A mid-sized personal injury firm in our network closes 180 settlement demands a year. Each letter takes an associate 90 minutes to draft: pull medical records, summarize treatment, calculate special damages, write the liability narrative, format exhibits, cite case law. That’s 270 hours of associate time at $300 an hour, or $81,000 in direct labor before you count the partner review pass.

The firm’s intake coordinator told me the real problem isn’t the cost. It’s the three-day lag between signing the client and getting the first demand out. Adjusters move fast. If your demand lands a week after the other firm’s, you’re negotiating from behind.

Most practices solve this by hiring another associate or leaning on paralegals who aren’t trained to write persuasive liability arguments. You get faster output but weaker positioning. The alternative is what we build at Enterprise DNA: an agent that drafts the demand letter from your case file, maintains your firm’s tone and citation style, and hands a clean Word document to the supervising attorney in under ten minutes.

This isn’t a template generator. It’s a system that reads your intake notes, pulls relevant medical chronology, applies your damages formula, and writes a narrative that sounds like your best associate wrote it. The partner still reviews and signs. The client still gets a human lawyer. You just stop burning billable hours on first-draft drudgery.

Why Demand Letters Eat So Much Time

Walk through what happens today when a new PI case converts. The client signs the retainer. Intake uploads the police report, photos, initial medical records, and insurance details into your case management system. Someone assigns the file to an associate.

That associate opens a blank document and starts from scratch. They read the police report to build the liability section. They review medical records to summarize treatment and prognosis. They pull prior settlement ranges from the firm’s internal database or their own memory. They calculate specials: past medical, future medical, lost wages, property damage. They draft a demand figure and write two pages explaining why it’s justified.

Then they format exhibits, add a signature block, and send it to the partner for review. The partner reads it, rewrites the liability argument, adjusts the demand figure, and sends it back for edits. Another hour of associate time. Another 20 minutes of partner time. Three days have passed since the client signed.

The bottleneck isn’t legal skill. It’s the manual assembly of information that already exists in your case file. The associate isn’t doing legal analysis for 80% of that 90 minutes. They’re copying dates from a medical record, calculating a sum, and reformatting paragraphs to match the firm’s style.

That’s exactly the kind of work an AI agent handles well. It’s structured, repetitive, and rule-based. The creative part (deciding whether to push hard on pain and suffering or whether the liability narrative needs a different angle) still belongs to the attorney. But the agent can hand them a draft that’s 85% done instead of a blank page.

What a Demand Letter Agent Actually Does

We build these agents on Omni Ops, the operational AI layer that connects to your case management system, document storage, and templates. The agent doesn’t replace your workflow. It plugs into the step where the associate opens a blank Word doc.

Here’s the sequence. A partner marks a case ready for demand in your CMS. The agent receives a trigger. It pulls the case file: intake notes, police report, medical records, insurance policy limits, and any prior correspondence. It reads through the medical chronology and extracts treatment dates, provider names, diagnoses, and total billed charges. It identifies the at-fault party from the police report and pulls the relevant liability standard from your firm’s template library.

Then it writes the demand letter. Liability section first, using the facts from the report and the legal standard your firm applies to that case type. Medical summary next, organized chronologically with emphasis on the treatments your firm typically highlights (ER visit, imaging, physical therapy, ongoing pain management). Damages calculation follows: special damages as a line-item table, general damages with a multiplier your firm uses, and a total demand figure.

The agent applies your firm’s tone. If your letters open with a two-sentence summary of the accident, the agent does that. If you always cite a specific case when arguing comparative negligence, the agent drops that citation in the right spot. If your damages section leads with a paragraph about the client’s age and occupation before listing medical bills, the agent structures it that way.

It outputs a Word document with your firm’s letterhead, the supervising attorney’s signature block, and exhibits attached as appendices. The whole process takes eight minutes. The document lands in the partner’s review queue with a note: “Draft demand ready, case #4782.”

The partner opens it, reads the liability argument, checks the damages math, adjusts the demand figure up or down based on their read of the adjuster, and approves it. Total review time: 15 minutes. The demand goes out the same day the case was marked ready.

You’ve collapsed 90 minutes of associate work and 20 minutes of partner rework into 15 minutes of partner review. The client gets a stronger letter faster. The associate who used to draft demands now handles discovery or depositions. The firm’s average time-to-demand drops from three days to four hours.

Keeping the Attorney in the Loop

The most common objection we hear is that demand letters require judgment. They do. That’s why the agent doesn’t send the letter. It drafts it and routes it to the attorney who owns the case.

The agent isn’t deciding whether to demand $150,000 or $200,000. It’s pulling your firm’s damages formula (typically a multiple of specials based on injury severity and liability strength) and proposing a figure. The attorney reviews that figure, considers the adjuster’s history, the client’s presentation, and the venue, and changes it if needed.

The agent isn’t deciding whether to argue pure comparative negligence or joint and several liability. It’s applying the legal standard your firm uses for rear-end collision cases in your jurisdiction. If the facts are unusual, the attorney rewrites that section.

The agent isn’t deciding which medical treatments to emphasize. It’s following your firm’s pattern: lead with emergency care, highlight imaging that shows objective injury, summarize physical therapy duration, note any ongoing prescriptions. If the client’s treatment was atypical, the attorney adjusts the narrative.

What the agent eliminates is the blank-page problem. Your associate doesn’t spend an hour assembling facts and formatting paragraphs before they can start thinking about strategy. They start with a complete draft and spend their time on the 15% that requires legal judgment.

One partner in our network describes it as “having a junior associate who’s already done the research and written the first draft, but who never bills and never complains when you rewrite the whole thing.” The agent doesn’t get tired. It doesn’t miss a medical record buried on page 47 of the file. It doesn’t forget to include the property damage estimate.

And because the agent learns from your edits, it gets better over time. If you consistently rewrite the pain-and-suffering paragraph to lead with the client’s testimony instead of the medical summary, the agent starts drafting it that way. If you always add a paragraph about the defendant’s prior violations when the police report shows a history, the agent picks up that pattern and includes it in future drafts.

Extending the Agent to Settlement Proposals

Demand letters are the first use case most firms automate, but the same agent handles settlement proposals when the other side makes an offer. The structure is similar: summarize the case, explain why their offer is inadequate, counter with a figure and justification.

The agent pulls the original demand, the adjuster’s offer letter, and any correspondence in between. It drafts a response that acknowledges their offer, restates your damages calculation, addresses any new arguments they raised (like comparative fault or pre-existing injury), and proposes a counter.

If the adjuster offered $50,000 and your demand was $180,000, the agent doesn’t just split the difference. It applies your firm’s negotiation strategy. Maybe you counter at $140,000 and add a paragraph explaining why the client’s testimony will be compelling at trial. Maybe you counter at $120,000 and emphasize the cost of defense if the case proceeds. The agent follows the playbook your firm already uses.

The partner reviews the draft, adjusts the tone (maybe this adjuster responds better to aggressive positioning, maybe they prefer collaborative language), and approves it. The response goes out the same day the offer arrived instead of sitting in the associate’s queue for 48 hours.

This speed matters in settlement negotiations. Adjusters are managing 80 files. If your response takes three days, they’ve moved on to other cases. If your response lands in two hours, you stay top of mind and signal that your firm is organized and ready to push.

The Economics of Automating Drafting Work

Let’s use the numbers from the firm I mentioned at the top. 180 demand letters a year, 90 minutes per letter, $300 per hour for associate time. That’s $81,000 in direct labor. Add another 20 minutes of partner review at $500 per hour for each letter, and you’re at $111,000 total.

With the agent, you eliminate the 90 minutes of associate drafting and reduce the partner’s time from 20 minutes of review-plus-rewrite to 15 minutes of review-only. You’re now spending 15 minutes of partner time per letter, or $2,250 per year. You’ve freed up 270 hours of associate capacity.

That associate can now handle 12 additional cases through discovery, bill an extra $80,000 in fees, or take on the intake work that used to get delegated to a paralegal. The capacity gain is more valuable than the cost savings.

The secondary benefit is speed. Faster demand letters mean faster settlements. If your average case settles 30 days sooner because you’re getting demands out the same day instead of three days later, your cash conversion cycle tightens. You’re collecting fees a month earlier on every case. For a firm doing $3 million in annual revenue, that’s a $250,000 improvement in working capital.

The third benefit is consistency. Every demand letter follows your firm’s best structure. New associates don’t spend six months learning how to write a good demand. The agent drafts it correctly from day one, and the associate learns by editing a strong draft instead of staring at a blank page.

If you want a practical framework for where else AI fits into your client intake and case workflow, we’ve built a checklist that walks through the eight highest-ROI automation points for law firms. You can grab it here: AI Client Intake Checklist for Law Firms. It’s a one-page worksheet you can use to map your current process and identify which steps an agent can handle.

What the Build Looks Like

We don’t deploy a generic tool and hope it fits. Every firm’s demand letter style is different. Some open with a summary of damages and then explain liability. Others lead with a narrative of the accident and build to the damages at the end. Some cite case law heavily. Others rely on factual persuasion and save legal argument for litigation.

The build starts with a 60-minute Omni Audit. We sit with a partner and an associate, walk through two or three recent demand letters, and map the structure your firm uses. We identify the data sources: which fields in your case management system, which documents in your file storage, which templates in your shared drive. We ask about edge cases: how do you handle wrongful death demands differently from soft-tissue injury demands? Do you adjust tone based on the insurance carrier?

Then we build the agent in Omni Ops. We connect it to your CMS and document storage. We train it on your template library and five to ten example letters that represent your best work. We configure the damages formula, the citation style, the exhibit format. We set up the review queue so drafts land in the right partner’s inbox with the case file attached.

We test it on three closed cases. You compare the agent’s draft to the letter your associate actually wrote. We adjust tone, structure, and emphasis based on your feedback. Then we go live on new cases.

The first ten letters get extra review. You’re checking that the agent is pulling the right facts, applying the right legal standard, and matching your firm’s voice. After that, it’s just part of the workflow. Case gets marked ready, draft appears in your queue, you review and approve.

The whole process from audit to production takes three weeks. You can see what the audit covers and how we scope the build at the AI audit for law firms.

Handling Complex or High-Value Cases

The agent works best on cases that follow a pattern. Rear-end collisions with soft-tissue injury and $40,000 in medical bills. Slip-and-fall with a fracture and clear liability. T-bone intersection crash with moderate property damage and three months of physical therapy.

These cases make up 70% of most PI firms’ volume. The liability facts are straightforward, the damages calculation is formulaic, and the demand letter structure is consistent. The agent drafts these letters at the same quality as your mid-level associate, and it does it in eight minutes instead of 90.

For complex cases (wrongful death, catastrophic injury, disputed liability, multiple defendants), the agent still saves time but requires more attorney input. It can pull the medical chronology and calculate specials, but the liability narrative and damages justification need a human to craft the argument.

In those cases, the agent acts as a research assistant. It summarizes the medical records, flags the key treatment milestones, calculates the economic damages, and pulls relevant case law. The attorney writes the demand from scratch but starts with a ten-page brief instead of a raw case file.

You decide where to draw the line. Some firms use the agent for all cases under $250,000 in demand value and handle everything above that manually. Others use it for all cases but add a senior partner review step for demands over $500,000. The agent adapts to your workflow, not the other way around.

Connecting Demand Drafting to the Rest of Your Workflow

Demand letter automation is one piece of a broader operational system. The same agent that drafts demands can draft discovery responses, summarize depositions, and generate status reports for clients.

If you’ve already built an Intake Voice Agent (one of the other agents we deploy on Omni), it’s capturing case facts during the initial call and populating your CMS with structured data. That data flows directly into the demand drafting agent. The client describes the accident to the voice agent, and two weeks later the demand letter includes that description verbatim in the liability section.

If you’ve built a Document Review Agent to handle medical record review, it’s already extracting treatment dates, diagnoses, and billed charges. The demand drafting agent pulls that summary instead of reading 200 pages of records from scratch.

These agents stack. Each one makes the next one more valuable. The intake agent feeds the document review agent. The document review agent feeds the demand drafting agent. The demand drafting agent feeds the settlement negotiation agent. You’re building a system where information flows through your firm without manual handoffs.

We walk through this system design during the Omni Audit. You don’t have to automate everything at once. Most firms start with one agent (often intake or demand drafting), prove the ROI, and then add the next one. But it helps to see the full map so you’re building toward a coherent system instead of a pile of disconnected tools.

You can explore more of how these agents connect by visiting our guides section or reading through case studies and insights from other firms that have deployed similar workflows.

Why This Matters for Firms Under $10 Million in Revenue

If you’re running a firm doing $2 million to $8 million a year, you probably have three to eight attorneys and a handful of paralegals and admin staff. You don’t have the budget to hire another associate every time volume increases, and you don’t have the leverage to absorb inefficiency.

Demand letter automation is one of the highest-ROI places to start with AI because the work is repetitive, the output is measurable, and the time savings are immediate. You’re not asking your team to change how they practice law. You’re giving them a tool that eliminates the part of the job that feels like data entry.

The firms we work with typically see 200 to 400 hours of associate capacity freed up in the first year. That’s half an FTE. You can reinvest that capacity into case work, business development, or client communication. You can take on more cases without hiring. You can improve your settlement timelines without working weekends.

And because the agent learns your firm’s style, it becomes a training tool for new hires. A first-year associate who’s never written a demand letter can review the agent’s draft, see how your firm structures the argument, and learn the pattern in a week instead of six months.

The cost to build and run the agent is a fraction of what you’d pay to hire another associate. Most firms recover the investment in the first 90 days just from the time savings. After that, it’s pure capacity gain.

If you want to see what this looks like for your firm specifically, book a 60-minute Omni Audit. We’ll map your current demand letter process, identify where an agent fits, and show you the economics. You’ll walk out with a one-page build plan, a capacity model, and a cost estimate. No deck, no sales pitch, just the numbers.

What Happens After You Deploy the Agent

The first month is a learning period. You’re reviewing every draft the agent produces and making edits. The agent logs those edits and adjusts its output. By week three, you’re approving 80% of drafts with only minor changes. By month two, you’re approving 90% with no changes at all.

Your associate who used to draft demands moves to other work. Maybe they take on more discovery. Maybe they start handling settlement negotiations directly instead of just drafting the letters. Maybe they focus on client communication and case strategy. The point is they’re doing work that requires a law degree instead of work that requires a good memory and attention to detail.

Your average time-to-demand drops. Cases that used to sit in the queue for three days now get a demand letter out in four hours. Adjusters start responding faster because your firm is consistently first to the table.

Your settlement timelines compress. Faster demands mean faster offers mean faster counters mean faster resolution. The average case that used to take nine months from signing to settlement now takes seven months. Your cash conversion improves. Your clients are happier because they’re getting paid sooner.

And your partners stop spending 20 minutes rewriting liability arguments that should have been right the first time. They spend 15 minutes reviewing a draft that’s already 90% correct, make a few strategic adjustments, and approve it. The work gets done faster and the output is more consistent.

That’s the compounding effect of automation. You don’t just save time on one task. You improve the entire workflow downstream from that task. Faster demands lead to faster settlements lead to faster fee collection lead to better cash flow lead to more capacity to take on new cases.

The Practical Next Step

If you’re reading this and thinking “we waste a ton of time on demand letters but I don’t know where to start,” the next step is simple. Book a 60-minute Omni Audit. We’ll sit with you and one of your associates, walk through two recent demand letters, map your current process, and show you exactly where an agent fits.

You’ll get three outputs: a one-page workflow map showing which steps the agent handles and which steps stay with your attorneys, a capacity model showing how many hours you’ll free up in the first year, and a cost estimate for the build. No deck, no follow-up calls, just the information you need to decide whether this makes sense for your firm.

Most firms know within the first 20 minutes of the audit whether this is worth doing. If your demand letters follow a consistent structure and your associates are spending 60 to 90 minutes per letter, the ROI is obvious. If your letters are highly variable or your volume is low, we’ll tell you that too.

You can see more detail on what the audit covers and book a time here: book my Omni Audit. We run these audits every week for firms across personal injury, employment law, family law, and commercial litigation. The process is the same regardless of practice area.

The firms that move fastest on this are the ones that are already feeling the capacity constraint. You’re turning down cases because you don’t have the bandwidth. You’re hiring another associate but you know it’ll take six months before they’re productive. You’re working evenings and weekends to keep up with demand letter volume.

If that’s you, this is the fastest way to buy back 200 hours of capacity without hiring. And if you’re not sure whether your firm is a fit, the audit will answer that question in the first 15 minutes. Either way, you’ll know more about how your firm operates and where the bottlenecks are. That’s worth an hour.