In September 2026, two rooms of Social Security disability practitioners joined Disability Peers in Practice, Chronicle’s monthly peer community where SSD practitioners work through real operational problems together. The topic: what you measure, what you have given up trying to measure, and what changed once you could.
The rooms were small enough that solos and operations managers sat in the same conversation, which turned out to be the point. A true solo with about 60 cases and no staff, a solo attorney carrying 800 cases with three legal assistants, and a 932-case disability department inside a large personal injury firm answered the same questions, and the answers lined up more often than anyone expected.
Twelve practitioners spoke across the two sessions, from Hawaii, California, Oklahoma, Washington, Florida, Pennsylvania, Louisiana, North Carolina, West Virginia, and Illinois. Roughly half run a practice under 300 cases; the rest run departments of several hundred to over a thousand. What they had in common was that nobody claimed to have the measurement problem solved, including the people who measure the most.
The Number Everyone Watches Is Cases by Stage
- Every firm, at every size, looks first at how many cases sit at initial, recon, and hearing
- Below about 100 cases, that count plus the bank balance is the whole dashboard, and it is a choice
- Watch the spread week over week, not the total; the trend is where the bottleneck shows up
- Solos who deliberately do not track intake said so without apology
Asked what number they look at most, nearly everyone described the same view. The difference by firm size was not which number, but whether anything sits on top of it.
I look daily at how many at initial, how many at recon, how many at hearing level without a hearing scheduled, how many at hearing level with a hearing scheduled. Because again, I have no plans of hiring another attorney. I have one full-time staff member. But I really don’t want to scale. That’s not what I’m looking to do in this life.
— Stephanie
The solos in the room were consistent that intake tracking is not worth their time when the phone rings enough. One predicts income from two numbers: win rate and hearings per month.
If my win rate is X percent, and I’m doing 3, 4, 5 hearings a month, plus you’re going to get a couple of initials or recons, to me, I can pretty well predict what my income is going to be over time.
— Joe
The refinement worth stealing came from a 320-case firm that looks at the same stage counts weekly as a spread, watching for where cases back up. That firm is at a record case count with hearings backing up faster than the total, which is the kind of thing a single number never shows.
What this means: the stage count is universal because it answers the two questions every firm has, is enough coming in and where is it stuck. Track the trend, not the total.
Nobody Has a Formula for Case-Manager Capacity, and the Range Was 150 to 300
- Two practitioners asked the rooms directly for a number; nobody had a derived one
- Stated baselines ran from 150 cases per case manager to about 300, plus 250 to 300 per records retriever
- The spread is explained by what sits underneath: whether the case manager pulls records, and how work is split
- Complaints and staff who feel safe saying “I’m falling behind” are the alarm most firms actually use
Our internal number, the sort of the baseline that, you know, subject to moving is 150 cases per case manager, that’s sort of where we’re operating, and now that we’ve broken through 300, we’re finding that that’s stressing the staffing levels.
— Josh
That 150 is for a case manager who does not pull medical records; a separate person does that on a records platform, and the firm’s working estimate is 250 to 300 cases per retriever. At the other end, one solo attorney carries about 800 cases with three legal assistants, and it works because the work is split by function, one on intake, one on appeals, one on records, rather than by caseload.
The large-firm practitioners were candid that their number is a guess.
We kind of have a random number in mind, but it’s just purely random. A lot of it depends on their skill.
— Jeff
His instrument is complaints: he stays out of day-to-day casework but every complaint reaches him, and staff know they can say they are overstressed. A 932-case operation described the same thing from the other side, ramping each new hire slowly and scaling back when client response times slip or work product suffers. One practitioner who came up through a larger firm put the cost of skipping this plainly.
I kept telling my boss, the problem is not the person, the problem is you’re asking them to do too much.
— Christy
What this means: the ratio is discovered per person, not set. Whatever your baseline, the number that matters is the signal that tells you it is wrong, and the firms with the best signal were the ones where staff could raise it themselves.
The Firms That Can Rattle Off Their Numbers Built One Report
- One firm runs a metric report every Sunday night with an activity index per staff member
- The first thing it surfaced was a data-entry problem, not a work problem, and that problem has a fee-petition cost
- A solo gets the same pulse check by asking her case management system’s AI a question before each one-on-one
- Contact rate is the number both rooms trust least, because a text counts the same as a conversation
One operations manager can recite her report’s columns: cases touched versus cases actually worked, documents generated, notes created or updated, tasks completed, documents deleted, and an overall activity index per person per week.
This one employee has an activity index of, let’s say, 800 for the week, another who is a high performer is around 750, and then my next one is at 600. They all have about an equal caseload. So that’s kind of a problem… Come to find out this one employee isn’t entering notes for all the interactions that they’re supposed to.
— Erin
Her rule that follows from it is the sharpest operating line of the month: no more work until the notes are consistent, because missing notes mean an inaccurate fee petition later. A solo attorney in the same room said he could find every one of those numbers in an hour or two, but could not rattle them off, and that gap, between data that exists in the file and data that arrives on its own, was the real subject of both sessions.
The lightweight version came from a practitioner on Clio who asks its assistant a question before meeting with staff.
It’s definitely not making any decisions, it’s more just, you know, like, before I meet with a staff member, I’ll ask it, tell me how many cases do they have at this stage? What are the number of contacts they’re having with clients.
— Kat
What this means: the report exists to prepare a conversation with a person, not to replace it. Build the one that runs on its own, then use it to decide who to talk to on Monday.
Cost to Carry a Case: Three Firms Are Computing It, Two Started This Year
- Log the date of every stage and you can price what a three-and-a-half-year case cost to hold
- Average in the zero-fee cases or the mean fee lies; one firm’s 2026 mean is $4,576 with 21 percent of closures at zero fee
- Intake to payment averaged 1.24 years at that firm
- The number the smallest practice in the room is proudest of is average fee per case
How long did we hold it? How much did it actually cost me to hold that case? Is there a real profit by the time I pay my staff, and what have you?
— Christy
She logs the date a case came in, the application date, every decision and appeal date, so a slow case can be priced instead of just resented, and so she can ask what would have gotten it approved sooner. The larger firm in the room does the same at the aggregate level.
I want to know a mean case value. I want to know what a case is worth to us… if I average in which I think you need to, all of our 0 fee cases, our average fee this year is 4,576. Average time from case intake to case payment is 1.24 years.
— Josh
A third firm is midway through a review of what each case costs per case manager, salaries included, and the takeaway one attorney named at the close was to start tracking average fee per case and time again after letting it lapse. The wish one attorney voiced for both rooms is the notice of award delivered electronically, because the fee history lives in the bank rather than in any case system.
What this means: the stage count is in every system; the unit economics have to be built from dates and payments. Two disciplines make it usable: log every stage date, and include the zero-fee cases.
Approval Rates Are Slipping at the ALJ Level, and the Response Is to Front-Load the Case
- Two firms in different states reported more ALJ denials this year; a third saw zero-fee closures rise
- The reaction across sizes was the same: do more at recon, with the initial-denial explanation in hand
- One large firm is retraining case managers from process work toward judgment about which cases need earlier records
- Logging the reason for each lower-level decision tells you which case types to keep taking
I recently have realized that more of my cases are being denied at the ALJ level. And I talked to some other practitioners who feel like they’re seeing the same thing… That used to be my favorite metric, now it’s my least favorite metric.
— Stephanie
One attorney wondered aloud whether adjudicators are using AI; nobody in the room could say. What the larger firm could say was what it is doing about it.
Review that initial denial DDE and say, okay, this is what they found missing, let’s try to address that while the case is at reconsideration. Instead of waiting till the hearing level, when someone else is going to look at it, and then there’s not enough time to do anything.
— Jeff
He is also moving his case managers away from purely process work, filing 3441s and ADL forms, toward identifying cases where earlier medical reports would make the hearing stronger. A smaller practice has started logging whether each lower-level approval came from a listing or a grid rule, to steer intake.
What this means: the approval rate is a lagging number. The leading ones are the reason for decision and how much of the hearing-level work you can pull forward to recon.
Cost Per Retained Case Is the Only Marketing Number, and the Referral Field Is Never Right
- Two firms independently land at a cost per retained case under a thousand dollars against a fee near $4,500, roughly a 5x return
- Both walked away from referral marketplaces on cost; one said the platform’s take is moving from 25 to 30 percent of the fee
- Every firm that tracks referral source said clients get it wrong at the point of entry
- Live chat is dying; a form and a call are what prospects actually use
Cost per retained case. I mean, maybe my thinking’s wrong, but nothing else matters to me, but a case.
— Josh
The number survives the referral-source noise because it needs only total spend and retained cases. The referral field itself was described the same way in both rooms: clients say they saw a TV ad the firm never ran, and word-of-mouth referrals arrive through Google, so Google gets the credit. It is still worth keeping for the one thing it catches reliably, a new source showing up in a cluster.
I noticed we had, like, 3 or 4 cases that were being referred to us from an actual pain clinic. So I was like, okay, this is important, like I need to track some time out, drive down there, introduce myself.
— Christy
One practitioner priced voice-AI intake at $11 per call and rejected it, because his firm turns away about 70 percent of callers and the AI would be billed for all of them; he is building his own. Another has moved chat to a form because, in his words, nobody uses live chat any more and it is one more thing to break.
What this means: compute cost per retained case by channel and ignore the rest of the funnel until that number is wrong.
Smaller by Design: Automation Shrank Staff, and the People Left Are Burning Out Faster
- One firm’s staff got smaller as it automated; another’s caseload got larger for the same reason
- A 900-case department wants to grow with the staff it has, because hiring for this work is harder than it was
- Client hostility toward staff was described as the worst in 22 years
- A solo named the three tools that let him carry more cases and do better work
My staff has actually gotten smaller. Which is a weird thing to say and to see. As we’ve automated more.
— Diane
She practices from an island with no labor pool to hire from, so the goal was forced on her. A 250-case practice in the same room said the mirror image, that its caseload has grown because it can handle more than it could five years ago. A large mainland firm arrived at the same goal because the people are hard to find and harder to keep.
Our clients are angry and impatient, and our staff get really burnt out much faster than they used to.
— one operations manager
Her firm’s rule is that staff are not doormats: empathy is the job, verbal abuse is not, and staff may end the call. The solo with about 60 cases credited three tools, a research assistant for reading records, a medical-records and brief-drafting tool, and daily ERE monitoring, with letting him handle more cases and do a better job on each.
What this means: three firms arrived at the same operating goal from different directions, more cases per person rather than more people. Protect the people you have first.
About This Series
The SSD community has been missing a regular virtual space where practitioners in similar situations talk through real operational problems together. Chronicle built Disability Peers in Practice to be that space.
Sessions run monthly, segmented by firm size, free and capped to keep them small. The format is peer-led and practitioner-first: structured conversation focused on what’s actually working in the room.
Register for the next session →
About Chronicle
Chronicle is an ERE monitoring and analysis platform built for Social Security disability practices. It checks the SSA’s ERE and e-file daily across your firm’s cases, surfacing status changes, new documents, and upcoming deadlines before they become problems. Chronicle is CMS-agnostic: it works with any CMS with an API, or with no CMS at all. Disability Peers in Practice is one of the ways Chronicle brings the SSD community together outside of conference season.