Methodology guide

How Athena produces next actions, and what stays human

Buyers do not just want “AI planning.” They want to know what is simulated, what is deterministic, what clients see, and where humans supervise. This guide explains Athena’s planning stack in plain language so a founder and a CCO can read the same page. Most wealthtech planning pages describe advisor modules—goals conversations, plan PDFs, portfolio-tied projections—which are useful and often meeting-centric. Athena’s planning platform is built for always-on household decisions that show up as ranked next steps in the client app and as context in Mission Control. If you are evaluating Athena against Orion planning, Advyzon goals workflows, or Nitrogen-style tax and income centers, start here, then read Planning for the product surface.

Book a demo

Most wealthtech “planning” pages describe advisor modules: goals conversations, plan PDFs, portfolio-tied projections. Useful — and often meeting-centric. Athena’s planning platform is built for always-on household decisions that show up as ranked next steps in the client experience and as context in Mission Control.

If you are evaluating Athena against Orion planning, Advyzon goals workflows, or Nitrogen-style tax/income centers, start here — then read Planning for the product surface.

Four layers behind the next step

Surfaces publish these outputs into client Health and chat, advisor plan automation, cards and emails via Communication, and supervision context in Mission Control. The table below is the stack in one place: shared household state, simulation that scores decisions, deterministic engines that produce recommendation records, and a roadmap that orders what the client should do next.

Surfaces publish these outputs into client Health and chat, advisor plan automation, cards and emails via Communication, and supervision context in Mission Control.

Layer What it does Primary output
Household graph People, linked accounts, cashflows, debts, goals, engagement signals Shared state for sim, engines, and surfaces
Simulation Year-by-year projection of wealth, Social Security, tax, RMDs, and contributions across poor / average / good percentile paths Outcome paths to score decisions
Advice engines Deterministic solvers for retire age, SS claiming, liquidity allocation, debt actions, contribution mix, and more Concrete recommendation records
Roadmap & Health Impact-ranked opportunities plus a Financial Resilience Score across weighted domains, explained in firm voice Ordered next actions clients can act on

Engines propose, models explain, and humans decide the exceptions

This split is the compliance-relevant story. Core recommendations come from engines and simulation that are auditable, repeatable, and tied to inputs; large language models explain that methodology in firm voice and power the conversational UX without inventing the recommendation itself. Humans handle exceptions, off-policy situations, and relationship judgment through Mission Control, instead of re-running every routine household manually. Supervised methodology with conversational delivery is the product; “AI advice” as a vibe is not. See also Athena AI and FAQ · methodology.

This split is the compliance-relevant story. “AI advice” as a vibe is not the product — supervised methodology with conversational delivery is. See also Athena AI and FAQ · methodology.

Layer Role Not responsible for
Engines & sim Propose the recommendation record — auditable, repeatable, tied to inputs Conversational tone or free-form inventing advice
LLMs Explain methodology in firm voice; power conversational UX Being the source of truth for the recommendation itself
Humans Exceptions, off-policy situations, relationship judgment via Mission Control Re-running every routine household manually

Domains the engines actually cover

Domain What engines cover What clients see
Retirement & Social Security Retire-age search, wealth needed to retire, sustainable spend, longevity of money, SS claim-age enumeration (including spousal / survivor-aware) scored via simulation Claiming and retire-timing next steps on the roadmap
Education & 529 College funding needs, excess-cash → 529 by wellness posture, plan-selection context where state tax benefit matters Contribution and funding actions
Tax-aware choices Federal tax inside simulation; pretax vs Roth targeting; backdoor / after-tax patterns where relevant; asset location and tax-aware rebalance and harvesting on investing Planning-relevant contribution and location guidance — not a tax-preparer replacement
Debt & credit Refinance gates, consolidation, accelerated payoff prioritization, utilization / payment gauges, score-change attribution Payoff and credit actions with “why it moved”
Cash, emergency, next dollar Emergency fund targeting; ranked allocate-liquidity across debt, match, Roth/pretax, 529, and new accounts; budgeting from linked accounts Where the next dollar should go
Insurance, home, resilience Home affordability and insurance-adjacent guidance; Financial Resilience Score across weighted domains Money-life roadmap, not investment-only health

What clients see instead of a 40-page PDF

Clients do not get a forty-page plan PDF as the daily interface. They get an ordered Health roadmap of next actions, cards and notifications tied to real household state, chat that explains why a recommendation exists and is grounded in the engines, and escalation to your team when the question exceeds automation. That is how planning reduces advisor hours: the methodology keeps working when the calendar is full, and the advisor walks into a review already knowing what changed.

Prospects can enter the same experience through paid clicks from Facebook, Instagram, TikTok, Google, LinkedIn, and YouTube that land in the wellness app. See Advisor ads that land in the wellness app. Many firms still keep a specialist tool for complex estate work or a formal plan document; Athena’s job is the always-on household simulation, the ranked actions, and the client-facing explanation under the firm brand.

Clients do not get a 40-page plan PDF as the daily interface. They get:

  • An ordered health roadmap of next actions
  • Cards and notifications tied to real household state
  • Chat that explains why a recommendation exists — grounded in the engines
  • Escalation to your team when the question exceeds automation

That is how planning reduces advisor hours: the methodology keeps working when the calendar is full. Pair with the wellness funnel for how prospects enter the same experience.

How advisors supervise

Advisors are not asked to re-run every household manually. Mission Control is the household CRM: it surfaces exceptions and overnight activity, household profiles carry plan context, and communication and Athena AI drafts remain reviewable under firm policy. Investing automation—household investment policy, location, and harvesting—stays on-policy with review on Schwab as the custodial connection. See Investing and Security for the engines-propose / models-explain / humans-escalate ladder a CCO will ask about.

Advisors are not asked to re-run every household manually. Mission Control surfaces exceptions and overnight activity; household profiles carry plan context; communication and AI drafts remain reviewable under firm policy. Investing automation (household policy, location, harvesting) stays on-policy with review — see Investing.

Versus suite planning modules

For suite positioning overall, see Orion and Advyzon versus Athena and Compare.

Approach Strength Gap vs Athena thesis
Orion / Advyzon-class planning Strong advisor workflows and portfolio-tied planning Thinner always-on next actions across debt, credit, and cash
Nitrogen-style centers Deep tax/income tooling in places Not a full practice platform with firm-branded wellness and Mission Control
Athena Simulation + engines + roadmap as the layer clients live in Humans stay on exceptions — not a meeting-PDF substitute claim

For suite positioning overall, see Orion & Advyzon vs Athena and Compare.

Questions this usually raises

Is every recommendation LLM-generated?

No. Core recommendations come from engines and simulation that are auditable and tied to inputs. Large language models explain methodology in firm voice and power conversational UX. They are not the source of truth for the recommendation itself.

Can we customize methodology?

Firm voice, brand, escalation policy, and rollout sequencing are part of implementation. Ask in a demo which levers are configuration versus roadmap, and how that maps to your investment policy and advice policies.

Where should compliance readers go next?

Security for SOC 2 Type II, Bedrock Guardrails, and reconstructability. FAQ · methodology for short answers. Athena AI for the client and Mission Control chat surfaces. A working session should walk your advertising and advice policies against the same ladder.

Does this replace our planning software?

Many firms keep a specialized tool for complex estate work or a formal plan document. Athena runs the always-on household simulation, the Health roadmap, and the client-facing explanation. Ask in a demo how that maps to your workflow.

How do prospects enter planning?

Paid clicks from Facebook, Instagram, TikTok, Google, LinkedIn, and YouTube land in the wellness app, where Health and Athena AI can start before the household is a funded wealth client. See From ad to wealth client.