When I started my journey in the data world over seven years ago, I was convinced that the key to success was perfect technological mastery. I thought that if I learned the most difficult functions in SQL, wrote flawless Python code, and built advanced models, the business would automatically appreciate my work. However, the years spent working as a data architect brutally verified this approach. Yes, technology is the foundation, but for decision-makers in a company, code is just a means to an end. The true language of business is numbers, specifically Key Performance Indicators (KPIs).
In May of this year, I closed a certain chapter by leaving my full-time job to fully dedicate myself to the development of KajoData. This decision gave me an even broader perspective on analytics from a business owner’s point of view. When you have to manage a budget yourself, optimize costs, and plan the development of products such as KajoDataSpace or online courses, theoretical metrics become a hard reality that determines your survival in the market.
In today’s article, I want to translate the knowledge from my latest video material into a concrete, analytical guide. We will discuss five absolutely crucial indicators that every data analyst, especially those working in the e-commerce or SaaS industry, simply must know. We will look not only at dry definitions but, above all, at the mistakes most commonly made when interpreting them.
Customer Acquisition Cost (CAC), or How Much a Customer Costs Us
We start with the absolute baseline. CAC, or Customer Acquisition Cost, is a metric an analyst encounters almost every single day. The concept is very simple: it determines how much money we need to spend on marketing and sales efforts to get one new customer to make a purchase or create an account. In an ideal world, it would be enough to divide the monthly marketing budget by the number of new customers. In reality, however, the matter is much more complicated.
The most common mistake I see in reports is the complete mixing of acquisition channels. Imagine a situation where your company spends money on search engine ads, sponsored posts on social media, and email campaigns. The average CAC might look fine, but if you don’t break it down into individual sources, you will miss the fact that one channel is burning through the budget without bringing any conversions, while another is delivering great customers for a fraction of the price.
Another aspect is scaling. When a business grows and the decision is made to double the advertising budget, many managers assume that the number of acquired customers will also double, and the CAC will remain at the same level. As an analyst, you must make the business aware that reaching a broader, and therefore colder, audience (people who have had no previous contact with the brand) drastically increases the cost of convincing them to buy. Keeping CAC in check during aggressive scaling is one of the biggest challenges in e-commerce. Of course, the cost of advertising itself is not everything. You have to add the costs of maintaining tools, the work of the marketing team, and even creating graphic designs. Only then do we get a reliable picture of the situation.
Customer Lifetime Value (LTV), or the Long-Term Game
The customer acquisition cost itself, even calculated most accurately in the world, is useless if we do not compare it with what that customer means to us in the long run. Here enters Customer Lifetime Value, or LTV for short. Bluntly speaking, it is the sum of money we will extract from a customer during their entire relationship with our company. It may sound a bit cruel and purely transactional, but the truth is that the best and most stable business is a repeatable business.
There is a common misconception that LTV is an indicator reserved exclusively for subscription models. In the case of KajoDataSpace, the matter is obvious: the user pays a monthly access fee, so to calculate their value, we multiply the monthly fee by the expected number of months they will remain active. It works similarly for gym memberships or VOD services.
However, LTV can and should be calculated also for classic products, even in the FMCG sector. Take a carbonated beverage manufacturer as an example. If a consumer buys a can of soda regularly every week for several years, the brand – with the help of appropriate research and receipt analytics – can calculate their lifetime value. Knowing at what age the peak of consumption occurs allows them to precisely target ads at young people to build loyalty for decades.
For an analyst, the real magic happens when we combine CAC and LTV, creating the LTV:CAC ratio. This is the indicator that determines whether a business model makes sense. Let’s return to our example with advertising channels. Suppose you acquire a customer from channel A more expensively (higher CAC) and from channel B much cheaper. If you only look at costs, you will immediately turn off channel A. But if you check the LTV, it may turn out that customers from the expensive channel A are much better informed, understand the value of the product, and buy from you four times more often than those from channel B, who churn quickly. Without comparing these two metrics, you risk cutting off the most profitable source in the company.
Average Order Value (AOV), or the Art of the Shopping Cart
The third key indicator is the Average Order Value. It allows us to understand how much money the average customer leaves during a single transaction. While LTV looks at the entire life of the customer, AOV focuses on one specific moment at the checkout (or in the virtual cart).
Calculating this indicator seems trivially simple: you divide total revenue by the number of orders. The devil, however, is in the operational details. The biggest disruption to AOV comes from periodic promotions. In systems where heavy discounts appear regularly – for Black Friday, the New Year, or back-to-school – the average basket value can fluctuate drastically. As an analyst, you must be able to isolate the “real AOV,” characterizing normal sales days, from the distortions caused by aggressive discounting.
The real difficulty level appears when the price spread in the offer is huge. Imagine an online store where the cheapest accessories cost 5 PLN, and professional equipment costs 500 PLN. In such a model, a full cross-section of products falls into the basket. Tracking AOV and properly segmenting it can show the business a very clear path: which cheap, low-margin products should be discarded to minimize warehousing and packaging costs, and which more expensive items are worth scaling more aggressively. The average order value is also the basis for good service pricing and creating free shipping thresholds.
If you feel you need a solid technical foundation to navigate these metrics smoothly, know that this is exactly what analytical tools are for. If you want to dive deeper into the world of databases or visualizations and learn how to build dashboards that calculate all this automatically, take a look at my homepage. You will find comprehensive courses on Excel, SQL, Power Query, Power BI, Tableau, and Python environments. It is an excellent way to turn theoretical indicators into real analytical queries.
Cohort Retention Rate, or the Loyalty Test
We are moving on to an indicator that is slightly more difficult analytically, but serves as the foundation for companies operating in the subscription model and hardcore e-commerce. We are talking about the Cohort Retention Rate. A cohort, in our case, is a group of users who started using the service at the same time – usually, we analyze this on a monthly basis, for example, “January 2025” or “May 2026.”
In KajoDataSpace, this is one of the most important indicators I look at. Taking a closer look at a group that joined in a specific month, I check what percentage of those people continue to renew their access to materials and the community in the following months. Comparing different cohorts over time (e.g., January cohort vs. April cohort after six months) allows me to unequivocally assess whether the changes I introduce to the product actually bring value and keep users for longer.
It is extremely important here to distinguish between two concepts that confuse even experienced researchers: retention and purchase frequency. You might have a customer who, in the first month, buys fifteen different products from you. They caught the bug, fell in love with the brand, and order like crazy. However, after a month, they disappear completely and never come back. Although their initial purchase frequency was huge, retention turned out to be zero. From a business perspective, this means a failure in delivering long-term value.
Retention, or more precisely its inverse, the churn rate, determines whether a company is doomed to a constant, costly search for new customers. If the retention rate is high, the pain of spending money on marketing is much smaller because the business is driven by the loyalty of existing users. In the SaaS industry, where the costs of switching business software are very high, a low retention rate is an alarm screaming that the product has critical flaws. Since customers preferred to go through the painful migration process to the competition, it means your service was simply very bad.
Customer Payback Period, or the Race Against Time
The last indicator on our list is the Customer Payback Period. Unfortunately, in the daily rush of e-commerce, it is often confused with LTV, but it actually carries a completely different kind of information. The Customer Payback Period answers one critically important question: how long does it take for a given customer to break even?
In the era of aggressive marketing, companies can fall for the illusion of growth. They spend huge sums on acquiring a customer, believing in their product and knowing the LTV is high. The problem is that LTV tells us about the total profit over time, and CAC tells us about a one-time cost. We forget, however, about what happens along the way. The company is hit by operational, technological, and financial costs, taxes, and so on. From this whole equation, we have to extract the net margin.
Imagine a situation where the LTV is, for example, 2000 PLN over two years, and the CAC is 400 PLN. On paper, it sounds like a great deal. However, after calculating the net margin, it turns out that very little remains in the register from the customer’s monthly fee, meaning the customer must pay for the subscription continuously for six or seven months for the business to just break even on the acquisition cost. If the average retention rate in the company is four months, then despite a high theoretical LTV, the company is… losing money with every new customer and heading towards bankruptcy.
That is why the Payback Period is so essential. You must know when the break-even point occurs. Reaching the break-even point must happen relatively quickly; otherwise, you need massive capital and scale comparable to Netflix to bear the growing holes in cash flow. A perfectly tuned business is one where the payback period is short, and after it ends, the customer happily remains in the company’s ecosystem for years to come.
Summary
Working with data is a constant connecting of the dots. As analysts, we have the extraordinary privilege of looking under the hood of the entire enterprise. Mastering the formulas and definitions of indicators is only half the battle. The true value of an expert is the ability to see the dependencies between advertising spend (CAC), customer behavior over time (LTV and Retention), cart value (AOV), and the financial liquidity of the entire process (Payback Period). When you start talking about the business from the perspective of these intersecting forces, you will stop being just a report creator and become an irreplaceable strategic advisor.
I hope this text has helped you systematize your knowledge about the most important e-commerce indicators. If you found this article valuable, I would be very grateful if you shared it with your friends on LinkedIn, Facebook, or Twitter. Your support on social media helps me reach a wider audience of people who want to consciously develop their skills and build a career in the world of data analysis. See you in the next posts!
The article was written by Kajo Rudziński – analytical data architect, recognized expert in data analysis, creator of KajoData and polish community for analysts KajoDataSpace.
That’s all on this topic. Analyze in peace!
Did you like this article 🙂?
Share it on Social Media 📱
>>> You can share it on LinkedIn and show that you learn something new every day.
>>> You can throw it on Facebook – and perhaps help a friend of yours who is looking for this.
>>> And remember to bookmark this page, you never know if it won’t come handy in in the future.
You prefer to watch 📺 – no problem
>>> Subscribe and watch my English channel on YouTube.
Prefer to read in Polish? No problem!
Other interesting articles:
