I'm on the sleeper train from London to Scotland,

and there's not much room in here.

Tonight, this train is going to roll through the English Midlands

and up through northern England, and while we're on our way,

I'd like to tell you the story of one of the biggest online hits I've ever been involved in.

The north-south divide is an important thing in English culture.

If you're northern, you're supposed to be proud of it:

industry, school of hard knocks, saying 'bath' instead of 'baath'.

And if you're southern, you're supposed to be proud of…

not being northern, I guess?

You'll see it in the northerners who move down to London, in the south, for jobs.

Complaining about the price of everything down there,

about how you can't decent chips-and-gravy,

and about how for some reason the locals insist on wearing a big coat outside when it's cold.

I mostly grew up in the Midlands, so both sides hate me.

And I brought my big coat.

My point is: the idea of being 'northern' or 'not northern'

is relatable for a lot of the people in this country.

In 2013, I was working for a website called 'us vs th3m',

and my job was to make web toys and games quickly.

So when one of our team came up with an idea for a 'How Northern Are You' quiz,

the whole office jumped on it.

Occasionally, you might get an idea that arrives fully formed in your head,

like a bolt from the blue,

a complete plan unfolding in your brain.

Generally, I get those ideas when I'm in the shower,

when I'm not really thinking about anything,

when my mind's got time to start putting together everything I've been thinking about

over the last day.

It's usually a really good plan to act on those ideas

if you're lucky enough to get one of them.

But as far as I can remember, this was not one of those ideas.

We workshopped this one in the office.

All the team suggested questions; we tested them on the folks around us.

And rather than just using a regular question-and-answer system for the quiz,

I coded up a full-screen game called the “North-O-Meter”

that would give you a big arrow that swung north and south

depending on the answers to your questions.

Another one of the team suggested that, as well as giving you a percentage,

it should give you a result that places you somewhere in the country,

in a town or a city.

It was maybe a day's work to put that all together.

Most importantly: the game had a big button that let you share your results.

Now, I should point out that we weren't doing any data harvesting here.

Unlike certain other companies,

we weren't asking people to log into Facebook and share their details with us,

we were just using the standard, simple sharing template

that just posted a thing to your wall.

Our mission was just to get people to visit our site, to visit our projects,

to pay attention to us.

Our job was, literally, to try and be popular on the internet,

and asking people to log in would have got completely in the way of that.

So it was simple.

See your friends, on Facebook or Twitter or whatever, share their result,

find out your own, share it onwards. That was our goal.

We knew we'd got a hit pretty quickly, but the numbers kept rising, and rising, and rising.

At one point, we had 13,000 people playing this at the same time,

and it only took a minute to play it.

By four days in, it had been shared on Facebook a million times.

Not clicked. Shared.

A million people had decided that they were going to click the button

to share their results to their friends.

We had made something relatable.

No matter if you were northern or southern, proud of it or not proud of it,

that quiz gave you something that you were probably willing to share.

Now it helped that we'd spent a while calibrating the questions and answers,

bouncing cultural references and results back and forth until everyone

had something they were at least happy with.

But lots of people told us that their result had identified the actual city they lived in,

which was basically just a lucky guess, but somehow it seemed better than that,

for some people it seemed almost spooky.

And anyway, if it gave you a result you didn't like,

if it was too southern or too northern or wrong,

that would still cause a reaction,

that was still fun enough that you might want to share the results.

We estimated that, even if they hadn't clicked through a link,

the name of it had still been seen by about 10% of the British population.

It was the biggest thing we'd ever made, and, if I remember right,

it was the biggest thing we would ever make.

The secret to success there was connecting with people:

provide something they're interested in and give them a reason to show it to others.

That is the same reason that people put faces in YouTube thumbnails.

Human connections,

the desire to see and be seen,

that's what powers sharing online.

Around the same time as the North-O-Meter went out to the world,

my YouTube channel was starting to see some traffic.

For years, I'd been ignoring YouTube's advice about building a channel,

about regular uploads and brand identity,

and I'd just used it as a dumping ground for any of my projects that happened to include video.

Usually, they were projects without my face in them.

I was just an anonymous British voice in the background,

but in the year or two before that, I'd started appearing in vision.

And I'd been doing some videos with a channel called Computerphile,

who were already doing some pretty good numbers,

and they were happy to link their audience to my channel.

Collaborating with others has to be mutual,

but that isn't just about subscriber counts or about money:

it's about what you can provide.

For Computerphile, I provided the expertise,

and they provided the audience.

Soon after that, I'd start a little series called Things You Might Not Know,

and that's where it really started to take off.

People connect with people, and now I was on screen presenting,

now I was putting out regular videos,

now there were lots of similar things to click on and a reason to come back,

now that I was collaborating with others,

now people realised that this YouTube channel, this thing here,

wasn't just a thing, it was a person,

people were starting to subscribe.

The first thousand fans, subscribers, whatever, they're the hardest ones to get.

Back when I was getting started 15 years ago, it was a case of emailing the right folks,

of showing up in the right blogs and newsletters.

And it was a case of persistence and perseverance, of grinding your way through.

It still is. At least, I think. The catch is,

I don't know what the right techniques and the right places are any more,

I haven't known for years because I'm out of touch with that now.

And what those places are will be different depending on who you are, where you are,

when you're watching this, and what you're making.

But as you make stuff, you will find out.

You'll bump into someone who'll say, oh,

'I should send that there', or,

oh, 'I like that, I'll pass it on'.

It's about human connections.

Not spam, not cold-calling: human connections.

I'd like to believe that everyone out there already has some connections,

some people who know you.

A message board you post on, people you've talked to in a comment thread,

a WhatsApp chat you're in, a Facebook group with friends.

Those are your first steps, those are your first contacts, and yes,

it absolutely sucks that there are going to be genius-level pieces of work

that aren't making it out of those early groups --

and mediocre work that gets lots of praise because it's from someone well established.

I'm not saying that's right or fair, and we should work to change that,

but that's the way things are now.

There's still a long way to go.

Let's not forget that a good part of the reason that Things You Might Not Know took off

is that people tend to trust a white guy with a British accent.

Anyway.

The site I worked for back in 2013, 'us vs th3m',

rode high for a long time on the North-o-Meter

and a lot of similar projects made by a lot of different people.

Our Twitter and Facebook and email subscriber counts continued to tick upwards,

and the team expanded: we gained more staff members,

and a second full-time developer in the office.

But while we had some great ideas and great traffic,

sustainable growth was tricky.

Suddenly there were a lot of social media quizzes out there.

I'm not saying we were entirely responsible for the fad of all those

“What Magical Unicorn Are You” quizzes that hit Facebook in 2014,

but we were certainly in the vanguard.

We were popular on the internet.

And the question that everyone has to ask for any project, though, is:

how long can it last?

Because 'us vs th3m' survived for a couple of years,

which was much longer than the initial experiment that it was designed to be,

but the site was sadly never able to pay its way in the

really harsh and expensive world of corporate media.

And even now, I'm aware that my current big project, all this YouTube channel,

could go away at any moment.

So how do you go from just throwing ideas out there to making it work long-term?

Part of the solution to that is up at my destination.

That was too loud.