Convergence Check
How your product work could move EBITDA
Work with your AI agent to turn your backlog, roadmap and strategy documents into product data Lorenz can read. See how your roadmap, backlog and planning decisions support your business priorities. Understand what’s working, where the gaps are, what to change, and what that could mean for your organization’s finances.
See an example reportStep 1 Choose how to generate your product data
Step 2 Copy the prompt into your agent session
How your product work could move EBITDA. Find out with the free Lorenz Convergence Check. Read https://lorenz.converge.studio/assess/llms.txt first and follow its steps in order. Your first message to me is the opening paragraph llms.txt gives, word for word; then tell me the path: Share → Check → Priorities → Clarify → Report. About two minutes. You get a report preview link after the read-back that fills in as you answer, and the final link at the end; the check is not finished until you have the final link. The recognise reply, the dry grade's reply and the stored grade's reply each carry `say` and `progress`. Relay `say` to the person word for word before anything else. Before each ask, name the step from `progress` as "Step N of 5 · Name", with the name from `progress.title` (the dry reply's reads "Step 4 of 5 · Clarify Your Approach"). When a step starts, say its purpose once, after its step line, as llms.txt lists them. In Clarify, number the questions as "Question 2 of N", where N is the dry reply's `question_count`: the one count, the same number the page shows and `progress.questions_left` carries. It does not change as you ask. The read-back before the questions is not a question. When the dry reply's `financial_question` is true, the last of the N is the optional financial question, counted in N like the others. 1. Gather the six inputs for ONE planning period, from whatever tools you can reach (up to 8 plain-text files: CSV, Markdown, JSON or text; 256 KB and 500 rows in all): - Required: Backlog (with descriptions and whatever you track); Plan (current roadmap, with supporting details); Strategy (mission, goals, priorities — and anything else that says where you're going). - Helpful for a full review: Business cases (or financial targets); Customer evidence (research, requests, usage); Capacity (people, points or a budget); Support and incidents (tickets and outages); Risks (known risks and promises); Outcome reports (results of past work). If something is not written down, do not invent it or hunt for it: tell me, and I will answer the check's questions instead. If I have no business case or figures written down, you may help me build one from what I know; tell me which figures are mine and which you worked out. 2. Show me each file and the privacy note before sending anything: what is in it and what you left out, and offer to mask names, sensitive terms and specific metrics. Then tell me what the next step is and why you need to send the files, and wait for my OK; when you ask for it, warn me that my app may ask me to allow each web request. 3. Read the columns and confirm with me: tell me how each file was read and show me only the columns the check was unsure of, in plain words (every column if I ask). Wait for my answer. Then, in a new message, say "Step 3 of 5 · Confirm Priorities": confirm with me the priorities list found in any strategy document and ask me the planning unit I work in, or, if I already confirmed the list, say "Step 3 of 5 · Confirm Priorities — completed from the priorities you already confirmed." 4. Read back what was found and ask me the questions one at a time, only the ones the check returns, in the order it gives, with the answers numbered so I can reply with a number. Only my own words count. When the check says so, the last question is my financial figures (annual revenue, and EBITDA or EBITDA margin): optional, counted with the others, and never sent. 5. Grade and store the report, and give me the link. The stored grade's reply carries `done: true`, `report_url`, `say` and `wrap_up`: relay the sentence, then give the person the link right after it. Give ONE link: `report_url` with the person's financial figures or bands after `#` when they gave them (step 4 says how), else the plain `report_url`. The check is finished only then; until the person has the link, it is not. 6. Present the result the way llms.txt describes.
Scroll to read the full prompt
The files you share are not stored. Only the neutral input read from them, the column mapping you confirm and the report are kept, for 30 days unless you leave an email, and none of it is ever used to train a model. To read them, the files are sent to TypeSafe, whose model answers the assessment's questions about them. TypeSafe is the only outside service that reads your files, and Lorenz does not keep the files themselves. The report quotes the answers you give and any passage you point at, up to 500 characters each. A fingerprint of what the model was asked, and its answers — not the text itself — is kept for 30 days so that checking the same files again reads the same answers. Anyone with the report’s link can open it.