On the surface, it may seem that calculating ROI is easy and you can't go wrong. After all, what could be difficult about the formula?

Return on investment - cost of investment ROI = ____________________________________________ x 100% (investment cost)

Nothing. If we count the ROI of a campaign, the profit from which was 2200 PLN, and which cost 500 PLN, we have simple mathematics, i.e. the result of 340% (or 3.4 - we will resign from multiplying by 100%), so nothing, but we can announce a spectacular success! However, if we take a closer look at the values included in the formula, it turns out that it is in this simplicity that the weakness of this indicator lies. Of course counting the return on investment (ROI), return on marketing investment (ROMI), or return on ad spend (ROAS) is useful. Especially if we want to calculate the return e.g. from a particular affiliate program or work on general data concerning marketing-mix of a given company. On the basis of ROMI it is possible to argue for further marketing expenditures, to settle accounts with conducted actions, or to illustrate the effects of work (every marketer knows that nothing is as convincing to the recipients as... numbers). For the management ROI and ROMI are also a convenience. It is enough, after all - assuming costs and risks - to establish their threshold, which is a positive number (and on this basis to assume that a given investment will pay off, and consequently to establish marketing investment strategies). And while calculating ROI, ROMI or ROAS itself is not troublesome, translating it into marketing (and business) practice is much more difficult.

First of all, when calculating ROI, it is important to consider what can actually be considered an increase in financial value for the company (returns on investment). Obviously, it's the financial value we've gained from a given campaign, i.e. sales revenue. It depends on us whether we are going to calculate it in general (for all sales channels together) or for each one separately (which is more recommended). Regardless of which option we choose, we will have to estimate the value that we would not achieve if we resign from a given campaign/activity. Usually the starting point would be sales data from e.g. one year ago. Most likely, however, last year we ran other marketing campaigns... As a result, we actually have no information completely "uncontaminated by marketing" to which we could refer. Of course, some solution could be to reach for older data, but working on it is always risky. In marketing, time runs faster. One, two, three - these are periods that are virtually incomparable. Three years ago content marketing or non-stock images were treated as an interesting novelty, today - due to Google algorithm changes - they are a must-have for every website. Two years ago, campaigns on Facebook did not require as much outlays as they do now (when after the change of the algorithm, all business fanpages record drops in the reach of published content). A year ago we did not spend a fortune on building a proper background for marketing communication with customers (which is now a necessity in light of RODO). A few months ago, the changes regarding the ban on Sunday trading seemed unrealistic, and today - it is necessary to run additional campaigns in order to slow down the decline in retail sales3 that the entire industry is experiencing.