Revenue, Income and Return on Investments

Business analysts study various business metrics, such as gross, operating, net profit and margin.

Turnover, or gross revenue, is the amount of money customers paid the company.

Prime cost is the money the company paid to acquire goods.

gross profit. Gross profit = turnover - prime cost. The ratio of gross profit to revenue is called the gross profit margin (or gross margin). Gross margin = gross profit / revenue.

Operating expenses are the company's expenses on core activities. If you subtract operating expenses from gross profit, you'll get operating profit. Operating profit = gross profit - operating expenses.

Operating profit gives a picture of overall business performance by indicating how much a company earns from its core activities. Calculating operating profit is easier and less time-consuming. If operating profit is negative, there's an operating loss but, as with negative gross profit, an operating loss doesn't mean that the jig is up. From time to time companies have planned unprofitability, so that they can invest all of their revenue in fast development and growth.

Dividing operating profit by revenue, you'll get the operating margin. This is the share of revenue that remains in the company after prime costs, salaries, rent, marketing, and other core-activity expenses have been paid. Operating margin = operating profit / revenue

Net profit is the amount of money business owners can take for themselves or reinvest to develop the company. Net profit = operating income - taxes and loans. A negative net profit is called a net loss. Much like other types of losses, it signals that the company failed to make money.

The most important metric for investors is ROI (return on investment). ROI = (net profit - investments) / investments Return on investment can be calculated not only for a business as a whole, but also for its components. **ROI of ad campaign = (income - expenses) / expenses.**

In order to distinguish between payoff for the business and payoff for an ad campaign, ROMI (return on marketing investment) is calculated instead of ROI. ROMI of ad campaign = campaign's gross profit / expenses.

Conversions

The sales process is essentially the process of turning a visitor into a buyer.

The more users are converted from visitors to buyers, the better things are going. The conversion rate is the share of people who change their status.

Funnels

A funnel illustrates:

1. the path a user takes to buy a product

2. the share of users that move on to the next stage

To plot a funnel, you first have to find the number of people that reach each stage. Knowing the number of users at each stage allows you to determine the share of those that reach a certain stage. You can also determine the share of users that complete every subsequent step.

Funnel analysis helps you ask the right questions, formulate hypotheses, test them, and track changes.

The Marketing Funnel: Impressions, Clicks, CTR, and CR

Potential customers sometimes click on advertising banners and get redirected to a landing page, which has a registration form that users fill out if they're interested in the product.

The goal is to get as many registered users as possible. Let's plot a funnel showing user activity for the landing page. We'll need to know:

• How many times the banner was shown (impressions)
• How many users clicked on it
• How many of them navigated to the landing page
• How many registrations occurred

Data on impressions and clicks can be retrieved from advertising systems. The click-through rate (CTR) is the ratio of clicks to impressions. It's a percentage:

Conversion from clicks to sign-ups is called the conversion rate (CR):

The Marketing Funnel: Grouping by Weeks and Months

When data is noisy, it's hard to find patterns. So sometimes you'll have to consolidate it by, for example, grouping it by week. Here's how we do it:

1. Add a week column to the DataFrame
2. Group the DataFrame by this column
3. Sum up the number of impressions, clicks, and registrations for each week
4. Calculate the mean weekly conversion rate

We can do something analogous for months.

Simple Product Funnels

Product funnels show you how users interact with your websites and apps.

It's the information used to plot the funnels that makes them different. As a rule, marketing data is aggregated. It presents the number of impressions, clicks, and registrations per day. It looks like the results of a survey.

But product data is usually "raw." Each table row is a single event and looks more like an individual survey form.

The easiest way to create a funnel is to calculate how many times each of the events occurred. For instance, we group the data into one DataFrame on the event_name field and find the number of rows. To find the number of unique users in each group (since each user can perform the same action multiple times), we call nunique (number of unique elements) when aggregating.

Product Funnels with Sequence of Events

Sometimes analysts need to study details of the transition from one stage to another. To do so, for each user, they first have to find the exact time a particular step was taken for the first time. Then they find those who took steps in the desired order.