Ace Data-Driven Marketing

Abhay Gupta

Sep 1, 2024

There are infinite ways you can do your marketing.

Infinite ways to design an ad. Infinite social media posts you could share.

So how do you know what to actually do?

On top of that, how can you determine what works and what doesn't?

"If I try everything, eventually something will work."

True. So we're done here.

Well, unless you really want to try everything.

Don't have the time? So let's speed things up a bit.

The most important thing is that there is an aspect of trial and error, so this process will still take some time.

How much time?

That can vary depending on several factors (niche, location, marketing strategy etc).

It can be anywhere from about a week to a few months.

So why use trial and error if it still takes a long time?

It's to find a marketing strategy that works.

Then, focus your efforts on the things that do work and scale them up.

You'll save a lot of time in the long run.

"How do I know if something works?"

Determining what works is not subjective.

To put it simply, "What is measured, improves." You might recognise this as Pearson's Law.

So how do you apply it?

Let's break this down into 3 steps.

1) Data Collection

How are prospects interacting with your marketing? Are they paying enough attention? Are they following CTAs to get in touch with you?

In the case of digital marketing, data collection is mostly done for you. Cookies, views, clicks and so on.

Sometimes, you will need to do this manually. The easiest way is to directly ask customers questions. For example, how did they find out about you?

2) Metric Calculations

Now we need to analyse it, calculating metrics such as cost per acquisition (CPA), return on ad spend (ROAS) and so on.

But how do you know if your value for a metric is good?

Firstly, you need to understand what the metric actually represents.

You also need to know the ideal case and determine how far away from the ideal case your numbers are.

I should also mention that the ideal case might not be possible. There will always be some "inefficiencies".

3) Comparison

Repeat 1) and 2) for a bunch of marketing strategies and cross-compare the results.

For example, say I have strategies A and B.

A has a CPA of £5. B has a CPA of £8.

CPA measures how much it costs to attract one customer.

The ideal CPA is £0.

A's CPA is closer to the ideal £0 than B's. So A is better than B.

"Ok, so I'll just do this for every metric on Earth."

Generally, the more metrics you use the better. But don't get carried away.

What if you have too many metrics?

You want it to be as easy as possible for you to make a decision.

If you use too many metrics, you're going to be overwhelmed with information.

You'll end up with decision paralysis... not ideal.

I would suggest an absolute minimum of 3 and no more than 10. But this will depend on what your trying to monitor.

The important thing is that you understand the metrics you are using, so you can make a proper decision.

So, let's start making data-driven decisions.

Test out a bunch of marketing strategies.

Scale up the ones that do well according to your chosen metrics.

Focus on the relevant metrics - don't become decision paralysed.

Talk Soon,
Abhay