# Engagement letter drafting: fill confirmed facts, leave scope and fees to the lawyer > A proposed engagement letter workflow: approved templates, facts from confirmed intake, visible gaps, no generated fee terms and a recorded lawyer approval. Canonical URL: https://techtideai.io/blog/legal-engagement-letter-drafting Markdown URL: https://techtideai.io/blog/legal-engagement-letter-drafting.md Publisher: TechTide AI, Columbus, Ohio (https://techtideai.io) Author: Alex Cinovoj, Founder & CTO, TechTide AI ยท CTO & Co-Founder, CPG Playbook Type: Field Note Published: 2026-10-29 Collection: September 2026 editorial collection (a label, not the publication date) Reading time: 5 min Topics: engagement letter automation, law firm engagement letter, legal document drafting, ABA Model Rule 1.5 Site index: https://techtideai.io/llms.txt Offer summary: https://techtideai.io/offers.md --- Once intake is complete and conflicts are cleared, someone has to prepare the engagement letter. In many firms that means opening last year's letter for a similar matter, replacing names, adjusting the scope paragraph, and hoping nothing from the old client survives. It is careful work done under time pressure, and it is easy to get subtly wrong. Drafting help is reasonable here, with a clear boundary. The workflow can assemble the letter from the firm's approved templates and the facts already confirmed at intake. The scope of the representation, the fee terms and any departure from the template are decisions for the responsible lawyer, and the workflow should never invent them. This is a proposed workflow, not legal advice. Ethics rules and their interpretation vary by jurisdiction, and the firm's own policies define what an engagement letter must contain. ## Start from the rule the letter serves [ABA Model Rule 1.5](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_1_5_fees/) provides in paragraph (b) that the scope of the representation and the basis or rate of the fee and expenses for which the client will be responsible shall be communicated to the client, preferably in writing, before or within a reasonable time after commencing the representation, with an exception for a regularly represented client charged on the same basis or rate. Changes in the basis or rate must also be communicated. States adopt their own versions of the Model Rules, and firms often go further than the minimum. I would ask the firm's general counsel or risk partner to define which clauses are required for each matter type before any drafting is automated. That definition becomes the template configuration. ## Use approved templates and clause variants only The workflow should draft only from templates and clause variants the firm has approved. If a matter needs a clause that does not exist, the draft should show a clearly marked gap for the lawyer to write, not generated text that looks like an approved clause. Each clause variant should have an owner and a version. When the firm updates a standard clause, new drafts use the new version, and the record of each sent letter keeps the version it used. ## Fill matter facts from confirmed intake Client name, matter description, opposing parties, billing contact and responsible attorneys should come from the intake record after conflicts are cleared, not be re-extracted from emails. Every filled value should link back to its intake field, so a reviewer can see where it came from. Where intake is incomplete, the draft should leave the field visibly blank. A plausible guess about the client's legal entity name or the billing address is worse than a gap, because it may survive review. ## Draft the scope paragraph as a proposal The scope paragraph is where drafting help is most tempting and most sensitive. A model can propose a scope description from the intake notes, but it should be marked as a draft for the lawyer's revision, and the workflow should highlight phrases that commonly cause disputes, such as open-ended descriptions or references to related matters. I would also have the draft list what is excluded when the matter type's template calls for exclusions. Clear exclusions protect both client and firm, and they are a decision the lawyer should make consciously. ## Never generate fee terms Fee terms should come only from the firm's approved rate tables or from values the responsible lawyer enters. The workflow should not calculate a blended rate, propose a cap, or describe an alternative fee arrangement on its own. If the matter type allows several fee structures, the lawyer chooses one and enters the specific values. The review screen should show the fee section separately from the rest of the letter, with the source of each number. A reviewer confirming fee terms should not have to search the letter to find them. ## Route the draft to the responsible lawyer The draft goes to the responsible lawyer for review with a short summary: template used, clause variants, filled fields and their sources, gaps, and anything that departs from the standard template. The lawyer edits, approves, and sends it through the firm's normal process. The workflow should record who approved the final letter and when, and store the sent version with the matter. If the client negotiates changes, the revised version and its approval should be recorded the same way. ## Keep client information in approved systems Engagement letters contain client identities, matter descriptions and fee terms. The drafting step should run inside systems the firm has approved for client information, and any model service involved should be reviewed under the firm's confidentiality obligations before client information is sent to it. The ABA's guidance on generative AI, discussed in a separate note on vendor review, is a useful reference for that review. ## Measure accuracy before speed In a pilot, have the workflow draft letters for matters the firm is opening anyway, and compare the drafts with the letters the lawyers would have sent. Track wrong or missing facts, clauses the lawyer had to replace, and gaps that were correctly left blank. The acceptance condition I would agree: for the pilot matter types, every filled fact matches the confirmed intake record, no fee term appears that the lawyer did not enter or the rate table did not supply, and every letter sent has a recorded approval. Bring a few anonymized engagement letters, your template set and the intake fields you already capture. [Send the project brief](https://techtideai.io/contact). ## Sources - [ABA Model Rule 1.5](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_1_5_fees/) ## Related Field Notes - [Law firm intake automation: prepare the file, keep the decision with counsel](https://techtideai.io/blog/law-firm-intake-automation) ([Markdown](https://techtideai.io/blog/law-firm-intake-automation.md)): A proposed legal intake workflow with source-linked facts, conflict-review routing, confidential data boundaries, and human approval. - [Evaluating an AI tool for a law firm: the confidentiality questions to answer before a pilot](https://techtideai.io/blog/law-firm-ai-tool-confidentiality-review) ([Markdown](https://techtideai.io/blog/law-firm-ai-tool-confidentiality-review.md)): Confidentiality questions for a law firm AI pilot: training, retention, access, ethical walls, output checks, data rules and exit, grounded in ABA guidance. - [Legal matter closeout: reconcile the checklist before changing access](https://techtideai.io/blog/legal-matter-closeout-workflow) ([Markdown](https://techtideai.io/blog/legal-matter-closeout-workflow.md)): A proposed matter-closeout workflow that reconciles the checklist across systems and reviews access before anything changes, under the firm's own rules. - [Commercial credit applications: assemble the file, keep the decision with the credit team](https://techtideai.io/blog/commercial-credit-application-review) ([Markdown](https://techtideai.io/blog/commercial-credit-application-review.md)): A proposed business credit workflow: completeness checks, sourced financial figures, organized references, analyst-owned decisions and approved notices. All Field Notes: https://techtideai.io/blog