Real-time sale alerts, right on your phoneSubscribe to updates
    Selling & sync4 min read

    Phoenix, the built-in AI assistant

    Phoenix is PalmFlow's built-in AI assistant. Ask plain-language questions about your own inventory and sales, draft listing copy, and get pricing context.

    Phoenix is the AI assistant built into PalmFlow. Unlike a generic chatbot, it can answer questions about *your* data — your inventory, your sales, your margins.

    What Phoenix is good at

    • Answering questions about your business. "Which items have sat unsold the longest?" "What was my real margin last month after fees and shipping?" Instead of building a spreadsheet, you ask.
    • Drafting listing copy. Give it an item's brand, model, and condition and it drafts a title that uses keywords buyers actually search, plus a clean description.
    • Pricing context. It can help you reason about where to price something based on what you're seeing — a starting point, not a guarantee.

    How to use it

    1. 1Open Phoenix from the dashboard.
    2. 2Ask in plain language. Be specific — "show me items in *To ship* over their SLA" beats "what's up with my orders."
    3. 3Use the answer as a fast first pass, then act in PalmFlow.

    Honest limits

    AI drafts are a 90%-done first pass, not gospel. A model will sometimes confidently invent a model number or a measurement it can't see, so:

    • Check listing drafts before you publish — especially specifics like model numbers and measurements.
    • Treat pricing suggestions as a starting point, not a promise. No tool has a crystal ball for what an item will sell for.

    What Phoenix can look up

    Phoenix isn't guessing from your question — it queries your account. It can pull orders over a date range, filter your inventory, compute performance metrics, compare one period against another, break revenue down by platform, rank your top sellers, and find your longest-sitting stock.

    That's the boundary worth knowing: questions answerable from those are answered from your real data. Anything outside them — what an item will fetch next month, what a buyer is thinking — is the model reasoning, not your ledger.

    What counts against your allowance

    Phoenix is metered per month, and both kinds of use draw on the same pool: questions you ask the assistant, and AI-drafted listing titles and descriptions from the listing composer. Running low on one means running low on both, which is worth knowing before you hit it mid-listing.

    Allowances scale with your plan, every plan includes some, and annual billing doubles them. See Plans and billing and /pricing for current numbers.

    Everything here is free to try.