Data fundamentals – Make the most of your data

by | Aug 2, 2019

The value of data has recently surpassed the value of oil¹ and every new piece of software or online tool comes with its own suite of data analytics in some form or another which means the average business has more data than they can accurately use.

Whilst it’s a good thing to have lots of useful data points, for a business without the resources to manage it properly it can become a burden. That, combined with not knowing what to look at and asking vague, indirect questions will be a significant blocker.

In this article I’m going to outline an easy framework for asking the right questions to get the most benefit from your data.

A good question should be made up of three components:


  1. Target segment.
  2. Detail.
  3. Action.


The Target Segment. The question will look at a specific subcategory or subset of data, such as a specific marketing channel or user group. i.e, email users.

The Detail. The question will include as much detail as required to make the question as specific as possible, such as the metric you’d like to track. I.e new users which made a purchase.

The Action. The question will be specified in such a way that the answer will give you insights to take action. I.e, marketing channel performance this week compared with last week.

Let’s look at an example of a bad question:

“How many sessions does our website get?”


Why is it bad?

Knowing the number of sessions on a site isn’t necessarily a bad thing, but pitched in this way will mean the answer will serve as nothing more than a vanity metric². The output will give you a single figure and nothing to compare it against.


How to make it better

“How many organic sessions resulted in new leads generated this year, compared to last year?”


Why is it better?

It takes into account the framework from above, and the output will tell you how your organic channel has been performing relative to last year. The data gained from this will tell you if you need to investigate why there has been a decline, or praise your SEO team for the work they’ve been doing.


Let’s break it down…

How many organic sessions resulted in a new leads generated this year, compared to last year?

In this case our target is organic sessions, as it’s focused on a specific area from where traffic is coming from. Our action is new leads generated and the detail is the time frame comparison between this year and last year.


Let’s take a look at another example:

“How many enquiries did we get?”


Why is it bad?

Like the bad example above, the answer to this question doesn’t give you any real insight into the health of the business or your product. If the answer is 40, is that good? Is that bad? Data exists to help drive decision making in businesses and is your most powerful tool if approached correctly³.


How to make it better

Which marketing channel drove the most enquiries within the duration of the latest marketing campaign?


Why is it better?

The answer to this question will inform you of the best performing channel within the time frame and will aid future marketing campaigns as you know where you can maximise ROI. It will have the benefit of helping you analyse weaknesses and inefficiencies in the latest campaign.


Let’s break it down…

Which marketing channel drove the most enquiries within the duration of the latest marketing campaign?

In this case the target segment we’ve defined is the duration of the latest marketing campaign (because we’re looking at marketing channels holistically) our detail is the most enquiries generated by the marketing channel, and the action is the marketing channel (So we can nail down the performance of the individual channels).

Rounding it all up 

This framework should help you avoid a lot of trial and error. You’ll be able to get actionable insights from the get-go and save time trying to interpret your data without running the risk of making vast assumptions or generalisations. It will take a few goes to perfect asking the right questions, but with experience you will save time and get more accurate answers.



1 – 

2 – 

3 –


Credit: Matthew Brandt, CXL

Author: Edd Saunders

Share Post

Related Posts

Talk to Us