Essay · Applied for Operators

The First Three Sentences

Autumn 2026 · Reading time · 15 minutes

Pick any first exchange your company had with a stranger this week, a new customer, a new candidate, a new counterpart. Now look only at the first three sentences your side said. That is the entire diagnosis. Everything you need to know about what your team has really been trained in, and what your standard actually is when nobody is watching, lives in those three sentences.

A working owner who has been paying attention will recognize this test. Every first contact is a compressed portrait of the company. The person on the other side arrives with no context. Whatever your side offers in the first three sentences is what the company genuinely is at that moment, unedited by relationship, unearned by history, unsoftened by the trust that develops later. It is the raw signal. And in a season when the first three sentences are, more and more often, half-written by a tool the owner has never audited, the raw signal has become the most important thing to inspect in the whole operation.

The compression the openings expose

Openings are the most compressed artifact any business produces. Three sentences is not much space, and every decision the company has ever made about who it is has to fit into that space or be sacrificed to it. That is why openings are so revealing. They cannot contain everything, so they contain what the company has, in practice, prioritized. If the priority has drifted, the drift shows up in the compression first. Long before the drift is visible in retention numbers or customer surveys, it is legible in the first three sentences your team is sending, because compression forces choice, and choice under pressure exposes the underlying operating system.

Owners who become fluent at reading their own openings often notice a strange phenomenon: they can predict, from the first three sentences alone, how the rest of the exchange will go. Not always. But often enough that the prediction stops being coincidence. That predictive power is not magic. It is the fact that compressed artifacts encode a lot of information about the sender, and a trained ear can hear it. Once the owner starts hearing it, she cannot un-hear it. That is one of the productive irreversibilities of the practice. The ear, once trained, becomes an always-on quality check that no dashboard can substitute for.

What the first three reveal

The first sentence reveals what the company thinks it is. It is the frame the team places around itself before the other person has said anything. A team that leads with credentials is telling you it is anxious about being taken seriously. A team that leads with a question is telling you it is confident enough to open the room to the other side first. A team that leads with a scripted greeting drafted by a tool is telling you the owner has not sat down with the intake in over a year. Every one of these is a real signal, and none of them require a survey to detect. They require the owner to open one first-contact transcript per week and read the first sentence out loud.

The second sentence reveals what the company thinks the other person is. This is where most of the invisible drift lives. A team that treats the other person as a problem to be handled will begin the second sentence with a policy. A team that treats the other person as an adult with information the company needs will begin the second sentence with a listen. The word choice will be small. The signal will be enormous. Read across a month of second sentences and you have an almost perfect map of how your company sees the people it exists to serve. That map is often unrecognizable to the owner, because the owner’s own second sentences do not sound anything like it, and she has never asked whether the training she thought was flowing outward was actually landing.

The third sentence reveals what the company plans to do next. It is the pivot. Does the team move the conversation toward a decision, a piece of information, or a piece of friction. Does it name the next step or leave it implied. Does it invite the other person into the process or push them into a queue. Every serious operator I have watched, when she reads back three months of her team’s third sentences, sees a pattern she did not know she had trained in. The pattern is honest. It is also almost never what she would have designed if she had noticed it earlier.

The opening that stopped being a formality

For a long stretch of business history, the opening of a first exchange was a formality. Everyone knew what a greeting was supposed to sound like. Templates existed. Manners were shared. The opening carried very little information because it was almost entirely convention. That has changed. In a market where every counterpart has already seen a hundred templated openings this month, the formality no longer functions. A conventional opening now signals absence, not politeness. The other person immediately reads the message as machine-generated, whether it is or not, and adjusts her attention downward. The opening that used to be safe is now the opening that quietly loses the room.

That is why the first three sentences have become a strategic surface. Not because they are the whole conversation, but because they are the piece the other person uses to decide how much of herself to bring to the rest. A distinctive opening earns attention. A generic opening earns half-attention. Half-attention is where deals slow down, where candidates disengage, where customers churn quietly. The opening has been promoted from social ritual to competitive instrument, and owners who have not noticed the promotion are competing at half-attention without realizing why their close rates have softened.

Why the machines make this even more diagnostic

In many companies the first three sentences are now co-authored. A tool suggests, a person edits, a message goes out. This is not a threat to the diagnostic. It is an amplifier of it. Because the tool’s default is generic. If the first three sentences your team is sending read as generic, that tells you the person on your side accepted the tool’s suggestion without shaping it, which tells you either the person is not experienced enough to know the standard, or the standard has not been written down where the person could apply it. Either way, the finding is actionable and the owner can start on it Monday.

There is a subtler version of this. Sometimes the person on your side edits the tool’s suggestion into something that reads well but is not yours. The sentences are competent. They are also indistinguishable from the sentences a competent adult at any competitor would have produced. That is the sound of the company losing its voice one message at a time. The remedy is not to remove the tool. The remedy is to give the tool, and the person, a written specification of what your voice actually is. Otherwise the median wins by default, and the median in your industry is now very fluent, very generic, and very forgettable.

The company that has never audited its openings

A company that has never audited its own openings is not a lazy company. It is usually a busy one, run by a capable owner who has been focused on the middle and the end of every conversation, because the middle is where the work gets done and the end is where the outcome gets closed. The opening feels like a formality, so it gets delegated to whoever is on rotation, to whichever template someone drafted a year ago, to whichever suggestion the tool put on the screen this morning. That delegation is entirely rational until you notice that the opening is the only part of the conversation the other person will ever remember with clarity, because it is the only part before context has softened the impression.

First impressions are not a soft topic. They are the compression of every other operating decision the company has made, packaged into three sentences a stranger will use to decide whether to keep listening. An audit of the openings is not a branding exercise. It is a compression audit. Owners who take it seriously find, without exception, that the compression is wrong in ways that the middle and the end of the same conversation would never have surfaced. The middle is where the team performs. The opening is where the company is.

The audit that changed my own week

The first time I did this audit seriously on my own businesses, I found three patterns I did not know I had trained in. The first was that my teams almost always opened with an apology, even when nothing had gone wrong. The second was that my teams described their own service in the second sentence far more often than they asked the counterpart a question. The third was that my teams routinely closed the third sentence with a next step that put the burden of the next action on the counterpart. None of these were catastrophic. Every one of them was fixable in a week. Together they were the reason a certain quality of prospect always left before the second conversation, and I had never once been able to identify why.

Fixing the three patterns did not require a large intervention. It required naming them, writing a short calibration piece for each function, and reviewing the outgoing openings for a fortnight until the pattern shifted. Within a month, the fortnightly review was no longer necessary. The team had internalized the calibration and the tools they were using had been given the calibration as context. The productivity gain from that single audit was larger than any campaign I had run that quarter, and it cost nothing but attention. That is the highest-leverage kind of intervention available to a working owner and it is available in the openings.

The exercise, weekly, small

The exercise is short and it is the single highest-leverage inspection routine I know for a working owner. Once a week, pick three first-contact exchanges from anywhere in the business, sales, support, hiring, partnerships, it does not matter which. Read only the first three sentences your side sent. Write down what each of the three reveals. Do not read the rest of the conversation until you have made your notes. Then read the rest and check whether the conversation confirmed or contradicted your read. Do this for a month and you will have a more accurate picture of your company’s actual behavior than any survey, any dashboard, or any team-health report will ever give you.

The exercise takes twenty minutes a week. The insight it produces is durable enough to reshape how you train, how you hire, and how you design the tools your team is using. Most owners resist it because it feels beneath them. It is not beneath them. It is the same discipline a serious craftsperson uses to inspect the beginning of every piece she is not yet allowed to inspect the end of. The beginning is where the piece decides what it is going to be. Everything after is a consequence.

The three-sentence draft I make my team keep

One practical piece of the standard I maintain across every function is a written three-sentence opening for each first-contact scenario the business regularly encounters. Not a template to be pasted verbatim. A calibration piece that shows a team member what a strong first three sentences look like in that specific situation, so that when she is drafting her own, aided or unaided by a tool, she has a concrete reference for what the standard is. Three sentences is short enough to hold in memory and specific enough to actually guide the writing. Longer specifications get skipped. Shorter ones do not calibrate.

The teams that have adopted this practice tell me two things consistently. The first is that the calibration piece becomes the most-referenced internal document in the function within a month. The second is that the tools they use start producing openings that match the calibration much more closely, because the calibration piece can be given to the tool as context. What began as a training aid for humans becomes, without any additional effort, a conditioning input for the machines the humans are working alongside. That double effect is the reason the three-sentence draft is now, in my opinion, one of the highest-leverage documents a working owner can maintain.

The first three sentences the owner sends

It is worth turning the audit on yourself before you run it on the team. Pick three first-contact exchanges you personally initiated in the last month. Read only your first three sentences. Ask the same questions. What did you signal you were. How did you treat the other person. What did you propose as the next move. Owners who do this on themselves before running it on the team almost always find that they are guilty of at least one of the patterns they were about to correct in others. That finding is the correct place to start. Fixing your own openings first gives you the moral standing and the practical experience to help the team fix theirs.

There is a version of this practice that stays with the owner as a personal discipline long after the team has internalized it. I still audit my own first three sentences on any exchange that matters. The audit takes seconds and it has saved me from sending openings I would have regretted more than once. It is the smallest possible quality check and it produces a disproportionate return on the exchanges that go on to matter, which is most of them, because the exchanges that begin well are the ones that get to continue.

What to correct, and what to leave

Not everything you find in the first three sentences deserves a correction. Some of it will be a person on your team responding correctly to a situation you have never seen, in a way you would not have chosen but that produced the right outcome. That is not drift. That is your team maturing past your default. Leave it. What deserves correction is any first three sentences that would have embarrassed you if a peer had read them. The embarrassment is the useful signal. Not because the sentences are catastrophic, but because they mark the gap between the standard you carry in your head and the standard your company is actually shipping. Close that gap, one small correction at a time, and the whole company begins to sound like itself again inside a quarter.

The final thing to notice, if you run this exercise for long enough, is that your own first three sentences begin to change. Not because you were trying to change them, but because reading a hundred of your team’s openings sharpens your ear for your own. You will find yourself pausing before you send the first message in any new exchange. You will find yourself asking, silently, what would I want the first three sentences of this company to be if I could hear them fresh. That is the effect the exercise has on the owner. It gives her back the ear she lost while she was busy running the business. And an owner who has her ear back can hear things in her own company, and in the wider market shifting around it, that no dashboard and no assistant is ever going to bring to her attention on time.

The signal you send by paying attention to openings

There is one further effect of this practice that only shows up after a few months and is worth naming. A team eventually notices which parts of the work the owner reads carefully. Openings are one of the parts owners are least likely to inspect, and the team knows it. When the owner starts to inspect openings, the team’s attention to openings rises within a week. Not because they were being careless before. Because attention flows toward the parts of the work the owner treats as important, and the openings were, until then, treated as filler. The moment they stop being treated as filler by the owner, they stop being treated as filler by anyone.

That effect is why the audit is worth doing publicly, at least in part. Not to shame anyone. To communicate, without a memo, that the openings now count. A quiet mention in a weekly review that you have been reading first three sentences across the business and have noticed a pattern is enough. The team internalizes the standard before the pattern is even fully described, because they understand what an owner-visible surface has come to be. That transfer of attention, from owner to team, is one of the cleanest ways to raise quality across a whole function without changing a single process document.

A short instruction, for this week

Before you close this tab, pick a first-contact exchange from your own inbox that happened in the last seven days. Read only the first three sentences your side sent. Say them out loud. Ask yourself, without softening the answer, whether those three sentences are the ones you would have wanted a stranger to receive from your company. If the answer is yes, note the person who wrote them and thank her by name this week. If the answer is no, do not correct her yet. Ask yourself what would need to change, in the standard, in the tool, or in the training, for the answer to be yes next week. That single loop, run honestly, is the smallest possible piece of quality work an owner can do on a Monday morning, and it will change more in the company by Friday than any offsite you have ever planned.