A director once asked me why the analysts were asking so many detailed questions. My answer was that an owner of the business would ask the same exact questions – and that shareholders should be treated like owners. A private owner, however, could look at the operating model, whereas public-market investors have to work backward from what management gives them.
That's how I think about earnings Q&A. Most investor questions are attempts to understand an assumption behind future revenue, earnings, or cash flow. The question may sound qualitative. The decision usually isn't.
When an analyst asks about sales productivity, he's trying to forecast revenue. When he asks about hiring, he may be trying to understand the margin path. Customer behavior can tell him something about retention. He's asking because the answer could change what he's willing to own and what he's willing to pay for it.
The model management already knows
CFOs spend all day inside their company's numbers. They know when a new salesperson becomes productive, how bookings turn into revenue, and which costs should disappear next year. After enough time with the model, those relationships start to feel obvious. Almost like second nature.
But they aren't obvious to someone coming into the business from the outside. Investors are analyzing hundreds of companies. They’re swamped in their review of other companies’ results, and they’re also deciding where to spend their time. They don't have the benefit of every operating review management attended during the quarter. So all of what you tell them is “news to them.”
That creates a meaningful disconnect. Management thinks it’s already explained the answer. The investor thinks a basic assumption is still missing. Repeating the original language more emphatically won't necessarily resolve it.
Suppose, hypothetically, a company says it plans to add salespeople. The expense begins when those people join. Their revenue contribution, however, comes later and ramps at a cadence unknown to the investor. If management explains the hiring plan without explaining the ramp, an investor can understand the strategy and still have the wrong margin forecast.
The useful answer connects the operating decision to the financial consequence. It doesn't require handing over the internal model, but it does require understanding which relationship the investor is trying to reproduce.
What number is behind the question
An investor asks how demand looks next year. Management says demand “remains strong.”
That might accurately describe what the CEO sees. But the investor still needs to know what “strong” means. Are existing customers spending more? Are sales cycles shortening? Is the improvement broad enough to affect revenue, or concentrated in a small part of the business?
The adjective leaves those assumptions unresolved. If investors keep asking, the problem could be missing evidence, evidence that doesn't connect to the forecast, or a history that makes them reluctant to take management's word at face value.
I'd rather a team work through that than decide the analyst wasn't listening or is “too demanding.” An answer can be clear English and still be financially unhelpful!
Prepare the assumption before the answer
Before an earnings call, take the questions you're expecting and write down what each one could change in an investor's model. Then decide which evidence supports the answer and where the uncertainty remains.
A competition question could be about pricing, retention, or the cost of winning new business. Don't assume every version requires the same response. If you're unsure which issue the investor means, ask.
There are limits to what a company can responsibly forecast or disclose. A useful answer can explain the drivers, distinguish what management knows from what it doesn't, and connect the discussion to the public financial framework. Precision doesn't require pretending you know a number you don't.
The test I'd use after a rehearsal is straightforward: what could the investor now do with that answer? If the response sounds confident but leaves the same assumption unresolved, the preparation isn't finished.
