Classification that used to take five minutes now takes four hours. How is anyone pricing that?

There are two problems here, and only one of them is about money.

The first is evidential. If you cannot show where the four hours went, you have no basis for the conversation with the client and you will lose it. Record time against the file, and separately record time spent on research that benefits the client but attaches to no single entry. The second category is the one that habitually goes unbilled, and it is often the larger of the two.

The second problem is contractual. A fee schedule written before the work changed does not bind you to lose money indefinitely; it is simply the schedule you have not yet renegotiated. Take the record you have built, show it, and ask to be paid hourly for classification specifically. If they refuse, you will at least know what that relationship costs you.

We took the simple route: the price follows the number of lines, not the hours.

If one article ends up carrying four provisional codes, that is four lines to build, four to check, and four to defend if any of them is queried. So it is billed as four. Nobody has argued with it yet, because the customer can count the lines on their own paperwork.

The advantage over hourly is that it does not ask anyone to trust your timesheet. The disadvantage is that it undercharges exactly the case you are describing, where the research runs long and the finished line is short.

Sorry to be the cautious one here, and please tell me if I have this wrong.

During onboarding we were told not to discuss fee levels with other brokers, and that the rule does not really care whether the conversation is anonymous. I have been reading this thread wondering whether describing a pricing method falls inside that or outside it.

My honest question is whether there is a line between saying how you structure a fee and saying what you charge. Those feel different to me, but I would not want to be confident about that on my own reading of it.

The concern is a real one, and it exists here as well under competition law, so it is better to state the distinction precisely than to avoid the subject.

What is prohibited is an agreement between competitors about price: to set it, to hold it, to adopt a common formula, to stop discounting. The element that carries the offence is the agreement. Observing what a market does, or describing how your own business organises its work, is not that.

Where I would be careful is the formula. Several firms converging on one additional charge per additional line, because they all read it in the same place, begins to look like a standard, and a standard adopted in common sits closer to the prohibited thing than any individual number ever does.

That is the right place to draw it, and it matches how we are told to behave.

Whenever brokers meet as a group, somebody reads the antitrust statement before anything else is discussed. Not because the subject is forbidden, but because the room is the risk. A number said out loud among competitors is a different object from the same number said to a client.

So: methods, yes. Rates, no. Which is what I asked for, and I have got more out of it than I expected.