Pattern

Opening signal

Thursday, 5:40 p.m. A client calls and says a sentence you know by heart: “look, I know it’s late, but this has to be today.” You stay until nine. On Friday morning you are tired, and you push back the thing you actually wanted to do — the one with no deadline, because it is yours.

On Saturday you think of it as bad luck. A difficult client. A bad week.

But look back through three months of your calendar. You will find four Thursdays like that. Different names, the same hour, the same phrase, the same result: Friday morning eaten, your own work postponed. That is no longer bad luck. That is a pattern — with the difference that bad luck is innocent, while a pattern has a cause, and a cause can be found.

Most of the people I know do not have a problem with a lack of information about their own lives. They have a problem with resolution. We see a single event in high resolution — we remember the details, the tone of voice, the specific request. But we do not see the intervals between events. And the pattern lives in the intervals.

The cost is expensive in a very simple way: we solve the same event over and over, each time from scratch. We pay the full price in attention for something we only needed to pay for once. This is the point at which work starts to resemble noise — not because there is too much of it, but because it has never been counted.

This chapter is about one skill: noticing repetition before it becomes a cost. Not about predicting the future and not about looking for signs — about reading your own records more carefully than they are usually read. Because almost everything that surprises us in our own working life has surprised us before. It just was not written down.

Source

The hermetic tradition. The whole tradition rests on a single claim: the levels of reality correspond to one another. In the Corpus Hermeticum it reads: “the cosmos is an image of God, and man an image of the cosmos” (CH X.11). Behind it stands the Stoic sympatheia, the assumption that the world is one connected whole, in which what happens at one level resonates at another.

What is not there. There is no ancient table in which a stone corresponds to a planet, a planet to an angel, and an angel to a letter. Those tables come from Agrippa (De occulta philosophia, 1533), from the Order of the Golden Dawn and from Crowley’s 777 — they are Renaissance and modern, not Egyptian. The very category of “correspondences” as a named doctrine of esotericism is Antoine Faivre’s scholarly construct from 1994. What is ancient is the intuition, not the inventory. Nor is there anything in these texts resembling a diagnosis of recurrence in working life.

Modern sources. The capacity to detect patterns has also been described from the side of its failures. In 1958 Klaus Conrad introduced the term apophenia for seeing meaningful connections where there are none. In 1971 Amos Tversky and Daniel Kahneman described the “law of small numbers” — the tendency to treat a small sample as representative, which makes three events in a row look like a rule. To this add the clustering illusion, the perception of clusters in random data, and survivorship bias: we look at the cases that survived into our field of view and draw conclusions from them about the whole.

Limits. These two bodies of material do not confirm each other and share no authority. The first says that the world has structure; the second, that our apparatus for detecting it fails in specific, predictable ways. We set them side by side because both pieces of information are needed, not because they form a single claim.

Our reading

Our position begins with a sentence that translates no source: chaos is a pattern at too low a resolution. Not all chaos — some events really are singular: accidents, coincidences, other people’s decisions you have no influence over. But a great deal of what we call chaos has structure. We simply have not seen it yet, because we were looking from too close.

In practice this means three levels of seeing the same thing.

• Event — client X called on Thursday at 5:40 p.m.
• Pattern — clients call with urgent requests towards the end of the week, and I usually agree.
• Structure — I have no fixed moment for closing the week, so Thursday afternoon is an open door; and since I price hours rather than the problem, every extra request looks cheap to the client.

Intervening at the level of the event changes nothing. You can refuse once, feel relief, and have exactly the same Thursday two weeks later. Intervening at the level of the pattern helps a little. Only intervention at the level of structure ends the matter — and it is usually surprisingly small: one sentence in a contract, one hour in the calendar, one change in how work is handed over.

An event can be survived. A pattern has to be named. A structure has to be changed.

From the hermetic claim about correspondence we take one thing and put it in our own words — this is our sentence, not a translation of CH X: a pattern rarely appears at only one level of life. If at work you cannot close a project and drag it out with revisions for two weeks past the deadline, then almost certainly there is somewhere else you cannot close either: a conversation, a day, an argument, an evening. If you do not set a boundary with a client, check who else you are not setting one with. Not because the cosmos sends signals. Because it is the same person with the same repertoire of reflexes, only in different settings.

That gives a diagnostic method: when a pattern is hard to see at work because it is padded with professional justifications (“that’s the industry”, “that’s the client”), look for its twin where you have no ready excuse. There it is usually visible to the naked eye.

And the attitude we inherit from this tradition most willingly: assume that what returns has a reason, and that the reason can be found. That assumption is self-fulfilling in a good sense — people who do not hold it simply do not look. But the modern half of the sources places a condition on it that must not be removed: the same apparatus that finds real patterns also produces false ones, and it does so with the same sense of obviousness. That is why in this chapter everything is counted and written down rather than recognised by intuition.

Operational translation

Patterns do not live in memory, they live in records. Memory is partial: it keeps the dramas and loses the repetitions. So reading patterns starts with the places that record themselves.

Where to actually look

• The calendar. Not what you planned, but what happened. Which meetings get moved? Which hours are regularly eaten? How many times a month has a block for your own work been given up to somebody else’s urgency?
• Inbox and chat. Not the content, the recurring sentences. If a fifth client writes “and could we just add”, that is not a feature of the fifth client — it is a description of your scope of work.
• Invoices and payment dates. The most dispassionate record of reality you have. Who pays late? Do those jobs share a feature — a discount, no deposit, a project sold through an acquaintance, no clear moment of acceptance?
• The places where projects stall. In most studios I know, delay is not spread evenly. It gathers at one point: at approval, at the client delivering materials, at the last ten per cent. That point is always the same, and almost nobody has ever named it.

How to tell a pattern from a coincidence

Three signals, and all of them have to appear together:

• Repetition — it has happened at least three times within a reasonable window of time.
• The same phrasing — people describe it in almost identical words. This is the strongest of the signals. When a fifth person says of your offer “but this is probably for bigger companies”, you have a problem with the offer, not with five people.
• The same reaction — you feel the same thing: the same tightness, the same irritation, the same relief. Here the emotion is a useful marker, not the subject.

Three patterns that show up most often

The scope pattern. A project ends later than it should, and always for the same reason: the end was never defined. Nothing is missing in skill or in time. What is missing is the sentence “at this stage we deliver this and this; further changes are a separate job”.

The intake pattern. The worst clients arrive through the same channel. Usually it is the channel that produces a quick yes without a conversation about the problem — a referral with no context, an enquiry with a price in the first sentence. If three difficult working relationships came in the same way, the way has to be redesigned, not the people blamed.

The postponement pattern. There is one category of task you always push back. For most of the people I have asked about it, it is the task that requires exposure: sending the proposal, naming a price, chasing a payment, publishing something of your own. This is not a problem with time management. It is one pattern impersonating a dozen separate arrears.

Embodied check

The body is the earliest recorder of patterns you have, and also the one that is read least often. The calendar shows the repetition after a quarter. The neck shows it the second time.

So it is worth treating a few things as records rather than complaints: when exactly the neck or the jaw starts to hurt, after what kind of week the sleep problems come back, on which evenings that particular pressure before Monday appears, after which conversations you eat differently than you otherwise would. These are data with dates. If for three months you do not sleep on Sunday nights, you have information about the structure of your week, not about your sleep.

The body can also be faster than judgement at recognising that something is repeating. The familiar tightening at the sight of a particular name in the inbox is the first signal of a pattern — it arrives before you can remember that with this person it has ended the same way three times. That signal is worth writing down and checking against the records rather than believing straight away: it is a marker, not proof. The feeling that “something is coming back here” can also be apophenia.

Recognising your state before deciding applies here in one specific place. A pattern review carried out in a state of exhaustion always gives the same result: everything is a pattern, everything is your fault, everything has to change. That is not sharper vision, only darker. You do not do the review after a sleepless night, or after a day when something fell apart — you do it in an average state, because only then are the proportions visible.

Practice

Pattern review — a seven-day experiment

A classic pattern review is done once a quarter. Here we need a version that can be run in a week — so the gathering happens on the first day, and over the following five you test the hypothesis live.

What you need: the last three months of your calendar, a list of closed and open projects, your invoices, and a sheet of paper. You do not need a tool. Paper is an advantage here, because it forces you to be brief.

• Day 1 — gathering (20 minutes). Go through three months and write out the events that cost you unplanned energy: overtime, your own work postponed, difficult conversations, revisions after the deadline, days taken out of circulation. The list only, no commentary. Usually it comes to between fifteen and thirty items.
• Day 1 — grouping (10 minutes). Put the things that rhyme into families. There is one criterion and it is strict: can it be described in the same sentence. If you have to stretch, they are two different things. Usually thirty items leave three to five families. Choose one — the one that came back most often, not necessarily the one that hurts most — and write it down as a hypothesis, in a single sentence.
• Days 2–6 — live testing, two minutes a day. Each evening you answer one question: did my hypothesis hold today or not, and how do I know. You write down the literal sentences that were said, and the times. Five days is too little for proof, but it is enough to see whether you are even looking in the right place.
• Day 7 — review (30 minutes). Four questions, in writing. How many times did it come back, and in what words? What exactly happens just before it — what is the trigger? What does it cost me over a quarter, in hours or in money; estimate roughly, but do count. And the deciding question: is this a pattern to change, or a parameter of mine to respect?
• One structural move. Write one specific sentence about what you are changing in the structure — not in your resolutions. Structure is: a clause in a contract, an hour in the calendar, a change in a form, a different moment in the process, someone else responsible. A resolution is: “I will be firmer.” The first works, the second does not.

A worked example. Family: “urgent requests at the end of the week”. Came back four times, always with the sentence “I know it’s late, but”. Trigger: no established moment at which I tell the client the week is closed. Cost: about fourteen hours a quarter plus two of my own deadlines pushed. This is a pattern to change. The move: on Thursday at three I send a summary of the week saying what goes into next week — one hour in the calendar, recurring, with a message template.

What this practice does not promise. It will not remove patterns whose source lies outside you — the market, somebody else’s illness, a client’s decision to freeze the budget. It does not replace therapy where the pattern is deep and old; an operational review sees the mechanics, not the history. And it will not work in a single pass: the first review is usually disappointing, because only the obvious things show up in it. The third shows what you did not want to see — which is why this weekly experiment is worth repeating next quarter rather than extending indefinitely.

Limits

Not every pattern exists. This is the most serious limit of this chapter, and it follows directly from the section on sources: the apparatus you are searching with produces false hits with the same sense of obviousness as true ones. Three events in a row are not yet a rule, and the most convincing patterns are usually the ones that confirm something you already believed. Hence the requirement to count and to write things down, and hence the question worth asking yourself at the end of every review: what would have to happen for me to conclude that this pattern is not there?

Do not interpret a pattern straight away. The great temptation is to jump from observation to a diagnosis of character — “because I’m chaotic”, “because I have a problem with assertiveness”. Such sentences sound like insight and work like closing the subject: there is nothing you can do with them on Monday. Stay with a description of the mechanics for as long as you can. Mechanics can be changed; character is an excuse.

Not every pattern is to be removed. Some patterns describe your limits, not your mistakes. If after two days of intense conversations you need a day of quiet, that is not a fault to be repaired — it is a parameter to plan the rest around. Reading patterns serves exactly this distinction: what is to be changed, and what is to be respected.

Not everything requires action. Not everything that can be changed should be changed.

Not everything in life is meant to be legible. This method is good for work, money and obligations. It is bad for friendship, curiosity, play, work made without a purpose, and evenings that do not have to lead anywhere. Anyone who turns every part of their life into a family of patterns with a quarterly cost will end up with a life fully counted and entirely without the places where something might happen for no reason. Part of life has a right to remain opaque, and that is not neglect.

Ethical question

Reading patterns works on people too. That is a real advantage and it has to be said plainly. Whoever recognises the recurrences in the behaviour of a client, a partner or a colleague knows at which point the other person will give way, what they are afraid of, and which sentence will move them. The same skill that lets you redesign your own process lets you lead another person somewhere they would not go on their own.

Influence and manipulation. The line runs exactly where it ran in the previous chapters: at openness. If you can tell the other side what you have noticed and how you are using it — “I see that by the third round of revisions you usually let things go, so I want to settle this in writing now” — that is influence, and it usually turns out well for both sides. If the mechanism has to stay unnamed in order to work, it is manipulation, however favourable the outcome.

People are not their patterns. Describing someone by a single repeated behaviour is convenient and almost always unfair: it closes off their route to change in your eyes before they have tried. A pattern is a description of a situation, not a definition of a person — and if your diagnosis of someone’s pattern means you stop listening to them, it is no longer a diagnosis but a verdict.

The cost to me. A life organised around detecting repetition can become a life in which nothing is new any more. Every so often it is worth deliberately not reading the pattern and letting something simply happen.

Agency increases responsibility. The more you can do, the more carefully you must choose.

Closing formula

Recognising a pattern is not yet change. It is the condition of change — and, more importantly, the condition of change being cheap. Whoever does not read the pattern pays the full price every time, as if it were the first. Whoever reads it pays once and moves on.

The next chapter is about what you do with a pattern once you can see it: how to turn recurring chaos into a system that runs without you.

What returns has a reason. What has a reason can be changed.