Understanding customers

Your customers are trying to tell you something

How to treat your store analytics as evidence, run useful experiments and learn what your customers want.

“The only way to win is to learn faster than anyone else.”

— Eric Ries

There is a fairly common way to run a small business.

Post something on Instagram. Hope people see it. Post again. Get a few more followers. Maybe boost a post. Someone buys something. Great. Do more of… whatever it was that caused that.

This works.

Sort of.

The problem is that you don't really know why it worked.

And if you don't know why something worked, it's quite difficult to do more of it.

Fortunately, the internet leaves clues.

Your store is basically an experiment

Imagine you post a picture of your new cupcakes on WhatsApp and 100 people visit your Earn+ store.

Ten look at the chocolate cupcakes.

Four add them to their cart.

Two actually buy.

That is considerably more useful information than “my post got 37 likes.”

You now know something about what happened between somebody hearing about your business and somebody giving you money.

Marketers call this a funnel.

At the top are all the people who encounter your business. Some visit your store. Some look at a product. A smaller number add something to their cart. An even smaller number check out.

Your Earn+ analytics show activity along this journey: visitors, storefront views, product views, add-to-cart actions, checkout starts and paid orders.

These totals are clues, not a neat line of the same people moving through every step. One visitor can view several products or come back later, so the numbers will not always get smaller.

Your job is to look for patterns in where interest turns into action — and where it seems to stop.

1,000 followers aren't necessarily better than 100

Let's say you have 1,000 Instagram followers and 20 people visit your store this week.

Your friend has 200 followers and 40 people visit hers.

Which audience brought more people to the store this week?

Your friend's. A smaller audience brought twice as many visitors.

Followers still matter. But the follower count alone can't tell you what's working.

Analytics helps fill in the gap: how many people reached your store, what they looked at and whether they bought.

Then comes the next question: where did those visitors come from?

Where did everybody come from?

Now things get more interesting.

Your Earn+ analytics don't just tell you how many people visited your store. Where source information is available, they also give you clues about how people found it.

Perhaps 30 came from Instagram.

Another 20 clicked your WhatsApp link.

Ten came through Earn+.

Five came from somewhere else.

Using your store link and the results from your Analytics page helps you build a clearer picture of where people find you, whether it's on Instagram, WhatsApp or somewhere else.

They also help answer a more useful question:

Where should I spend my time?

If you spent three hours making Instagram content and it brought two visitors, while one WhatsApp status brought 30, that is useful information.

It doesn't mean Instagram is useless.

It means you have learned something.

And now you can test it again.

Welcome to the extremely sophisticated world of experimentation

“Running a marketing experiment” sounds like something that requires a marketing degree, several dashboards and somebody who uses the phrase brand activation.

It doesn't.

An experiment is basically:

I think X will cause Y. Let's try it and see.

For example:

I think posting my products in three local Facebook groups will bring 50 people to my store this week.

That's a perfectly good experiment.

Share the full link from your Store dashboard. Note where and when you posted it. Wait a week.

Then look.

Maybe 60 people arrive.

Promising. Try it again and see whether the pattern holds.

Maybe three people arrive.

Also useful. Perhaps Facebook groups aren't the answer. Or perhaps the post was bad. Change something and try again.

Compare similar time periods and change one thing at a time. A rise in visits is a clue, not proof that every visitor came from that post.

The important bit is that you are no longer guessing blindly.

Find the leak

Traffic isn't the whole story either.

Suppose 200 people visit your store and nobody buys anything.

Getting another 1,000 visitors might not be your biggest problem.

There may be something worth improving inside the store.

Look at the activity in your analytics.

Are people visiting but not viewing products?

Maybe your store isn't immediately clear.

Are they viewing one product much more than everything else?

Interesting. Your customers may be telling you what they actually want.

Are lots of people viewing a product but very few adding it to their cart?

Perhaps the price, photographs, description or offer needs work.

Are people adding products to their cart but not checking out?

Now we have a different problem.

This is what makes the funnel useful.

It helps turn:

“Nobody is buying 😭”

into:

“This product gets views, but few add-to-cart actions. What might be getting in the way?”

The second question gives you something specific to investigate. Ask customers too; the numbers alone cannot tell you the reason.

Your customers vote with clicks

One of my favourite things about analytics is that customers leave clues even when they don't tell you what they think.

They show you what they do. Conversations help you understand why.

If one product gets five times more views than everything else, that's a clue.

If people suddenly start arriving from WhatsApp after you change the way you post there, that's a clue.

If your visitors double but sales don't move, that's a clue too.

None of these numbers gives you the answer.

Analytics isn't magic.

It is evidence.

And the more experiments you run, the more evidence you collect.

So try something

This week, pick one thing.

Not ten.

One.

Maybe you'll post your store in a Facebook group. Ask ten friends to share your link. Put your best-selling product on your WhatsApp status every morning. Try a different Instagram post. Run a special offer.

Before you do it, write down what you think will happen.

Then come back to your Earn+ analytics and see what actually happened.

How many people visited?

Where did they come from?

What did they look at?

What did they add to their carts?

Did anybody buy?

Then change something and go again.

Because building a business is, to a surprisingly large extent, just this process repeated over and over:

Have an idea. Try it. Measure it. Learn something. Try again.

The businesses that get good at that have a rather unfair advantage over the ones that are still guessing.

Put the clues to work

See what your customers did next.

Open your analytics, choose one useful signal and decide what you want to test next.

Open your store analytics