# AI for Financial Advisors: The Financial Advisor Marketing Agent Product page: https://www.winstondigitalmarketing.com/products/ai-for-financial-advisors/ Price: $47 USD // product / financial advisors ## AI for financial advisors: the problem is not that compliance rejects your marketing. It is that compliance improves it to death. A marketing agent for an advisory firm, built so the review pass is fast instead of destructive. Twelve workflows with the prompts, eight lines you do not cross, and a script that scans your existing site for the ones you already crossed. **$47** [Buy the kit](/contact/?product=ai-for-financial-advisors) [Or read the free guide first](/playbooks/financial-advisor-marketing-compliance-trap/) Instant download after checkout. Fourteen day refund, no questions, including that you read it and decided it was not for you. > General information, not legal advice. Advisory firm marketing is regulated, and what applies to you depends on whether you are state or federally registered, which states you are notice filed in, and your own firm's policies. Nothing here states what any rule requires. Your chief compliance officer is the authority on every question in this kit. ## Three things it writes without being asked. **A performance figure.** A growth number is persuasive and it has read thousands of them, so you get "our clients have averaged" or an illustration of a hypothetical account. Performance is the most regulated category in the field, and the requirements are not about whether the number is true. They are about how it must be presented and what has to accompany it. **A sentence telling the reader what to do.** "You should roll it over." That is how helpful writing sounds, and it is the exact shape of personalized advice delivered to an unknown reader. The tell is second person plus an imperative, and once you can see it you will find it on your own site. **A testimonial with nothing attached.** It does not know you carry disclosure, conflict, supervision and recordkeeping obligations that a restaurant does not. None of that is the model being careless. It is what marketing copy looks like in the material it learned from, and almost none of that material came from a regulated firm. ## The actual problem, which is the opposite of the one people describe. Advisor marketing rarely fails review. It fails because of review. A draft goes to compliance, the specifics come off one round at a time, and what publishes is a page of true, safe, entirely generic statements that could sit on any of ten thousand advisory sites. It clears precisely because it commits to nothing. Then it ranks for nothing, converts nobody, and everyone concludes that compliance makes marketing impossible. It does not. What is missing is that nobody wrote the line down in advance, so every draft rediscovers it by trial, and the safest path through any review is always to say less. This kit inverts the order. You have one conversation with your compliance officer, using the eleven questions in file 04. You write the answers into the profile. Every draft then starts inside the constraints instead of being sanded into them, arrives with its claims and their support next to each other, and carries your approved disclosure blocks already in place. Review becomes a check on placement rather than a rewrite. Being specific about a topic is not the same as being specific about a reader. Almost everything good in this field lives in that gap. ## Where it actually helps. - **The situation page.** The highest-converting asset in this field and the one most firms do not have. Not "retirement planning" but the actual circumstance: what happens to the account when you leave the company, what changes when one spouse dies, what the year of a business sale looks like. Searched by people in the middle of the event, specific enough to rank, and exactly the page an unprepared review destroys. - **Making the review pass fast.** The real product. A draft that arrives with its support attached, the disclosure blocks in place and the questionable lines already flagged is a different object from one your reviewer has to excavate. The bottleneck usually moves before the output improves. - **Reviews, which you may well be allowed to use.** Client testimonials and third-party reviews are available to advisory firms today, and a lot of firms still operate as though they are forbidden. In plenty of markets that leaves the local map result uncontested. The agent will help you run a compliant programme and will not produce a word of it without your disclosure block attached. - **Bios and team pages.** Among the most-visited pages on a small firm's site, and the most neglected. They are also what an AI assistant reads when deciding whether your firm is a real entity worth naming. - **Reading your own site honestly.** It has no attachment to copy that survived a painful review in 2019. Usually the most useful first session anyone has with it. ## Where it does not help, whatever anyone selling it says. - **Anything with a number in it.** Performance, returns, illustrations, projections. Not a limitation to work around. That material has presentation requirements detailed enough that it belongs with your reviewer and your approved templates, and nowhere near a general purpose assistant working from a prompt. - **Telling you where your line is.** Your line depends on your registration, your firm's policies and your reviewer's judgment. The agent enforces a conservative line and marks what to ask about. It does not pretend to know what your compliance officer will accept. - **Anything client-specific.** No plans, no proposals, no account reviews, no client communications. The kit refuses client data entirely and the profile says so at the top. - **Prospecting outreach at volume.** Deliberately out of scope. Solicitation rules, supervision obligations and retention requirements make automated outbound a bad idea at this size, and a cold sequence is poor strategy in a business built on referrals. - **The meeting.** The business closes in a room, on a call, at a kitchen table. The hours this saves are worth having mostly because they go back into that. ## The archive point, which almost nobody thinks about. For a restaurant, an autoposted draft is a bad post. For you it is a records problem. Your public communications sit under supervision and retention obligations. Copy generated in a chat window lives outside every system your firm uses to review and retain. The failure is not the draft, because a draft is not a communication. It is the shortcut: someone pastes it into a scheduler and something reaches the public without ever entering the review or the archive path. So the workflow never changes. Draft here, move it into your firm's review process, publish from your firm's own systems. The agent says this every time it produces something for an outward channel, and it will not suggest a way around it. ## It is not just prompts. **A compliance pre-read you run before a person sees it.** Workflow 9 reads a draft the way your reviewer would and returns every line needing support, every implied recommendation, every forward-looking statement, every missing disclosure, with the compliant rewrite next to each one. It is instructed never to soften a point into meaninglessness: if something cannot be said, it says so and offers a different point. **A script that actually runs.** Point it at your whole site copy in one command and it returns every flagged phrase with a line number and a replacement. Most firms find a performance reference on an old page, a designation displayed in a form its owner does not permit, or a sentence telling the reader what they should do. Python 3, no dependencies, no network, nothing leaves your machine. **The rules as data.** Thirty-two flagged phrases with their reasoning, in a file you can edit. Every time your reviewer rejects a phrasing you add it, and from then on the assistant and the script both enforce your firm's actual line rather than a generic conservative one. **Eleven questions for your compliance officer.** Written to be brought as a single list. Most compliance officers would rather answer eleven questions once than reject the same sentence eleven times, and that hour is the highest-value one in the whole thirty day plan. ## What it is not. Not software, not a course, not a list of advisor AI tools. A folder of markdown files you read and then hand to an assistant you already pay for. It never touches client data. No accounts, no balances, no holdings, no plans. The agent is instructed to stop if it sees something that looks like client information. It drafts public marketing copy from facts you approved, and every draft goes through your firm's review path before it reaches anyone. ## Refunds. Fourteen days, no questions, for any reason including that you read it and decided it was not for you. Email john@winstondigitalmarketing.com. ## Frequently asked questions. ### Can financial advisors use AI for marketing? Yes, and the useful question is how rather than whether. An assistant is good at situation pages, planning explanations, bios, review replies, newsletters and event copy. It is unreliable on anything containing a performance figure, a forward-looking statement or a recommendation, and it produces all three confidently if nothing stops it. The workable setup is a written instruction file that refuses those categories every time and attaches your approved disclosure blocks automatically, with every draft still going through your firm's normal review and retention path. ### What should a financial advisor never let AI write? Eight things. Performance figures, returns, backtests and hypothetical illustrations. Market predictions stated as fact, including hedged ones. Testimonial or review content without the firm's required disclosures and process. Personalized advice in public copy, whose tell is second person plus an imperative. Specific securities or products presented as recommendations. Registration or credential claims beyond what the firm holds, including anything implying a regulator endorses it. Statements of what a rule requires. And anything that publishes or sends outside the firm's review and retention path. ### Why does compliance ruin marketing copy, and can this fix it? Because the line is usually not written down, so every draft finds it by trial, and the safest route through any review is to say less. Round after round of that produces a page that is true, safe and generic. The fix is not a cleverer draft, it is deciding the constraints in advance with your compliance officer and writing them into a file the assistant reads first. Drafts then start inside the line rather than getting sanded into it, and review becomes a check on placement and support rather than a rewrite. ### Are advisors allowed to use client testimonials and Google reviews? Testimonials and third-party reviews are available to advisory firms, and many firms still operate as though they are not, which in a lot of markets leaves the review-driven local result uncontested. The permission carries obligations around disclosure, any compensation or conflicts, written agreements where they apply, and supervision and recordkeeping. This kit does not tell you what your obligations are, because they depend on your registration. It gives you the question to put to your compliance officer, and the agent refuses to produce any testimonial content until your approved disclosure block is in the profile. ### Is it safe to use AI with client information? This kit says no and is built so the question does not arise. The agent works from a profile of public, approved facts about the firm. It is never given accounts, balances, holdings, plans or client communications, the profile says so at the top, and the agent is instructed to stop and tell you if something that looks like client data appears in the conversation. ### Does this replace my compliance officer? No, and it is written to make that explicit. The kit contains no rule citations at all, on purpose, because what applies to your firm depends on your registration and your own policies, and a confident citation from a marketing kit is worse than none since it feels settled and stops you asking. Your chief compliance officer is the authority. The kit is a way to capture their answers once and have them enforced automatically afterward. ### What do I need to run this? An AI assistant you already pay for, or a free tier. The kit is plain markdown and works with Claude, ChatGPT, Gemini or Copilot. The included checker script needs Python 3, which is already on every Mac and most Linux machines. Nothing else, and no subscription beyond the assistant. ## Related reading, free. This kit is built on guides already published here, which cost nothing. Read those first if you would rather not pay for anything. - the financial advisor marketing compliance trap - financial advisor SEO - GEO for financial advisors - digital marketing for financial advisors - the rest of the products ## AI for Financial Advisors The marketing agent, twelve workflows, the compliance pre-read, the checker script, and the eight lines you do not cross. $47. [Buy the kit](/contact/?product=ai-for-financial-advisors) [Read the free guide first](/playbooks/financial-advisor-marketing-compliance-trap/)