Unit economics of advertising: how to tell that a channel pays off

A channel pays off in Yandex Direct in the formula the help calls profit: (conversion value − conversion price) × number of conversions. Yandex’s own example uses 100 leads, 50,000 ₽ of spend, and a value of 500 ₽. The conversion price in the example is 50,000 / 100 = 500 ₽, and profit is (500 − 500) × 100 = 0. The example numbers are not a normal lead price. They show that profit goes to zero when value equals price, and turns negative when price is above value.

How to get the price of one lead from spend and leads is covered in the article on CPA. Here value and price are already split: the first is your estimate of the goal, the second is spend per conversion.

Value is an estimate, not a cash fact

The Direct glossary calls conversion value a money estimate of reaching the goal and the margin the goal brings the business. The higher the value, the more that goal weighs in the campaign. It is the basis of the profit calculation for promotion. That is the advertiser’s estimate, not a cash fact.

In the performance-campaign statistics guide, “Conversions” is how many goal visits the ads brought, CR is the share of those visits among all visits, and CPA is the average spend per goal visit. The report is built in Report Wizard: period, goals, attribution model, then metrics. CR and CPA need Metrica goals. Revenue, profit, and ROI appear in the same report when a conversion value has been passed. Until there is a value, there is no profit in this report, even if clicks and visits exist. These fields are for Direct goal visits. They are not equal to “per goal completion” CPA and “per visitor” CPA from Metrica’s expenses report.

CTR is not in the profit formula. Many clicks at a conversion value of zero do not create profit: there is nothing to subtract from, and the “number of conversions” multiplier is about goals, not clicks.

Two ROIs, and neither subtracts fraud

The Direct glossary calls ROI the return on investment coefficient and defines it as the ratio of the return from promotion to the spend on promotion. The formula on the help’s picture is (business metric − ad spend) / ad spend. The page allows the business metric to be sales revenue, profit as revenue minus cost, or another return metric. The picture does not multiply the result by 100. The word ROMI is not on the page: the account calls the metric ROI. The formula has no separate deduction of invalid traffic.

The Metrica report opens at “End-to-end analytics → Sources, expenses, and ROI.” ROI by revenue = (total revenue − spend) / spend × 100. ROI by goal = (goal value − spend) / spend × 100. ROI by goal needs a “goal event” goal and revenue on the goal. The help treats a higher ROI as more profitable advertising. Zero means no extra revenue: only the costs came back. A negative value means more was put into advertising than came back. Spend of campaigns billed per impression is not shown in Metrica reports. This is not the glossary formula: there is no × 100 there. The page does not use the word ROMI. It does not measure the percent by which fraud lowers ROI.

A teaching example on the help’s numbers. Value 500 ₽, conversion price 500 ₽, 100 leads: profit 0. If the value were 800 ₽ at the same price, profit would be (800 − 500) × 100 = 30,000 ₽. Glossary ROI on a return of 80,000 ₽ and spend of 50,000 ₽ is (80,000 − 50,000) / 50,000 = 0.6. The same fraction in Metrica with a × 100 would be 60. Mixing the glossary coefficient and the report percent means comparing 0.6 and 60 as if they were one scale.

This weekly profit does not replace a long payoff. Under “Reports → Retention,” the help suggests comparing sources on who returns more often, makes more visits, and brings more revenue. It describes long-term value as the number of visits, page views, conversions, or revenue of a group over time. The abbreviation LTV is not on the page. There is no formula for revenue per person over a lifetime and no adjustment for invalid clicks. A campaign with a good near-term conversion can bring people who never come back, while a campaign with a weaker first conversion can bring a more active audience later.

The ROI formulas do not subtract the share of spend with no buyer by themselves. That share is counted separately, as in the article on budget loss. The robot percent in a report is not an ROI adjustment either — what it means is in the article on robot share.

The budget without which there is nothing to train the strategy on

For strategies, the budget help recommends a weekly budget of at least 10 × the margin per conversion per week for “Maximum profit,” and at least 10 × the conversion price for “Maximum conversions.” The minimum weekly budget is 300 ₽. The multiplier of 10 is a recommendation for the strategy to work, not a payoff threshold and not a sign of wasted spend.

Maximum spend per day is calculated on Moscow time. With a schedule of one day, up to 100% of the weekly budget and the carryover; with two days, up to 50%; with three or more, or with no schedule limits, up to 35%. As a teaching figure: a weekly budget of 10,000 ₽ and a schedule of three days or more gives a daily ceiling of 3,500 ₽. That is a daily ceiling, not a sign that the morning “burned because of fraud.”

A channel pays off in the account when conversion value is above conversion price and that repeats across the number of conversions, and when ROI is read in the formula you actually opened: the glossary coefficient or Metrica’s percent. Neither subtracts invalid traffic on its own.