Marketing Analytics

One of the most common uses for analytics is marketing. In this this article you will get an overview of the most common uses and methods used in marketing.

---

Knowing the Customer

Knowing about your customer is pretty important. When you understand the customer it helps you to tailor products and service that they want to buy. You do this by collecting information from them in a non-creepy way. Some companies like to follow people around the internet using tracking methods. It's called surveillance capitalism. Don't be those guys.

The correct way is when the customer opts-in voluntarily. Examples of this may be an email or text sign up, contact forms on website or an email from them requesting information or service. They explicitly expressed interest.

Also keep in mind of PII which is short for Personally Identifiable Information. There are many standards that establish how you handle this data depending on your industry, location and business size. Some of these are:

• PCI for those who accept card payments
• GDPR for those in the EU
• HIPPA for those in the medical field
• CCPA for those in California

There are others, these are just a few examples. Check with a licensed attorney for details on how these apply to your situation.

Surveys

Surveys are also a great way to collect information from customers. You see many big companies do this. Fill out our survey get a free whatever on your next visit. Your goal with that is to get some get demographics about your customer base and you are giving the customer a discount in exchange for that.

Purchase Patterns

Depending upon the business type you may or may not know who bought what. For example:

• Cash based businesses like food trucks may not know who bought
• Invoice based businesses know exactly who bought

If you are group 2 that helps. But even if you are group 1 and don't know who bought what you can still examine purchase patterns. You should be asking key questions. For example:

• What did you sell?
• How much did you sell?
• When did you sell it?
• How did you sell it?

You should be tracking this. A simple database is a good start but a data warehouse is ideal. Good sources of information are your POS system, CRM and customer service systems, if you have them.

Methods

There are a variety of methods that can be used. The most common ones are listed below. We can utilize a variety of technologies to do these things such as databases, visualization tool and machine learning. As this is an overview article, I'm not going to dig into technical details.

Customer Churn

Customer churn is when someone stops doing business with you for a variety of reason. It could be your service was awful and they left, it could be your competitor offer a better price, it could be you are local and they moved.

Just a side note on price, if you are competing only on the lowest price you are going to lose, even if you get the business.

"The problem with a race to the bottom is you might win"
~ Seth Godin

You should be looking at customer churn and see who is likely to churn. It is a good idea to have some version of automation or machine learning helping your do this. But you can't do this if you haven't collected data like we discussed above.

What is an indicator or churn is based on your individual situation but you should be looking out for customers likely to churn. Then do something to cause that not to happen. Such as an offer, discount, etc. But some people will always leave.

Customer Value

Some customers are worth more than others. They buy more are less of a hassle to serve, don't complain, etc. You want more or these people. Start making notes of what these people have in common and what they buy. You can use automated systems to do this but don't completely disregard human intelligence and intuition in the process. Which leads into the next item.

You may also want to calculate customer value. Essentially take sales / customers. That is your average value per customer. Then apply that to a given timeframe.

Customer Profiles

Customer profiles take your customer base and segment them into groups. The number of groups is up to you, but you should be doing this. For two reasons:

• You can offer different things to different groups. This helps you optimize your marketing spend.
• You want to know what the best customers look like, so you can find more of them.

Recommendations

You even been on an online store and you are looking at a product and they say something like we think you would also like X, whatever X may be? It seems like a magic trick. It is not. It is based on data and it is called a recommendation system.

Recommendation systems use two primary methodologies:

• Look at the current user and find users similar to them and see what they bought.
• Look at the product and find items similar to that item.

Just a bit of inside baseball these use processes called collaborative filtering and basket analysis. They are common in E-Commerce and media streaming. However, you don't have to be in these fields to use them. You many look at your customer base and have a system that will recommend who else of my customer base would be interested in product or service X.

---

Conclusions

To summarize you should be collecting data in a secure, legal and ethical manner. Then use that data to analyze customer churn, create profiles, calculate value and make recommendations.

The ultimate goal is to serve people and you can't service people if you don't know anything about them or what they want.