What a 46,057-impression post taught me about marketing

Ninety times my average, from two minutes of typing

I post on LinkedIn regularly and my posts get, on average, around 500 impressions. That is my normal. Then one post did 46,057. I did not boost it, I did not tag anyone, I did not post at a magic hour. I wrote it in under two minutes, on my phone, still slightly annoyed.

When I opened the analytics I found the interesting part. Only 6 per cent of the people who saw it were in my network. The other 94 per cent were strangers — the platform pushed the post to more than 30,000 members I had never interacted with. It produced 191 profile views and 12 new followers, from a post that contained no advice, no tip and no lesson. Just what happened at the airport.

I have spent years writing careful posts with useful frameworks and neat conclusions, and none of them came within a tenth of this. So I did what anyone in marketing should do when reality embarrasses their theory: I looked at what the outlier had that the others did not. Below is the story, and then the seven things I took from it. Some of them are about LinkedIn. Most of them are about how brands talk to people.

The story: a passport with five months and twenty-nine days

It was a one-day work trip to Cairo, and Murphy's law arrived early. Online check-in was down. The self-service kiosks were down. So I queued at the counter, where the agent looked at my passport and told me it had less than six months of validity — five months and twenty-nine days — and that I would have to renew it before flying.

I went to the passport office inside the airport. The officer there told me the opposite: it is valid, you are fine by two days, go and fly, and do not renew here because renewing at the airport costs double when the passport still has more than six months. Two counters, thirty metres apart, two confident and opposite answers.

Back at check-in, I opened an AI assistant on my phone and had it count the days precisely, then showed the agent the screen. She mocked the AI. Not the calculation — the idea that I would trust a chatbot over her. The clock was running and the plane was not going to wait for us to settle the philosophy of machine intelligence, so I agreed to whatever was needed and caught the flight.

That is the whole post. No moral. I did not win the argument. Nobody learned anything except that airports are airports. And 46,000 people read it.

Tell what happened, let someone else be the character, keep the lesson inside

1. Tell something that actually happened, not something you know. The feed is full of people explaining things. It is not full of people who were at a counter at six in the morning being told two opposite facts. People are exhausted by advice. They have never once been exhausted by a story. Every piece of brand content we now produce at Nplay starts with the question “what actually happened?” before “what do we want to say?”.

2. You do not have to be the hero. The check-in agent made that post, not me. She had the best line — mocking the AI — and I was just the person standing there with a phone. When a company tells a story, it almost always casts itself as the hero, and that is exactly the moment the audience stops believing. Let the customer, the technician or the visitor be the character. The brand can hold the camera.

3. Keep the lesson inside the story and in the comments. I did not explain what the story meant. The comments did, in about twenty different directions, and every one of those comments carried the post to a new group of people. When you hand people the analysis, you close the conversation. When you leave space, they open it for you.

Do not preach like Aristotle, and let the ending be true

4. Do not preach. There is a temptation, particularly for those of us who run companies, to end every story with “and the moral is…”. Aristotle could get away with it. On a phone screen it reads as a lecture, and people leave lectures. If the story is good, the moral is already in the room. If the story is weak, the moral will not save it.

5. The ending should be true, not heroic. I gave in. I agreed to whatever the agent wanted and I caught my flight. I did not produce a devastating argument, the airport did not change its policy, and I did not fly home with a lesson for humanity. That ending is the reason the post worked. People recognised it because it is how their own mornings end.

Compare that with the average brand case study, where every project finishes with a smiling client and a percentage. We write case studies too — see the memorial hologram project — and the ones people actually read are the ones that admit the dome projectors that had to be taken down and relocated, the content that went through several review rounds with the client, the cabling routes nobody had drawn before our technicians surveyed them. True endings are more persuasive than triumphant ones because they are the only kind the reader has ever lived through.

Shoot with your phone — and admit you do not fully know why it worked

6. Shoot with your phone. The feed is now wall-to-wall AI-generated images and polished designs, every one of them perfect and every one of them identical. An ordinary photo, slightly crooked, taken by a tired man at a counter, is what stops the thumb. This sounds like a LinkedIn tip but it is actually a brand tip. We run a 3D content studio and we make holograms for a living, so believe me when I say that polish has a place — on the exhibition stand, in the product film. In the feed, the honest phone photo of the hologram being unloaded from the van beats the render of the hologram every single time.

7. Maybe a keyword got lucky. I am not a growth expert and I do not pretend to understand the algorithm. It is entirely possible that some accidental word in the post — “AI”, “passport”, “Cairo” — matched something the platform wanted to promote that week. I am not going to build a theory on one data point. These are simply the numbers that were in front of me, and I would rather show you the numbers than sell you a formula.

What this means when the storyteller is a company

I run a company that sells immersive technology, so I have a stake in the answer to an obvious question: does any of this apply when the author is a brand rather than a slightly annoyed man at an airport? I think it applies more, not less, because brands start from a deficit of trust.

Here is how it has changed our own work. When we plan content for a client through our marketing services, we now ask for incidents before we ask for messages. What happened at the last exhibition? Which visitor asked the AI avatar a question nobody expected? Which screen went dark ten minutes before the opening and how did the technician fix it? Those are the posts. The message hides inside them.

It has also changed how we think about the physical experiences we build. A hologram on a stand is not just a display; it is an incident. Visitors photograph it on their own phones, badly lit, and tell the story to people who were not there. That is a 94-per-cent-outside-your-network moment, generated by the stand rather than by the marketing department. We design for it deliberately now: something on the stand should be worth telling someone about at dinner.

And we try not to be the hero. The visitor is the hero. The technician is the hero. We hold the camera.

A one-week challenge

I ended the original post with a challenge and I will end this one the same way. For one week, write only about situations that actually happened to you. No advice at the end. No “here is what I learned”. No framework. A photo from your phone if there is one. Then tell me what happened to your numbers.

If you are a brand, do the same with your company page: one week of incidents instead of announcements. If it goes badly, you have lost a week. If it goes the way mine did, you will have learned more about your audience than a year of careful posting taught me.

And if your incident happened in front of a hologram, a video wall or a virtual exhibition stand we built, send it to us. We collect them.