Marketing is an important function in any company but essential for startups — especially in the scaling phase. But, what is marketing actually and how should you approach it?
Marketing in 2019
A well-known marketing guru Philip Kotler defines marketing as “the science and art of exploring, creating, and delivering value to satisfy the needs of a target market at a profit. Marketing identifies unfulfilled needs and desires. It defines, measures and quantifies the size of the identified market and the profit potential. It pinpoints which segments the company is capable of serving best and it designs and promotes the appropriate products and services”.
Kotler’s last sentence is important as it points out the need to define your desired customer segments. You cannot serve all the customers and on the other hand, not all the customers are profitable for you. Marketing is all about finding that perfect target customer — and then getting your solution in front of them.
Ideally, you are able to define your target customer and know where he or she can be targeted with your marketing messages. In reality, you often learn by testing. Even a lot of established marketers use the 70/20/10 rule i.e. spending 70 % of their marketing spend and efforts towards activities that currently perform well, 20 % for developing their marketing further and learning from it, and 10 % to completely new high-risk high-reward activities.
But in order to know whether your marketing activities have been successful and you have been able to find profitable customers you need to be able to measure the results. And depending on your business model this might be the part where it gets tricky.
Often the value of an individual customer varies. Think about an e-commerce store. One customer might buy a pair of socks for 3 euros and another one two jackets, three pairs of jeans and one pair of boots for 500 euros. It is clear which one of them was more valuable. Similarly, a lot of tech startups have business models that are based on subscriptions resulting in varying customer values based on how long the subscription lasts or what has been the chosen monthly plan. There are also freemium and premium models. E.g. in the gaming industry there are a lot of examples in which you can either play for free or pay for certain upgrades.
If not all of your customers are equally valuable to you it is important to evaluate your marketing activities not only based on whether they brought you customers but also on how valuable were those customers.
Putting it all together
In order to tell what is going on, you need a solid data model that connects the source of a new customer, linked to the marketing cost and estimate their lifetime value. All of this needs to connect to the other business data too, so eventually, you’ll want it in your data lake.
If your product has a landing page and customer sign up through it, you can use the so-called UTM-parameters to capture where the signups are coming from. You need to capture that, bring it over to your product upon signup and store it along with the account data. If you have a CRM system, it might have a way to capture this. You can always build a custom solution too, as we did at Zervant.
When you have an app and the first thing your customers do is download it, things do get a little trickier, but the principle is the same. Google allows tracking the path through the play store into your app with a referral parameter, which can be used to accomplish this. With iOS, the solutions rely on a method called fingerprinting, which isn’t as elegant or perfectly accurate, but definitely do-able. Bottom line is, the source data needs to connect to the created account in your data lake.
Figuring out the cost is easier. All you need is your marketing spend data at the right granularity, so you can allocate it to the different sources. The likes of Google, Facebook, and Bing all have APIs that will be able to provide this. This part is likely the most straightforward of all of them.
In the realm of your own data warehouse, estimating a user’s lifetime value should be simple enough. You are likely already estimating your customers’ lifetime values, so all you need is to adapt them to customer subsets, or develop methods of estimating them on a per-account basis.
What’s in it for you again?
Imagine if you can pinpoint exactly how valuable a specific group of users is and compare that to the effort it took from you to acquire them. You could treat the groups almost as different businesses and distinguish how profitable they are.
Once you’re able to evaluate your marketing efforts based on their ROI, you’ll have an accurate picture of where you are tapping into your segment: the kind of customers who stay with you month after month. The kind of customers your product was built for. Not the kind that is flimsy: quick to convert and leave you soon after. Stop wasting money on the wrong customers.
Ilkka Peltola is a seasoned data analytics professional currently transforming Zervant into a data driven company.