As we know, marketing campaign may help businesses reach more customers and furthermore, increasing their revenue. But that’s only the case if we direct our campaigns towards the right direction. So today we will be doing an analysis for a company that have done a few campaigns to see how they are doing and see if we can find something to recommend to them.
Scenario
A company have done a few campaigns and now want to see how they are doing and see if there are any recommendations we can give. You can download the dataset here.
What does the data look like?
The data more or less look like this after cleaning (The query will be given at the bottom of this page).
We can see that we have a lot of columns to work with. But don’t worry, because we will be using Tableau to see the bigger picture of the data.
Questions
- Where do our customers come from?
We can see that most of our customers came from Spain, followed by Saudi Arabia and Canada.
2. What is the age of our customers?
To help us see the bigger picture, it’s easier for us to bin the age. And now we can see that our customers are mostly 41 years and older, while there is also a noticeable number of people in their 30s and some 20s.
Okay, now we can see that most of our customers come from Spain, Saudi Arabia, and Canada and is mostly older than 40 years old. So, let’s dig deeper into the campaign.
3. How successful was the campaign?
Before we answer this question, we need to answer this question first.
What defines a successful campaign?
Is it the total number of customers that accepted the offer? Or is it the revenue generated from it? For this case, let’s go with the former. Let’s say that a successful campaign means that we are able to make a certain number of customers accept our offer.
Now let’s focus on the overall stats of the campaign grouped by age. We can see that a lot of our target of age (>40 years old) are accepting our offer at least once. We successfully made about 19% of the age group of 40 years and older accepted our offer, which is amount to 361 people. So, on average, each campaign earns us about 60 conversions. That’s a not a bad number!
For now, let’s focus on people who are older than 40 years old so we can utilize what we have to gain more from our biggest group of customers.
4. Is there any preference of campaign types for each country?
It seems that our top 3 countries responded well to the sixth campaign and the opposite was the case for second campaign. For now, we might want to focus on our top 3 countries because the distributions are pretty much similar. We can compare the result later with all the countries involved (Maybe in part 2 :D).
The reason I choose to focus only on top 3 country is because the difference between top 3 country and the rest is just too big and might cause our analysis to become biased.
5. Is there any preference of products for each age group?
We can see from the graph that the age group of 40 years and older are very similar in product preference, where wines holds the first place with meats as the runner-up. While the least liked products are sweets and fruits.
Recommendations
After doing some digging into the data. We can conclude that most of our customers come from Spain, Saudi Arabia, and Canada with the age of 40 years and older.
In the end, we can confirm that there is a very similar pattern of preference for campaigns for our top 3 countries. Where campaign 6 was the most ‘achieving’ while campaign 2 was the least.
We can also see that for the age group of people with the age of 40 years and older have a very similar preferences of product. Where the most liked products were wines and meats, while the least liked were fruits and sweets.
Based on our analysis, we can give some recommendations:
- We might want to do campaigns similar to campaign 6 in the future for Spain, Saudi Arabia, and Canada
- We might want to find out why people didn’t respond very well to the campaign 2
- We might want to recommend more products related to wine and meat
- We might want to dig deeper to see why people older than 40 years old don’t buy our fruits and sweets
Bottom Line
While we have done some digging into the data and came up with some recommendations. I believe that it is wise to do some more data collections (like doing surveys, track more data, or some other methods) to back up what we suggested so that we can get a more accurate picture of what is really happening and what the company really needs.
That being said, this is not a thorough analysis of this dataset. There are still other things we can explore in this data, but for now I want to highlight what I could see after taking a quick glance of the data, while also considering our main topic, which is to see how the campaigns are going. I might continue the exploration of this dataset in the future; like asking more questions, involving more columns, and seeing correlations between columns (questions like: does higher income means higher chance of accepting campaigns?). I would be grateful if you have any feedback for me, see you!
Click here to download the query that are used in this post (I’m using SQL Server for this case)
