There is no shortage of lists of cognitive biases. Most of them are catalogues. A catalogue is a way to feel educated without changing a decision. I keep a short list because these patterns show up in real marketing work, on my side of the table as much as on the reader's. The point is to notice them. Collecting names is a hobby.
A bias, here, is a reliable lean that arrives before the argument. That is why this note sits next to decisions happen before the argument. If the lean is invisible, you will explain it later with a story that flatters you. The story will be fluent. It will also be late.
Name the lean while it is happening. A named bias after the launch is a eulogy, not a tool.
The few I actually watch
I do not try to remember forty. I watch for these, because I have seen them spend money.
Confirmation. You already like the channel, so the week produces evidence that the channel is working. Awkward comments get called anecdotes. A friendly reply gets called signal. The cure is to write the disconfirming result before you look. What would count as "this did not work"? If you cannot say it, you will recognize only the success.
Availability. The last campaign, the loudest customer, the tactic a peer just posted. It feels like the market because you can picture it. The quiet customers, who never write, are less available and often more typical. The cure is a boring sample: the last ten conversations, not the one you tell at dinner.
Anchoring. The first number in the room becomes the topic. A competitor's price, a vanity metric, a goal invented in January. Later discussion edits around the anchor instead of asking whether the anchor was the right object. The cure is to ask what decision the number is for, before debating the number.
Loss aversion, including inside the company. Killing a message feels like losing the work that made it. Keeping a weak page feels safer than the blank of a refusal. The loss you are avoiding is often smaller than the loss of a year of confused readers. The cure is to price the kept thing: what does it cost to leave this sentence up?
Social proof as a substitute for a point of view. "Everyone in the category says this" is not a reason. It is a fear of being the one who refuses. Sometimes the category is wrong together. The cure is the refusal: who you will not copy, even if the copy would feel companionable.
| Lean | How it shows up on a marketing team | A check before you ship |
|---|---|---|
| Confirmation | The dashboard is read for good news | Write the failure in advance |
| Availability | The loudest customer writes the page | Use the last ten notes, not the best story |
| Anchoring | The first target becomes the strategy | Ask what decision the number serves |
| Loss aversion | Weak work stays because killing it feels wasteful | Name the cost of leaving it live |
| Social proof | The category's sentence becomes yours | Say who you will disappoint on purpose |
Why a longer list makes you worse
A long list creates a new bias: the feeling that naming a bias is the same as being free of it. You can say "that's availability" and still obey it. The name becomes a badge of sophistication. Sophistication is not a decision. I would rather apply five checks badly and in public than recite thirty and change nothing.
There is also a cruelty available in this material. Biases are often taught as levers to pull on other people. A page that "uses scarcity" because a list said scarcity works has skipped the question of whether the scarcity is true. If it is not true, you did not use psychology. You lied, with a citation. I am not interested in a menu of techniques. Robert Cialdini's book is in the library as a set of forces to notice, not to deploy. The reading note is Influence. Notice when a page is pressing. Decide whether the pressure is honest.
Rory Sutherland's point, in the note on Alchemy, belongs here too. A purely rational plan can miss the lean that actually decides. Studying biases is not an invitation to be irrational on purpose. It is a warning that the rational memo arrives after the lean. Plan for the lean. Do not sneer at it, and do not fake it.
On the reader's side
Readers bring the same leans. They anchor on the first price they see. They treat a familiar logo as safety. They fear losing a current tool more than they desire a better one. You can write with that in mind without tricking them. An honest anchor is a clear price. An honest treatment of loss is to say what they will stop doing, not only what they will gain. Pretending there is no loss is how promises curdle.
The reader's confirmation bias will finish your vague sentence for you. If the sentence is fog, they will pour in whatever they hoped you were. That feels like agreement in a meeting and like mismatch after the sale. Specificity is how you stop them completing your thought in a way you cannot deliver. Vague copy is not neutral. It is a Rorschach test you will be held to.
Availability works on them too. The case study you repeat becomes the whole company in their memory. Choose the story that is typical, not the one that is cinematic. A cinematic outlier teaches the market to expect the outlier. Then ordinary work looks like failure.
A weekly practice
I do not run a bias workshop. I add one line to the weekly review: which lean did I obey? The answer is often availability. I let a single vivid conversation set the week's priority. Writing that down does not make me pure. It makes the next week slightly harder to fool, because I can see last week's method.
Pair the line with a decision. "I obeyed availability, so next week the priority comes from the last ten notes, not from Tuesday's call." A bias note without a change in method is a diary of self-awareness. Self-awareness is pleasant. Method is what the customer meets.
If you want one study session, take a live page and mark it. Where is the first number, and did it deserve to be first? Where did you keep a sentence because deleting it felt like loss? Where did you sound like the category because sounding different felt lonely? Those three marks are worth more than a poster of forty icons.
Study the leans that already spend your budget. Leave the catalogue on the shelf until one of these five stops surprising you. It will not stop. That is the reason to keep the list short enough to actually use.