What ROI tells you
Return on investment, usually shortened to ROI, compares what you gained from an activity with what it cost. In marketing, it answers a simple question: for the money and effort we put in, how much came back?
It is a useful summary, but only as good as the inputs. Before the formula, it helps to be clear that ROI is an estimate, and that marketing rarely lets you draw a perfectly clean line from spend to result.
The basic formula
ROI is commonly expressed as a percentage:
ROI = (revenue - cost) / cost x 100
Here, revenue is the income you attribute to the activity, and cost is what you spent on it.
An illustrative example
These numbers are invented to show the arithmetic, not typical results.
A small campaign costs 100 and brings in revenue of 250.
- Revenue minus cost: 250 - 100 = 150
- Divide by cost: 150 / 100 = 1.5
- Multiply by 100: 1.5 x 100 = 150 percent
So the ROI is 150 percent. Put another way, every 1 spent produced 1.50 more than it cost, on top of getting the 1 back.
Compare this with a similar-sounding figure: revenue divided by cost, which here is 250 / 100 = 2.5. People sometimes call this a return multiple, and in paid advertising a related measure called ROAS (return on ad spend) works this way. It is not the same as ROI, because it ignores the cost you already paid. When someone quotes a figure, check which one they mean.
Revenue versus profit
The example above uses revenue. That can overstate the picture, because the product itself has costs. A more careful version uses profit, or gross margin, in place of revenue.
Suppose, again as an illustration, that the 250 of sales came from goods that cost 150 to supply. Profit before marketing is 250 - 150 = 100. Subtract the campaign cost of 100 and nothing is left: the campaign broke even. The ROI would be 0 percent, a very different story from 150 percent. Which version is right depends on what you are trying to learn, but you should know which one you are using.
What to count as cost
Be honest about the full cost, not only the invoice from the advertising platform. Possible items include:
- Ad spend or sponsorship fees
- Software or tool subscriptions used for the activity
- Fees paid to freelancers or agencies
- Content production costs, such as photography or video
- Staff time, even if it is hard to price exactly
Leaving out time and tools makes ROI look better than it is. If you cannot price everything, at least list what you left out so the result is read with that in mind.
What to count as revenue
Count only what the activity plausibly produced. Useful questions include:
- Would this sale have happened anyway, for example from a regular customer who happened to click an ad?
- Does the sale include later repeat purchases? A single order may understate a customer’s worth. Customer lifetime value looks at the longer view, though any projection is uncertain.
- Were there returns, refunds or discounts that reduce the real revenue?
The attribution problem
Attribution is the question of which marketing activity deserves credit for a sale. It is the hardest part of ROI.
A typical customer might see a social post, search for your name a week later, read a blog article, then click an email and buy. If your reporting credits only the last click, the email gets everything and earlier steps appear worthless. If it credits only the first touch, the opposite happens. Different models give different answers, and none is perfect.
Other complications:
- Time lags. Some marketing, such as content and brand work, pays off months later, so a short window understates it.
- Offline effects. A visitor may phone or walk into a shop after seeing an ad online, and your analytics may not capture it.
- Overlap. Several channels run at once and influence each other. Separating their effects precisely is often impossible.
- Privacy limits. Tracking is restricted by consent choices, browser settings and local laws, which vary by country and change over time. Check the rules that apply to you; this is not legal advice. Part of the journey may simply be invisible to you.
Because of these issues, treat ROI as a guide for comparison and direction, not as a precise fact. Understanding the customer journey can help you see which touchpoints to give credit to.
Using ROI sensibly
- Define the period. State whether you are measuring one month, one campaign or a full year.
- Compare like with like. Using the same method across channels lets you rank them fairly, even if the absolute numbers are imperfect. Cost per acquisition is another comparison that works well for paid channels.
- Do not chase a single number. A channel with a modest ROI may still feed others, or build an audience that pays off later.
- Combine with other measures. Conversion rates, repeat purchase rates and customer feedback add context.
- Start with what is measurable. If you cannot track everything, a rough but consistent estimate usually beats none.
- Remember uncertainty. A good ROI last quarter does not promise the same next quarter. Results vary with season, competition and many other factors.
Used with these cautions, ROI is a helpful way to ask whether your spending is doing its job. It works best as the start of a conversation about what to try next, rather than the final verdict.