by TradeCRM (LX AI)
TradeCRM — A CRM built for one trade Use it as decision-support: demo mode works without a live key; live runs require configuration. No fabricated metrics, and no claims for SSO/CSV/Slack unless that surface is actually shipped.
Decision-support only — confirm before you publish or act.
I built TradeCRM after watching solo tradespeople juggle paper notes, generic CRM screens, and endless spreadsheets that never fit the rhythm of a single trade. The product is a lightweight decision‑support tool that presents a simple pipeline tuned to the steps you actually follow: lead, quote, job, follow‑up, invoice. It does not try to be a sales hub for every industry; it focuses on one trade at a time, letting you pick the trade when you first open the app and then showing only the fields that matter for that trade.
The core idea is to give you a ready‑made checklist that appears as you move a card from one column to the next. When you drag a lead to the quote column, the app shows a set of prompts that remind you to capture material costs, labor hours, and any permits needed. When you move a card to the job column, you see a list of tasks that are typical for that trade, such as rough‑in, finish, and cleanup. The prompts are static; they are not generated by AI unless you enable live mode and provide your own API key.
TradeCRM is meant for solo pros, small crews, and anyone who feels that a generic CRM adds more noise than signal. If you spend more time customizing fields than actually tracking work, this tool tries to cut that overhead. It works for electricians, plumbers, HVAC techs, carpenters, painters, and any other trade where the workflow follows a repeatable pattern of lead → quote → job → follow‑up → invoice.
The application has two runtime modes: demo and live. Demo mode runs entirely in the browser with a built‑in dataset that lets you explore the UI without any external keys. Live mode requires you to configure a language‑model endpoint and an API key; the model then generates the next‑action hints based on the data you have entered.
In demo mode the app loads a static JSON file that contains sample leads, jobs, and follow‑up notes for three example trades. You can create, edit, and move cards just like in live mode, but any changes disappear when you reload the page. This mode is useful for training, for showing a colleague how the pipeline looks, or for testing a new trade configuration before you commit to a key.
Live mode starts when you provide a valid API key and endpoint URL in the settings panel. The app then sends the current card title and description to the model and receives a short list of suggested next actions. The model does not store your data; it only returns text that you can copy into the card’s note field. If the key is missing or the endpoint returns an error, the app falls back to demo mode and shows a banner that explains the fallback.
Below is a concrete example of how an electrician might use TradeCRM from first contact to final invoice. The steps are written as they appear in the UI; you can replace the trade‑specific details with those that match your own specialty.
When a new request arrives — whether by phone, email, or a website form — you create a lead card. The lead card contains fields for contact name, phone, email, and a short description of the need (e.g., “panel upgrade in basement”). In demo mode the app pre‑fills a sample lead; in live mode you type the details yourself. After you save the lead, the card appears in the Leads column.
Moving the lead to the Quote column triggers a set of prompts that ask you to list materials, estimate labor hours, note any permit requirements, and add a margin. You fill in each prompt, and the app calculates a tentative total that you can edit. The quote total is shown on the card as a read‑only summary; you can copy it to your quoting template or send it directly from your email client.
When the customer accepts the quote, you drag the card to the Job column. The Job column shows a checklist of typical phases for an electrical job: rough‑in wiring, fixture installation, testing, and cleanup. You tick off each phase as you complete it. If a phase needs a sub‑task (for example, “run new conduit to garage”), you can add a free‑form note that stays attached to the card.
After you mark the Job column as complete, the card moves to the Follow‑up column. Here the app shows a short list of actions that often help close the loop: schedule a walk‑through, collect final payment, request a review, and schedule maintenance. In live mode the model may suggest a custom hint based on the notes you have added (e.g., “customer mentioned future lighting upgrade — offer a quote”). You accept or dismiss each hint, and any accepted hint becomes a new task card in the Follow‑up column.
When all follow‑up tasks are done, you move the card to the Done column. At that point you can generate an invoice from the quote total, apply any change‑order adjustments, and mark the invoice as paid. The card stays in Done as a reference; you can archive it later if you need to keep the board tidy.
| Pipeline stage | Typical fields shown | Common actions |
|---|---|---|
| Lead | Contact info, request description | Create lead, assign source |
| Quote | Materials list, labor hours, permits, margin | Fill prompts, calculate total, send quote |
| Job | Phase checklist, free‑form notes | Track progress, add sub‑tasks, log time |
| Follow‑up | Walk‑through, payment, review, maintenance | Schedule, collect, request, plan next service |
| Done | Invoice total, payment status, notes | Generate invoice, mark paid, archive |
TradeCRM stores all user‑entered data in a folder named .data that lives next to the executable or, when run in a browser, in the browser’s IndexedDB under the same name. The folder contains a single JSON file per board, with each card represented as an object that includes its column, title, description, and any custom fields you have filled. No data is sent to an external server unless you have configured live mode and provided an API key; in that case only the text you explicitly send to the model leaves the machine, and the model’s response is returned to you for local copying.
Because the data stays inside the .data boundary, you can back it up by copying the folder to a USB drive or to a cloud sync service of your choice. The format is plain JSON, so you can also import it into another instance of TradeCRM if you move to a new computer. The app does not automatically sync across devices; you must manually copy the folder if you want the same board on multiple machines.
In demo mode the .data folder is populated with a read‑only sample set when the page loads. Any edits you make are written to a temporary overlay that is discarded on page reload. This means you can experiment freely without worrying about leaving test data behind. If you want to keep a demo session, you must switch to live mode and provide your own key before making changes.
To export a board, open the Settings pane and choose Export Board. The app downloads a file named tradecrm-board.json that contains the full board state. To import, choose Import Board and select a previously exported file. The import will replace the current board; a warning appears if the current board has unsaved changes.
I want to be clear about the limits of TradeCRM so you can decide if it fits your workflow. First, TradeCRM does not promise any increase in revenue, job speed, or customer satisfaction. The output is strictly decision‑support; you must review every suggestion before you act on it. Second, the tool does not include built‑in SSO, Slack notifications, or bulk CSV import/export unless those surfaces have been shipped in this specific build. If you see mentions of those features elsewhere, treat them as placeholders, not guarantees. Third, the AI model used in live mode is not fine‑tuned on your proprietary data; it is a general‑purpose language model that you point to with your own key. For guidance on responsible model use you can review the AI risk management framework from NIST (https://www.nist.gov/itl/ai-risk-management-framework) and the LLM‑focused Top 10 from OWASP (https://owasp.org/www-project-top-10-for-large-language-model-applications). Finally, TradeCRM is not a replacement for a full‑featured CRM that manages marketing campaigns, territory planning, or multi‑product catalogs. It is a simple pipeline helper built for one trade at a time.
A CRM built for one trade
Operators and builders who need a fast first draft or checklist from TradeCRM.
Yes in explicit Demo mode. Live AI requires a configured key.
No. Outputs are decision-support; you still review before publishing or acting.
Only if those features are implemented in this product build — do not assume them from marketing copy.
Runs may be stored locally under the product's .data/ boundary; treat demos as ephemeral.