Why order flow moves to software and client trust does not
The work here splits into two kinds of task. One kind is structured and repeatable: completing sales order tickets and submitting them for processing, pulling quotes, reviewing market data and financial periodicals, and keeping accurate records of every transaction. Software has chipped away at that side for two decades, and language models now draft the research notes and client summaries that used to take an afternoon.
The other kind is judgment under rules. Interviewing a client to establish their financial position, goals and tolerance for loss is a conversation, not a form. Explaining how markets, margin and trading practices work to someone who is scared of losing money takes patience and reading the room. Recommending a specific security means someone licensed puts their name to it, and answers for it later. That accountability is why the hands-on part of the job has not moved.
Scale matters too. The Bureau of Labor Statistics counts about 489,570 people in this occupation, with median pay of $78,660 and projected growth of 1.4% from 2025 to 2035 (BLS, 2025). Slow growth and good pay is the setting where firms hold headcount flat and push more accounts through the same desk. No robotics are involved: this is screen, phone and meeting work, which is part of why the software side moves faster here than in trades. For the wider group, see the services sales representatives family.
What AI handles, what it assists, and what stays with you
Tasks our task list puts in the AI-does group account for 36% of task time on this job. These are the mechanical pieces: routing and recording client-requested transactions, monitoring market data for price moves, compiling account and performance reports, and screening prospect lists before anyone picks up a phone. The “Can AI do it?” figure for this occupation is 51, where 100 would mean every hour of task time is already covered; the coverage method explains how that is built.
The assisted group covers 59% of task time. Here a person stays in charge and the tool speeds up the draft. Preparing financial plans and investment proposals is faster when a model pulls the holdings, runs the scenarios and writes the first version. Explaining stock market terminology and the mechanics of an order also sits here: the tool can produce the plain-English version, but the agent decides what this client actually needs to hear.
Tasks our list leaves with a person make up 5% of task time. That is the smallest slice, and it is the part that carries the licence: taking a suitability decision for a named client, signing off a recommendation, handling the call when a position has gone against someone, and keeping a long relationship through a bad quarter. The honest read is task erosion plus fewer junior seats, not a job disappearing.
Good to know: firms can automate the paperwork around a recommendation long before anyone lets software make the recommendation itself.
The evidence: no head-to-head test yet
Our “Is it better than a person?” evidence grade for this occupation is D, which means there is no direct, published test of an AI system against licensed sales agents doing this job’s real tasks. So we publish no parity number for it. Plenty of finance benchmarks measure document reading, summarizing and market question answering, but none of those measures client suitability, disclosure or the outcome of a recommendation over years.
What would settle it is specific: a supervised trial where an AI system and licensed agents work the same client scenarios, scored by an independent reviewer on suitability, disclosure quality and client understanding, with results published rather than described in a vendor case study. Regulator-reported findings on supervised AI recommendation pilots would count too. Until something like that exists, treat confident claims in either direction as opinion. Our grading rules are set out in the quality parity method, and the full approach sits on the methodology page.
When the mix could shift
Most likely between 2034 and 2043 (8 in 10 of our scenarios). Two things would pull that earlier. First, cost: the annual cost of tooling for this work sits far below the cost of a licensed agent, and the page above shows both ranges. Second, self-service. Every simple account that migrates to an app is an account a junior agent no longer learns on, and entry-level hiring is where this kind of change shows up first; the entry-level hiring tracker follows that signal.
Two things hold it back. Licensing and supervision rules keep a named, qualified person responsible for a recommendation, so firms cannot simply hand suitability calls to a model. And client trust is slow to move, especially for larger portfolios where one bad decision is expensive. Those brakes are why the structured half of the job changes faster than the licensed half. The replacement-year method explains how we build the range above, and the finance and insurance sector page shows how this role compares with its neighbors.
How to stay needed in this role
Lean into the tasks our list leaves with a person. Own the suitability judgment: be the one who can explain, in writing, why this product fits this client. Own the hard conversations, the loss calls and the risk talks that no tool will take for you. And own the long relationship, including referrals, family money and the slow business that compounds over a decade.
Two skills pay for themselves. One is supervising AI output: checking a model-drafted plan or client letter for errors and compliance problems before it goes out, and documenting that check. The other is compliance fluency, because the agent who understands disclosure and recordkeeping rules becomes the person a firm keeps when the drafting work shrinks.
Close neighbors are worth a look if you are weighing a move. Compare this role with insurance sales agents, personal financial advisors and financial and investment analysts, which share tasks with this job but split differently between automated and human work. You can put any two of them side by side on the compare tool, or see where finance roles land against everything else in the full job rankings.