Know your customers?

What is a customer really worth to your business? Figuring out this question is the cornerstone of most successful enterprises. If you know the future value of a customer over the next 3 years is $1,000 you'll likeley be more than happy to pay $100 dollars to convert them to a customer today (for example through marketing or incentives). But Life Time Value or LTV for short is oftentimes a mystic part of the business coming from the depths of the analytics or finance department. This blog is about shedding light on this sort of topics. Knowledge is power. Truth is power. Dumi-truth. As we were...

What is Life Time Value

To be or L-T-V? as Shakespeare often wondered
Some call it CLV. Some call it LTV some all it LCV. WUT. Let's define it here broadly as all the financial value your business will derive over a predefined time horizon from the business interactions with a customer.

There are many businesses that will not need to think about this issue. If your product is either a very large item, for example you are selling luxury, custom made airplanes, you'll likely not get a lot of repeat purchases. 

On the other hand if you are in the category of Fast Moving Consumer Goods (like toothpaste), your customer relationship is probably quite fickle so you'd likely be better of in thinking of aggregate trends rather than granular life time value.

Types of businesses that should really know about LTV

Any subscription business - e.g. Spotify or Netflix - would need to get a really good grasp on that. We will go into the components of LTV for such a business in the next paragraph.

Any financial services providers such as insurance companies or credit card companies need to know the value of a customer really really well. This is generally true of super competitive markets where the margins are really low.

Another great example are e-commerce websites. You get to sign-up a user for a cost of X. If you can get revenue greater than X from that user over their life-time, you get profit.

The Peter Drucker quote goes "what gets measured gets managed". I don't have a snarky twist here. It's true (and probably the only quote I like). If you can measure your cost of acquiring a customer and the value you can get from that customer, you're managing your bottom line.

*Drops mic*

All customers are created equal

Except for the ones that aren't. Which is most of them. [so basically the title is clickbait]

It's extremely likely that your customers will not behave the same at all. The things of interest here are:
  • What are they buying out of your product lineup?
  • How much of the stuff are they buying?
  • How often are they buying?
  • How long will they be doing that?

When I say "buying" I don't necessarily mean a financial transaction. If you're Facebook and you're selling ad-space to publishers your "customers" will also be your regular users and you'll care about engagement metrics such as how often do people log in and how much time do they spent on our platform. More time spent, more ads you can show them and more charged to businesses trying to advertise. It's all about selling eyeballs. The ad industry has some strange fixation with eyeballs.

The most contentious part of this is the time horizon. If I asked you for a million dollars ca$h today and GUARANTEED to give you back $2 million over the next million years that is not a great deal for you. Or if it seems like a great deal, please get in touch with me.

How long is forever?

Not that long. You usually have real world constraints such as cash flow if you are a bootstrapping startup or EBITDA goals if you are a supergiant Evil Corp. 

So, in thinking about Life Time Value you will likely want to ask how much money will this customer bring in the next 3 to 12 months. If you have VC mulla and can burn through cash like crazy you might be tempted to take a 5 to 10 years value of your customers. That's very unwise as the market usually changes. Fast. And you just don't know what that will look like. If you're a VC and someone pitches you a 10 year LTV, run. Fast.

Ok so let's take an example. You have a product that costs $100 with a profit margin before acquisition cost of 60%. Your customers buy it on average twice a month. Over a year that's $1440. Your acquisition cost is $500. Profit.

Now there is a fair bit of intuition and business judgement to be made. If your understanding of the market and priors (beliefs) make you think customer will be on average with you for a year, that is great. But ideally you'd want to have some industry data or even better your past data. Now, the future is always uncertain and the quality of new customers might change over time. This is where common sense is important.

What if I had more money than I could spend?

You don't. As I mentioned in the past paragraph, at the end of the day it boils down to cash-flow/ebitda goals/appetite for risk 

Understand risk. Embrace it. And treat it like any type of financial investment for your business.

Averages are bullshit 

As reddit would say, "the average human has one breast and one testicle". Averages are great big picture but if you are actually trying to bootstrap/be competitive averages are bullshit.

You know it is true. The distribution of users is much more important. If you don't get that right you'll get a lot of "potential customers" for a great price but they will never convert at the same rate as your "average customer". Cheap users from click farms  are not worth $1440. They are worth $0.

So you will want to understand the demographics of your acquisition - who are they, do they have enough disposable income to afford your product? And if so, how likely are they to buy it? Are there any triggers/events you can use to predict future conversion? Think time spent on your ecommerce site or the channel they came from (for example for insurance brokers inbound customers are much more likely to convert so you might want to incentivize them a bit).

Don't be fooled by averages. Understand your "types" of users.

Bad things good people do

Short term metrics and long term metrics ARE correlated very very strongly. So if you see no effect on your direct business metrics such as number of sales, the likelihood you will re-coup your investment down the line through "increased awareness" is close to 0. People forget. Fast. And everyone is shouting as loud as they can, asking for attention. So yeah, good luck with that.

Also, don't burn through your audiences. If you don't see a business response it means the product market fit was bad. Improve that first then reach them again. Otherwise you'll either piss them off or lose their interest.


Ok, I think I got it out.

So how do I do this?

3 simple steps:

1. Understand your user types
2. Understand their value in a reasonable time horizon
3. Use common sense.

Would you be interested in using a free tool to calculate LTV built by me? There are other tools out there but they are either not user friendly or miss important nuances.  Please comment in the section below to let me know or drop me a message on twitter @dumitruth. If there is enough interest I will put a tool up next week. Thanks for reading!

My name is Dan and that was the Dumitruth.