When two deal teams at the same firm negotiate similar contracts to different outcomes, most observers attribute the difference to negotiating style, client-specific factors, or the relative leverage of the parties. Sometimes those explanations are correct. More often, the inconsistency reflects something more tractable: one team knew what the firm had done before, and the other team did not.
This is not a matter of attorney skill or diligence. It is a matter of access. A firm's institutional knowledge about its own positions, what it has historically accepted on key clauses, where it has held firm and where it has conceded, is distributed unevenly across the people who built that knowledge. When those people are on the deal, the knowledge is in the room. When they are not, it is largely absent.
Why the Problem Is Structural, Not Individual
It is tempting to frame inconsistency across deal teams as a training problem or a supervision problem. If junior attorneys had better training on the firm's standard positions, or if partners reviewed drafts more thoroughly, the inconsistency would narrow. There is something to that, but it misses the structural layer.
The reason institutional knowledge is distributed unevenly is not that some attorneys work harder at learning it. It is that institutional knowledge, in most transactional practices, is primarily transmitted through direct experience: being on enough similar deals, under the right supervising attorneys, that you internalize what the firm does. That transmission is slow, it depends on the accidents of staffing, and it does not scale as the firm grows or as personnel turns over.
A firm that handles a high volume of commercial agreements is staffing those matters across multiple teams. Not all of those teams include the partner who has done the most work in that deal type. Not all of those associates have been on enough similar matters to have internalized the firm's positions. The inconsistency is an output of normal operations, not a failure of any individual.
What Inconsistency Actually Looks Like
The most visible form of inconsistency is the position variance: two similar deals with different liability caps, or two similar deals where one has a mutual indemnification structure and the other is one-sided. Clients who use the firm regularly across similar transactions sometimes notice this. Counterparties who deal with the firm on multiple occasions notice it too. Neither group is able to predict what positions the firm will take in negotiation, because the firm's positions depend partly on which team is doing the deal.
There is also a less visible form of inconsistency that involves the handling of counterparty proposals. When a counterparty proposes aggressive language on a specific clause, a team with deep context on the firm's past practice knows quickly whether this is something the firm has resisted consistently or something it has occasionally accepted. Without that context, the response to the counterparty's proposal is less calibrated. A concession that would have been quickly rejected by an informed team might be accepted by an uninformed one, not because the latter team thought it was good practice, but because they did not have the information needed to push back with confidence.
The Form Book Partial Solution
Most firms with significant transactional volume have invested in some form of standardized forms or playbooks. These are genuine attempts to codify the firm's standard positions and make them accessible to attorneys who lack the individual experience to have internalized them. For firms that maintain these well, they narrow the inconsistency problem on the clauses that are covered.
The gaps in form books are the clauses that are not covered, the positions that have evolved since the form was last updated, and the deal-type-specific variations that a general form cannot capture. A playbook for commercial agreements does not necessarily reflect what the firm does in commercial agreements in a particular industry, or in a particular deal size range, or when representing a particular type of counterparty. Those distinctions matter in practice, and they are where experienced attorneys deviate from the general form in ways that less experienced attorneys do not know to do.
There is also a maintenance problem. Form books require investment to keep current. The firm's positions evolve through deal experience, and if the form book does not keep pace, it becomes a record of what the firm used to do rather than what it does now. For firms with limited bandwidth for form book maintenance, the playbook gradually diverges from current practice without anyone intending it to.
The Precedent Bank Approach
August takes a different approach to the consistency problem. Rather than trying to codify the firm's standard positions into a maintained playbook, it derives those positions from the actual closed deals in the precedent bank. When an attorney on a deal team reviews a clause against the bank, they are seeing what comparable deals looked like and how they were resolved, with specific citations to the underlying agreements.
This has a different relationship to the form book maintenance problem. The bank does not require a separate maintenance investment: each new deal that closes and enters the bank updates the picture automatically. The most recent deals are always in the bank. If the firm's position on a clause type has shifted in the past two years, recent deals reflect that shift, and the bank comparison surfaces it when the relevant clause comes up in a new review.
It also handles deal-type variation naturally. A comparison run against the bank for a technology licensing matter produces comparisons to technology licensing matters. A comparison for a commercial real estate transaction produces real estate comparables. The attorney does not have to know which subset of the general playbook applies to this specific deal type. The system identifies the relevant comparable set based on the deal's characteristics.
What This Does Not Solve
We should be precise about the limits of the precedent bank approach to consistency. It narrows the knowledge gap between teams with deep institutional context and teams without it. It does not eliminate the judgment variation that comes from different attorneys weighing client priorities differently or having different assessments of negotiating leverage. Those variations are not a problem to be eliminated. They reflect the exercise of professional judgment that is what clients are paying for.
The precedent bank is also only as consistent as the underlying deal history it draws on. If the firm has been inconsistent in its positions over time, the bank reflects that inconsistency. The goal is not to force consistency where deliberate variation was appropriate. It is to ensure that where inconsistency is unintentional, driven by information gaps rather than deliberate judgment, the information gap can be closed.
Over the past year, we have been building August alongside transactional practices that range from general commercial work to more specialized deal types. The consistency problem looks similar across those contexts: the underlying cause is almost always the same information distribution issue, and the precedent bank addresses it in the same way. The specific deal types are different, the particular clause variations are different, but the mechanism of the solution is the same. That consistency in the mechanism is one of the things that gives us confidence it is addressing a real structural problem rather than just a surface symptom.