + SEC Rule 204-2 five-year retention with two-year accessible, annual outside-counsel review.
• $1.4B RIA case study: 18 advisors, 1,650 households, Q3 2025 - Q2 2026 restructure. ~9.5 hours/week recovered per senior advisor, $1,650/household/year cost-to-serve compression, AUM schedule held constant + scope expanded + new mass-affluent flat-fee tier ($2,000/year) launched + margin / comp distributed residual. Q3 2026: zero AUM-tier departures attributable to pricing, 180 households on new tier, ADV off-cycle amendment filed, Marketing Rule substantiation cleared outside counsel, AI maturity attribute per L4 Ch8 + Mercer Capital / ECHELON Q3-Q4 2025 framing improved from +0.7x to +1.2x EBITDA multiple impact within the top-quartile ~8x-10x premium-top ~11.6x band.