This is the second in a short series of blog posts examining the pricing of legal services in the era of AI.
Pricing legal services in the era of AI (2): Why existing pricing thinking doesn’t solve the problem
In my last post, I suggested that AI is breaking the economic model of legal services.
The uncomfortable truth is this: we’ve been talking about the alternatives for years — and they still don’t work. Value pricing, fixed fees, subscriptions, alternative fee arrangements. The profession has not been short of ideas. But when you look at how these are actually used in practice, very little has really changed.
Take value pricing.
Almost every partner will now say that they price on value. Yet if you look closely, very few actually do. What tends to happen is something much more familiar: an estimate of time is made, a rate is applied, and the number is then adjusted to reflect the client, the matter and the relationship.
That is not value pricing. It is the billable hour, slightly disguised.
Fixed fees are more interesting — and more nuanced.
They are not the problem. In fact, they are likely to be a large part of the future. Clients want certainty, transparency and, increasingly, choice — different price points for different levels of service. Fixed fees can deliver that.
The difficulty is that most firms are not using them in this way. Instead, the fee is derived from an internal estimate of time and presented as a single number. In competitive situations, that quickly becomes a race to the bottom, with firms invited to match or undercut.
Used properly, fixed fees look quite different. They are not a single price, but a set of choices — different scopes of work, different levels of input, different allocations of risk, each clearly defined and priced. The client is not simply given a price; they are given options.
That, in turn, requires clarity. Scope needs to be properly defined. Assumptions need to be explicit. Boundaries need to be understood. And pricing needs to be consistent enough across the firm to be credible.
Most firms are not there yet.
Subscriptions are often presented as the next step.
The idea is attractive — predictable revenue, stronger relationships, closer alignment with clients. But again, the difficulty is execution. Firms struggle to define scope tightly enough, or to price the variation in demand with any consistency.
What emerges is rarely a true subscription model. More often, it is a loose retainer with unclear boundaries and uneven profitability.
This is the pattern.
Each “new” approach is pulled back into the old model. Not because lawyers resist change, but because the underlying capability has not changed. Pricing decisions are still made individually, subjectively and inconsistently, partner by partner, matter by matter.
The firm still has no reliable way of answering a very simple question: what is this matter worth, to this client, right now?
And that is the real issue.
The problem is not a lack of pricing ideas. The profession is not short of models or terminology. The problem is that there is no consistent way of turning judgment into price.
AI does not fix that. If anything, it exposes it.
So we end up in an uncomfortable place. We know the old model is under strain. We experiment with new ones, but nothing quite holds.
Which leaves a simple but uncomfortable question. If we are no longer selling time, and we cannot reliably price “value”, what exactly is it that clients are paying for?
That is the question we need to answer.