testing

At Some Point, You Have to Launch

by
Rocket launching from a website interface surrounded by design tweaks and launch notes

Hello!

My last post was about how just because something works doesn’t necessarily mean it’s ready.

That’s true.

But there’s a problem on the other side too.

At some point, you have to launch.

“Ready” Can Become a Moving Target

I’ve been working on my new project for quite a while now.

And lately, I’ve noticed how easy it is to keep finding things to improve.

Rewrite this sentence.

Move that button.

Adjust the padding on that box.

Make this section a little wider.

Change the wording again.

Then change it back.

I wish I were exaggerating.

There is almost no end to it.

And the tricky part is that most of those changes really do make things a little better.

That’s what makes it so easy to keep going.

You aren’t necessarily wasting time.

You’re improving things.

But eventually there comes a point where another hour spent adjusting padding probably isn’t going to teach you as much as letting a hundred real people use the site.

Real Users Change Everything

I can click through every page.

I can test different screen sizes.

I can create test accounts.

I can try to break things.

But there’s one problem.

I already know how everything is supposed to work.

That changes the way I use the site.

A brand-new member doesn’t know what I expect them to click next.

They may completely miss something I thought was obvious.

They may misunderstand wording I thought was perfectly clear.

They may use a feature in a way I never considered.

They may find some strange combination of clicks that I never thought to test.

And once a few hundred real people start moving through the site, I fully expect them to uncover a whole new list of things that need fixing.

Honestly, I’d be surprised if they didn’t.

That doesn’t mean the launch failed.

It means I finally have information I couldn’t get by testing everything myself.

How Do You Know When to Launch?

Before launch, a lot of improvement is based on prediction.

Will people understand this?

Will they notice that button?

Will this page make sense?

Will they know what to do next?

You can make educated guesses.

You can test.

You can stare at the page for the 50th time.

But eventually you’ve learned about as much as you can from testing it yourself.

After launch, the questions start answering themselves.

Instead of wondering whether people understand something, you can see whether they actually do.

Instead of guessing which features matter, you can see what people use.

Instead of polishing things nobody notices, you can spend your time fixing the things people actually struggle with.

That seems like a much better use of effort.

And Yes, I Still Haven’t Launched

There is some obvious irony in me writing this.

The project I’ve been talking about still isn’t live.

I’m close.

Very close.

And there are still a few things I want to finish before I open the doors.

I think they’re worth doing.

But I’m also becoming very aware that there will never be a morning when I look through the entire site and think:

Perfect. There is absolutely nothing left I could improve.

That day isn’t coming.

There will always be another sentence I could rewrite.

Another button I could move.

Another little bit of padding I could adjust.

Eventually, the next useful step isn’t another tweak.

It’s letting people in.

Because you don’t stop improving a product when you launch it.

You just finally start improving it with real users.

After 20+ Years Online, I Think We Quit Too Soon

by
Online marketer at a crossroads choosing steady progress over constantly starting over

Hello!

One thing I’ve noticed after more than 20 years online is that marketers are really good at starting things.

Sticking with them?

That’s a little harder.

There’s always something new competing for our attention.

A new program. A new traffic source. A new page. A new strategy.

And lately, of course, a new AI tool every five minutes.

I’m certainly not immune to it.

But the longer I do this, the more I think constantly moving on to the next thing can keep us from learning what actually works.

There’s a Difference Between Testing and Starting Over

I test things all the time.

But testing only teaches you something if you give the test a chance to produce useful information.

Let’s say you send an email and don’t get the response you hoped for.

Next time you change the subject line, rewrite the email, promote a different page, and use a different traffic source.

Maybe the next mailing does better.

Great.

But what did you learn?

Was it the subject line?

The email?

The page?

The audience?

You really don’t know.

Sometimes the boring approach—doing something consistently and paying attention—is where the useful information comes from.

My Safelist Experiment Reminded Me of This

My Safelist Attention Project started with a pretty simple question:

Are people actually paying attention to the pages they’re clicking?

I could have run the experiment for a few days, looked at the first batch of numbers, and moved on.

Instead, I kept going.

That’s when it got interesting.

I started seeing differences between traffic volume and attention.

Some smaller traffic sources produced surprisingly strong engagement.

Repeated exposure started showing patterns of its own.

And eventually I was looking at things I hadn’t even thought to measure when I started.

None of that would have happened if I had looked at the first few hundred visits and decided I already had my answer.

Sometimes People Need to See Something Again

That experiment also changed the way I think about advertising.

It’s easy to assume that if somebody sees an ad and doesn’t click, the ad failed.

But people don’t always respond the first time.

Or the second.

Sometimes something has to become familiar before it becomes interesting.

I explored that more in How Many Times Do People See an Ad Before They Actually Click?.

That doesn’t mean you should run a bad ad forever.

It just means that replacing something because it didn’t produce an immediate result may be a little premature.

Persistence Isn’t the Same as Being Stubborn

Of course, some things really don’t work.

Some programs die.

Some traffic sources stop producing.

Some offers just aren’t very good.

There’s nothing wrong with walking away when the evidence tells you to.

But I think there’s a big difference between quitting because something isn’t working…

…and quitting because something newer caught your attention.

I’ve probably done plenty of both over the years.

These days, I try to know which one I’m doing.

Give It Enough Time to Teach You Something

When I try something now, I’m less interested in whether it immediately hits a home run.

I want to know what happened.

Who responded?

What didn’t happen?

Did anything surprise me?

Would the result look different if I kept going?

You can’t answer many of those questions if you keep changing everything.

That may be one of the biggest differences between how I approached Internet marketing when I started and how I approach it today.

I don’t feel nearly as much pressure to chase every new thing.

Sometimes the better move is to keep doing what you’re already doing…

and pay closer attention.

Thanks for reading,

Jerry

P.S. Before you abandon something you’re working on, ask yourself one question:

Has it really failed, or have I just not given it enough time to teach me anything yet?