Why I Check AI Visibility Before I Read the P&L

I look at what AI says about a business before I look at its books. Many founders find that backwards, so let me explain. They expect a finance person to open with the P&L, because the P&L is where the money is. They are right that it is where the money is. They are wrong about which direction the money flows from, and that one error is why so many slow months never get explained.

A P&L reflects past performance. It is the most honest document in your company, and it is a rear-view mirror, and those two things are not in tension. The numbers are true. They are also late. By the time a slowdown appears in the numbers, the underlying causes may have occurred weeks or even months prior. So I have spent my whole career doing one thing with that fact: refusing to be surprised by it. Trace the dollar back to its source, and watch the gauge that turns before the number does.

What is a leading indicator, and why does a CFO chase them?

A leading indicator is a signal that moves before the result it predicts, which is exactly why a CFO chases it instead of settling for the report. The report is the lagging indicator. It confirms. The leading indicator warns, and warning is the only thing that leaves you time to act. By the time revenue confirms a problem on the income statement, the window to fix it cheaply has usually closed.

This is not a new instinct, and it has nothing to do with AI. I have always tried to anticipate issues rather than just report what already happened. Monitor cash conversion before the cash crunch hits, which is the whole argument behind why a profitable business can still run out of cash. Assess channel margins before a blended average obscures them and one quietly bleeding line of business disappears into an average that looks fine. A bookings trend turns before recognized revenue turns. In every case the job is the same: find the gauge that moves first, and stop reading the dashboard like a history book.

Why is a P&L a lagging indicator?

A P&L is a lagging indicator because it can only record transactions that already closed. It is a settlement document. By design, nothing reaches it until the sale is made, the cost is incurred, and the period is over. That is what makes it trustworthy, and that is what makes it slow.

Read a P&L and you are reading the end of a long chain of events. A customer had a need. They looked for an option. They found you, or they did not. They considered you, chose you, bought from you, and only then did a number appear. The income statement sees the last link in that chain. Everything that determined whether the link ever formed happened upstream, before a single figure was entered. So when an owner asks me to explain a flat month from inside the P&L, I usually cannot, because the answer is not in the document. It is in what happened before the document had anything to record.

Where does AI visibility sit in the chain to revenue?

AI visibility is just the latest tool, and it provides insight earlier in the process than anything found in the financial statements. Long before a customer makes a purchase, an engine has already decided whether your business even makes the short list it hands them. That happens before the visit, before the order is placed, before anything ever touches your books. The decision comes first, and everything the P&L eventually records depends on it.

Think about how buying actually starts now. Someone opens ChatGPT, Claude, Perplexity, or Gemini and asks for the best option in your category, in your area, for their situation. The engine returns a handful of names. That short list is the new top of the funnel, and it forms before any behavior you can measure with a traditional tool. If your business is in that consideration set, you get a shot at the visit, the order, the revenue. If you are not in it, the chain never starts, and nothing downstream of it ever gets a chance to happen. I have written separately about how this is its own discipline in the difference between SEO, GEO, and AEO, but the financial point is simpler than the acronyms: this is now the first gauge on the line.

Why is this a finance problem and not a marketing problem?

It is a finance problem because it is a revenue problem, and revenue is finance's job to protect at its source. It is a leading indicator of revenue, and most owners have never once checked it. Marketing owns the message and the channels. Finance owns the question of whether the dollar starts at all, and right now the dollar starts the moment an engine decides whether to put you on that short list. That is upstream of every campaign, and it lands directly on the top line.

I am not trying to annex marketing's work. I am pointing out that the earliest input to revenue has quietly moved to a place no one on the team is watching with a financial eye. Marketing tends to measure AI visibility, when it measures it at all, as a brand or awareness metric. I measure it the way I measure cash conversion or channel margin: as a leading indicator of dollars. My goal remains the same one it has always been. Trace the dollar back to its source, and monitor the signals that others often ignore. The source just moved, and it moved into the model.

How does invisible revenue hide from your books?

Invisible revenue hides from your books because the customer you lost never reached them, and you cannot record a transaction that never started. This is the part founders find hardest to accept. A bad month at least announces itself. Invisible revenue does not announce anything, because there is nothing to announce.

It does not just manifest as a bad month. It results in a silent flat line in your revenue that you cannot explain from inside the organization, because the potential customer never even knew you existed. Sales are not collapsing. They are simply not growing the way the market should be handing you, and every internal number looks normal, because every internal number only counts the people who found you. The customer who asked an engine for the best option, got a list, and never saw your name is invisible by definition. They did not bounce off your site or abandon a cart. They leave no trace in any system you own. That is the most expensive kind of loss precisely because it is the kind you cannot see, which is the whole subject of what it costs to be invisible in AI search.

What does this mean for you?

What it means is that your earliest revenue signal has moved upstream of everything you currently track, and reading the P&L harder will not bring it into view. If your numbers look flat and you cannot find the reason inside your own data, that absence is itself a clue. The reason may be sitting one step before your data begins, in the answer an engine is giving about your category without you in it.

This is the work I built Main Street IQ around, and it is why I do not treat finance and AI visibility as two separate jobs. They are one job: trace the dollar back to its source, and watch the gauge no one else is watching. For years the most upstream gauge I could reach was cash conversion or channel margin. Now there is a gauge further up the line, and it determines whether the dollar enters the chain at all. A CFO who ignores it is choosing to be surprised, and getting surprised is the one thing the whole discipline exists to prevent.

Your Monday-morning next step

Monday morning, name one leading indicator your business does not yet track on purpose, and make the AI-visibility check the first one. Open ChatGPT, Claude, Perplexity, and Gemini, and ask each the question a real customer would ask to find a business like yours. Then read the answers the way you would read a finance document, not a marketing one. Are you in the consideration set? Who is? If you are not in the answer, you have just found the most upstream gap in your revenue chain, and you found it before it ever reached your books.

So here is the question I would actually ask you. For your business, what is the earliest signal you track before a number changes? If the honest answer is the P&L, you are reading the rear-view mirror and calling it a windshield, and the gauge that saw it all coming is sitting one step further upstream than you have been willing to look.