# Measuring marketing with a baseline > Why "the webinars returned twelve times their cost" is wrong: a baseline from the same months last year and a control group, incremental margin instead of revenue, and an honest ROMI. - Course: [B2B Marketing and Sales: Win Your First Company Customers](https://productovi.com/courses/b2b-marketing-and-sales.md) - Module: Building demand in a long sale - Lesson 15/32, ~24 min - Source: https://productovi.com/courses/b2b-marketing-and-sales/measuring-with-a-baseline ## Twelve times its cost? **Leo:** Great news for the board. The spring webinars for HR managers cost 20,000 UAH. In June to August new Teams clients bought 310 lessons: 248,000 UAH. Twelve times the cost! **Sofia:** 248,000 divided by 20,000 is 12.4. The arithmetic checks out. **Alex Tutor:** The division is right; the question is wrong. Two problems. First, revenue isn't what Klasna keeps: the tutor gets 430 of each 800 and video and AI cost 31. Klasna keeps 339 a lesson. **Alex Tutor:** Second, it credits every lesson to the webinars. Some of those clients would have come anyway, through referrals, the founders' contacts or plain luck. **Marta:** So the honest question is: how many lessons would we have sold without the webinars? > **Exercise:** Leo divided 248,000 UAH of new-client revenue by the 20,000 UAH the webinars cost. What is wrong with "twelve times the cost"? Choose all that apply. ## What would have happened anyway **Alex Tutor:** That number is the **baseline**: what the metric would have been without the campaign. You can't observe it directly, so you estimate it. **Alex Tutor:** Start with the same months last year. June to August 2025, new clients bought 200 lessons. Same months, so the summer dip is in both numbers. **Sofia:** So 310 minus 200: the webinars brought 110 extra lessons. **Alex Tutor:** Not yet. The market moved too. We promoted the webinars only in Kyiv and Lviv. In the other cities, with no webinars at all, new-client lessons grew 15% on last year. That's our **control group**. **Marta:** So without the webinars, Kyiv and Lviv would probably have grown 15% as well. **Alex Tutor:** Exactly. Baseline: 200 × 1.15 = 230 lessons. Only what's above that can be credited to marketing. > **Exercise:** New clients in Kyiv and Lviv bought **310** lessons in June–August 2026 and **200** in the same months of 2025. Cities without the webinars grew **15%**. How many lessons above the baseline can be credited to the webinars? ## From lessons to money **Alex Tutor:** Now turn the extra lessons into what Klasna keeps. **Incremental margin**: 80 × 339 = 27,120 UAH. **Alex Tutor:** Try that step yourself before we subtract the spend. > **Exercise:** The webinars brought **80** lessons above the baseline. Each lesson costs **800 UAH**; the tutor gets **430** and video and AI cost **31**. What is the incremental margin, in UAH? **Alex Tutor:** Subtract the spend: 27,120 − 20,000 = 7,120 UAH. That is what the webinars earned after paying for themselves. **Alex Tutor:** **ROMI**, return on marketing investment, divides that by the spend: 7,120 ÷ 20,000 = 35.6%. **Leo:** From twelve times to 35.6%. That hurts. **Marta:** It's still positive, and it's a number I can defend in front of the board. > **Exercise:** The webinars cost **20,000 UAH** and brought **80** lessons above the baseline. Contribution is **339 UAH** a lesson. What is the ROMI, in percent? ## Choosing a baseline **Sofia:** Which baseline is right: last year or the control cities? **Alex Tutor:** Use both if you can. Last year's same months handle the season. A control group handles the trend. It must be comparable and untouched by the campaign. **Alex Tutor:** Which control would you trust? > **Exercise:** Klasna plans an autumn campaign in some cities. Which control group gives the most trustworthy baseline? **Alex Tutor:** Two more cautions. In a long sale, results land months later, so measure when deals land, not in the campaign's month. And small numbers are noisy: 80 lessons could easily have been 50 or 110. **Alex Tutor:** One quick check on timing, true or false? > **Exercise:** **True or false:** in a B2B sale that takes months, the fairest way to judge a campaign is by the deals signed in the month it ran. **Alex Tutor:** For a stronger answer, pick the cities that get the campaign at random and plan the test in advance. The Product Analytics course shows how to set up and read such an experiment. > **Exercise:** Put the steps for measuring a marketing campaign in order. ## Telling Leo **Leo:** So should I tell the board the webinars were a waste? **Alex Tutor:** No, and no twelve times either. They earned about 7,000 UAH above their cost, which is a modest win on a small, noisy sample. Worth repeating as a planned test. **Sofia:** And I show how we got there: the baseline, the control, the margin, not just the final percentage. **Marta:** A board trusts a modest number with its working more than a big one without it. > **Exercise:** The spring webinars cost 20,000 UAH. New-client lessons in Kyiv and Lviv went from 200 to 310, against a baseline of 230 (cities without webinars grew 15%), and Klasna keeps 339 UAH a lesson. Which sentence for the board describes the webinars honestly? ## Your turn: a note to Leo **Leo:** One more. In September we sponsored an HR breakfast for 15,000 UAH. Same check, please, but in words I can repeat. **Marta:** The numbers: new-client lessons in October and November were 150, against 100 in the same months last year. The control cities grew 20%. **Alex Tutor:** Show each step: the baseline, the extra lessons, the margin, the spend, the ROMI. Then one sentence on what to do next. **Sofia:** Baseline first, then everything else follows. > **Exercise:** Write Leo a short note (4–6 sentences) on the HR breakfast: it cost **15,000 UAH**; new-client lessons in October–November were **150**, against **100** in the same months last year; cities without the breakfast grew **20%**; contribution is **339 UAH** a lesson. Give the baseline, the extra lessons, the incremental margin, the result after spend and the ROMI, then one recommendation. > **Key ideas** > > - Never credit all growth to marketing. Credit only what is above the **baseline**: what would have happened without the campaign. > - Build the baseline from the same months last year (the season) adjusted by a **control group** that did not get the campaign (the trend). > - Count margin, not revenue: **incremental margin** = extra lessons × 339 UAH; subtract the spend for the result. > - **ROMI** = (incremental margin − spend) ÷ spend. Report it with its working and its caveats: small numbers, results that land months later. > - For a stronger answer, choose the control group at random and plan the test in advance. **Alex Tutor:** Next: a whole quarter, five channels and 325,000 UAH. Which of it really moved, and where should next quarter's money go? - Previous lesson: [Outbound that respects consent](https://productovi.com/courses/b2b-marketing-and-sales/outbound-with-consent.md) - Next lesson: [Case: a quarter's budget and what really moved](https://productovi.com/courses/b2b-marketing-and-sales/quarter-budget-case.md) --- Productovi Learning. Free, no sign-up needed to read.