The short answer on paralegal versus automation cost

A good paralegal is not a software subscription with a pulse. They bring judgment, client awareness, process knowledge, and accountability. If your firm needs someone to manage case files, support attorneys, communicate with clients, prepare filings, and handle exceptions all day, hiring may be the right call.

But many law firms hire a paralegal to solve a narrower issue. The phones aren’t answered quickly enough. Intake forms pile up. Conflict checks are inconsistent. Associates lose days to first-pass document review. Matter updates live in email threads. Partners are doing work that should never reach their desk.

In that situation, the right comparison isn’t simply salary versus software price.

The real question is this: how much repeatable work can automation remove before you need another full-time hire, and how much revenue or capacity does that create?

For firms with 2 to 20 attorneys, legal automation often makes financial sense when the work is high-volume, rules-based, and time-sensitive. Client intake, matter routing, follow-ups, document classification, and first-pass review are common examples. These workflows don’t need an AI system to replace legal judgment. They need it to handle the first mile reliably so your people can handle the parts that require expertise.

At Enterprise DNA, we usually see a leakage band of roughly $80,000 to $250,000 each year in firms of this size. That isn’t one hidden expense. It’s the combined cost of missed calls, unbilled admin, slow response times, associate hours spent on repetitive review, and staff capacity absorbed by manual handoffs.

The useful comparison is not paralegal or automation. It is where each belongs in your operating model.

You can see Omni for law firms to understand how we map that work before recommending a build.

What a paralegal actually costs the firm

A paralegal’s base salary is only the starting point. Depending on market, seniority, practice area, and location, a full-time paralegal may carry a base salary in a broad range of $55,000 to $95,000 per year. In specialist practice areas or higher-cost markets, the number can go well beyond that.

Then add the costs that rarely appear in a simple headcount comparison:

• Payroll taxes, health benefits, retirement contributions, and insurance
• Recruiting fees or the partner time spent interviewing
• Onboarding, training, and supervision
• Software seats, equipment, and workspace
• Paid leave and capacity gaps during absences
• Turnover risk, which can be expensive in legal support roles

For a planning model, many owners use a loaded employment cost that is 20 to 35 percent above base salary. A $70,000 paralegal can become an $85,000 to $95,000 annual commitment before you account for the work required to get that person fully effective in your firm.

That cost can be justified quickly. A strong paralegal who keeps matters moving, prevents missed deadlines, and frees attorneys for client work can produce a clear return.

The issue comes when the role is expected to be a switchboard, intake coordinator, file clerk, document reviewer, client chaser, calendar manager, and process owner at the same time. That usually creates a bottleneck, not leverage.

A person can only respond to one interruption at a time. A new employee also takes time to learn your matter types, client qualification rules, conflict-check process, partner preferences, and case management system. If intake demand spikes at 7:30 pm on a Thursday, a new hire doesn’t solve that problem unless they are working then.

Automation changes the economics for specific tasks because it can operate outside business hours, apply the same first-step process every time, and handle volume without adding another salary.

The legal work where automation has the clearest ROI

Automation is a poor fit for work that depends on nuanced legal strategy, relationship repair, negotiation, or final legal advice. Those are human responsibilities.

It is a strong fit for repeatable operational work that already follows a pattern. For most firms, the first opportunities sit in intake and document handling.

Client intake and follow-up

Many firms still have a version of this workflow:

1. A potential client calls after hours or submits a form.
2. The inquiry lands in voicemail, an inbox, or a generic CRM queue.
3. A receptionist or paralegal sees it the next morning.
4. They gather missing information and try to reach the lead.
5. A lawyer or partner reviews the details.
6. The firm decides if it is a fit and schedules a consultation.

That process can work with a low number of inquiries. It breaks down when calls arrive after hours, staff are busy with active matters, or the intake information is incomplete.

The Intake Voice Agent changes the front end. It answers calls after hours, during lunch, and on weekends. It captures the caller’s details, asks your approved intake questions, runs an initial conflict-check workflow, and books a consultation into the firm’s calendar when the lead fits your rules.

It does not give legal advice. It does not decide a conflict is cleared without your process. It captures the information consistently and escalates exceptions to a human.

The Matter Triage Agent handles the digital counterpart. It reviews incoming forms and emails, classifies the likely practice area, scores fit against the criteria you set, and routes the matter to the right partner or intake owner. The recipient gets a one-paragraph brief rather than an unstructured message chain.

This is where the cost of delay becomes visible. Firms commonly report that 30 to 40 percent of after-hours inquiries never convert when there is no fast response process. Not every inquiry is a good client, and not every missed inquiry was a viable matter. Still, a single retained matter can cover months of operational automation spend.

If your team is looking at Omni Voice as a way to protect inbound demand, focus first on call handling rules, calendar access, qualification questions, escalation paths, and conflict-check safeguards. The technology is the small part. The operating design is the real work.

First-pass document review

Document review is another area where firms often use expensive legal labour for repetitive early-stage work.

A junior associate might spend hours identifying clauses in a contract set, sorting discovery documents, comparing versions, or creating an initial factual timeline. Their time can cost the firm $200 to $400 per hour in billable value, depending on the practice and market. Sometimes the client will pay for it. Often, a portion is written off because the time is too administrative, too slow, or beyond what the client will accept.

The Document Review Agent can perform the first pass on contracts, discovery batches, and matter files. It flags clauses and topics based on your review framework, summarises each party’s position, identifies missing information, and produces an associate-grade memo for human review.

That last point matters. The output is not a final legal conclusion. An attorney or qualified reviewer remains responsible for accuracy, privilege, legal analysis, and the final work product. The gain is that your team starts with a structured review package instead of a blank page and a large file set.

For an overview of how these operational agents fit together, look at Omni Ops. It is designed around the back-office and matter-flow work that consumes time without improving legal judgment.

A practical breakeven model for 2 to 20 attorneys

The right breakeven model starts with workload, not vendor pricing.

Take a 5-attorney practice with two paralegals and a steady intake flow. Each attorney loses an average of four hours per week to matter administration, status chasing, initial document sorting, and incomplete intake handoffs. That is 20 attorney hours every week.

Not all of those hours can be recovered. Some are necessary client-facing work. Assume automation removes or redirects only 35 percent of that time. That is seven hours per week returned to the attorney team, or about 350 hours a year.

If even half of those hours become billable or allow the firm to serve more matters, the return can be meaningful. At a conservative blended value of $250 per attorney hour, 175 recovered hours represent $43,750 in annual capacity. Add two or three retained matters from faster intake response and the economics can shift quickly.

Now compare that with hiring.

A new paralegal at a loaded annual cost of $90,000 needs to create or protect at least that much value before they are a net gain. That is entirely achievable in a busy firm. But if the reason for hiring is primarily intake coverage, email classification, status updates, and first-pass document work, an automation layer may reduce the required scope of the role.

Here is a useful planning lens:

Firm size	Common pressure point	When automation often makes sense	When hiring still makes sense
2 to 4 attorneys	Missed calls, partner-led intake, scattered admin	Before adding a full-time intake coordinator, especially with after-hours calls	When active matters need daily client contact and filing support
5 to 10 attorneys	Paralegals overloaded by intake and file movement	When the firm has repeatable intake, routing, and review tasks across practice areas	When matter complexity requires hands-on coordination across courts, clients, and counsel
11 to 20 attorneys	Volume, inconsistent process, associate review bottlenecks	When each team is doing the same administrative work differently	When automation has already stabilised the workflow and caseload justifies dedicated human capacity

A small firm may not need to replace any role. The more immediate outcome is avoiding a premature hire while improving client response. A larger firm may keep the same staff count but redeploy paralegals toward client communication, deadline management, and complex case support.

That is a better result than asking a paralegal to race an inbox all day.

If you want the numbers grounded in your own workload, Book a 60-min Omni Audit. We will work through the workflow, time burden, and likely financial range rather than hand you a generic automation estimate.

How to calculate your own breakeven point

Use a simple five-step calculation. You don’t need perfect data. Directionally correct figures are enough to identify where to investigate.

1. Measure the repeatable workload

For two weeks, track the volume of:

• Inbound calls and form submissions
• Calls received outside staffed hours
• Follow-up attempts before a consult is booked
• Documents or discovery batches requiring first-pass review
• Hours attorneys spend on non-billable matter administration
• Hours paralegals spend copying, routing, chasing, and updating information

Don’t ask people to fill out a detailed timesheet for every minute. Give them a short daily estimate. The goal is to find recurring work, not create another administrative burden.

2. Price the current method

Calculate the loaded cost of the employee doing the work. For attorneys, calculate two figures: their direct cost and the billable capacity they could reasonably create if the work moved elsewhere.

A partner spending three unbilled hours each week on intake follow-up is not just an efficiency problem. It is a leadership capacity problem. That time often displaces business development, client strategy, supervision, or billable work.

3. Estimate the recoverable percentage

Don’t assume 100 percent of a process can be automated. A realistic first estimate is often 20 to 50 percent for a defined workflow.

For example, the Intake Voice Agent can capture information, classify urgency, and schedule an appropriate consultation. It cannot replace the partner’s judgment about taking on a complex matter. The Document Review Agent can create a review memo. It cannot take responsibility for the legal position.

Use the lower end of the range if your processes are undocumented or vary heavily between attorneys.

4. Include recovered revenue and avoided cost

Your return may come from four places:

• Fewer qualified leads lost to slow response
• More attorney hours directed to client work
• Less associate time written off
• A delayed or narrower hiring decision

Don’t count all four at their maximum. That creates a business case nobody trusts. Choose the two effects most likely to occur in your firm and use conservative numbers.

5. Compare against implementation and ongoing spend

Automation has its own cost. There is workflow design, integration work, testing, staff training, governance, and ongoing support. A low subscription price without proper implementation is often false economy.

The key test is straightforward. If the annual cost of the automation program is lower than the recoverable value of the workload, with a sensible buffer for adoption, it deserves a serious look. If the work is too varied, too low-volume, or too dependent on human judgment, hire or improve the existing process instead.

Don’t automate a broken intake process

AI can make a clear process faster. It can also make confusion faster.

Before deploying an agent, agree on the firm rules it will follow. Define what counts as a qualified inquiry, who owns each practice area, what information is required before booking, and how potential conflicts are escalated. Set clear boundaries around legal advice, confidentiality, data retention, and human review.

This is especially important in law. You are not just routing sales leads. You are handling confidential information, potential conflicts, time-sensitive matters, and people who may be distressed or under pressure.

Our resources and guides can help your team think through adoption, but the best starting point is a process map of what happens from first contact through consultation and matter opening.

For intake specifically, download the AI Client Intake Checklist for Law Firms. It is a practical worksheet for defining your call questions, response rules, routing criteria, and escalation points before you ask software or staff to carry the process.

You can also access the direct checklist download if you want to work through it with your intake lead.