99% of Mass Arb Cases Never Reach an Award. Your Admin Stack Is Built for the 1% That Do.

Here is a number that should reframe how you think about your operations budget: fewer than 1% of consumer mass arbitration cases that reach the merits end in an award. AAA administered 104,556 consumer demands across 81 mass-arbitration caseloads, and the vast majority resolved before an arbitrator ever ruled.

That is not a failure of the system. That is the system working as designed. Mass arbitration is a settlement mechanism with an arbitration wrapper.

But if that is true, and the data says it is, then why is every claims administrator in the market building infrastructure optimized for hearings, evidence submission, and award processing? The answer is legacy architecture. Most admin platforms were built for class actions, where the endpoint is a court-approved settlement and the admin's job is notice and distribution. They bolted on arbitration features the way you bolt a roof rack onto a sedan. It holds, until you load it up.

The 99% Problem Nobody Is Staffing For

If your matter is going to settle (and it almost certainly is), the operational work that determines your outcome happens in three places:

1. Claimant engagement between filing and settlement. The defendant's entire strategy is delay. Bellwether selection, batching, procedural motions. Every month that passes, your claimant pool decays. People move. They change phone numbers. They lose interest. They forget why they signed up. Your admin stack needs to keep 5,000 or 10,000 or 50,000 people warm across a timeline that could stretch 12 to 24 months.
2. Settlement execution at scale. When a deal lands, you need signatures on releases and closing statements from a supermajority of your claimants, typically 75% to 85%, within a compressed window. This is not a mailing. This is a sales funnel with a deadline. The firms that hit threshold planned for it operationally before the settlement term sheet was signed.
3. Disbursement without leakage. Getting money out the door, handling returned payments, processing 1099s, managing escheatment for unreachable claimants. Every dollar that sits in an escrow account past a deadline is a dollar that erodes your client's trust and your firm's reputation.

Notice what is not on that list: arbitrator scheduling, evidence packet assembly, hearing logistics. Those matter for the 1%. The 99% lives and dies on engagement, execution, and disbursement.

What "Built for Hearings" Actually Costs You

When your admin platform was designed around the hearing endpoint, the cost structure follows. You are paying for arbitrator coordination tools, case management dashboards oriented around procedural milestones, and document repositories built for evidentiary submissions. Those are real features. They are also features you will use on maybe 50 of your 5,000 claims.

Meanwhile, the features that determine whether you hit your release threshold are treated as add-ons or afterthoughts. SMS outreach? Licensed from a third party. Document signing? Another vendor. Caller ID branding so your claimants actually pick up the phone? Another contract. AI-driven inquiry handling so claimants can get answers at 11 PM on a Tuesday? If it exists at all, it is a recent bolt-on with its own billing line.

Every one of those vendor layers adds cost. But more importantly, every vendor boundary adds friction. Your SMS provider does not talk to your signing platform. Your signing platform does not talk to your engagement dashboard. Nobody is tracking which claimant opened an email but did not sign, then triggering a follow-up text 48 hours later, then routing their callback to an AI agent that can answer "where is my money" without a human in the loop.

That orchestration is the product. Not the hearing prep. Not the evidence management. The orchestration across channels, across months, across tens of thousands of individual humans who each need a slightly different nudge at a slightly different time.

The Math on Owning vs. Renting the Mid-Campaign Stack

Let me put numbers on this because that is what actually moves decisions.

A legacy administrator running a 10,000-claimant matter will typically charge $20 to $25 per claimant for the full administration. That covers notice, outreach, signing, and disbursement. Sounds reasonable until you realize that the outreach component, the part that determines your threshold, is where they have the least margin. Every additional SMS costs them money. Every additional call costs them money. Every AI minute costs them money. Their incentive is to send fewer touches, not more.

Now compare that to an admin that owns the entire stack natively. Native document signing (not licensed from DocuSign or HelloSign). Native SMS and email delivery (not routed through Twilio with a markup). An AI super agent built on top of that same infrastructure, handling over 80% of claimant inquiries on its own. When the marginal cost of an additional touch is near zero because you own the pipes, you send the twelfth text. You make the ninth call. You keep showing up because showing up is not a cost center, it is the product.

That is how you get from $20 to $25 per claimant down to $9.99 per fully signed claimant. Not by cutting corners on the hearing side. By eliminating the vendor stack that makes every outreach touch a line item someone has to approve.

What the AAA Data Actually Tells You About Provider Selection

The 104,556 demands across 81 caseloads give us an average of roughly 1,290 claimants per mass arbitration. But averages hide the distribution. Some of those caseloads are 200 claimants. Some are 20,000. The operational challenge scales nonlinearly.

At 200 claimants, a decent paralegal team with a shared Google Sheet can probably manage engagement. At 2,000, you need real infrastructure. At 20,000, you need infrastructure that was designed from the ground up for persistent, multi-channel, months-long engagement with individual humans who did not ask to be in your campaign and will ignore you unless you make it easy to respond.

When you are evaluating an administrator, ask this question: what percentage of your platform's development budget went into hearing-side features versus engagement-side features? If the answer is anywhere close to even, you are paying for a product built for the 1%.

The better question: how many outreach touches can you send per claimant per month without a per-touch surcharge? If the answer involves a pricing conversation, you have your answer about their cost structure.

Build for the Settlement, Not the Award

I am not saying hearings do not matter. They do. The credible threat of 5,000 individual arbitrations is what brings defendants to the table. But once they are at the table, your operational challenge flips entirely. You are no longer preparing for hearings. You are running a completion campaign against a clock.

The firms that consistently hit release thresholds are the ones that understood this before the settlement term sheet arrived. They chose an admin stack built for engagement, not adjudication. They locked in flat per-claimant pricing that does not penalize persistence. They built their outreach cadence around the reality that a claimant who ignores you in week one will sign in week twelve if you keep showing up in the right channel at the right time.

Fewer than 1% reach an award. Build your operations for the 99%.