"How much will ads cost?" is the first question on almost every call we take. An honest answer starts with the fact that you pay for two different things — and mixing them up is the main reason quotes from different agencies are impossible to compare.
Ad spend goes straight to the platform — Meta, Google, Yandex or a Telegram channel owner. It pays for impressions. The agency should not receive it; if there is a commission on spend, it belongs in the contract as a separate line.
The work is strategy, campaign setup, creatives, analytics and day-to-day management. This is the part worth comparing between vendors, and the part most often hidden inside an unexplained "turnkey package".
When an agency quotes a single number, ask how much of it actually reaches the ad account. The answer tells you a lot.
This is not an agency price but a practical floor where the algorithms have enough data to learn. Meta and Google optimise on events — leads, clicks, purchases. With only a handful of events per week the system can't tell whom to show ads to, and results become a lottery.
Smaller budgets still work as a spot tool — boosting one post or checking one offer — but not as a steady source of leads.
As a reference, here is our own price list, which we publish openly:
Other agencies will quote differently. What matters is not the figure itself but what it includes: how many creatives a month, how often you get a report, and who owns the question of where leads end up.
Number of channels and campaigns. Each channel means separate creatives, separate analytics and separate specialist time.
Two languages. Russian and Uzbek are not one text translated twice — they are two audiences with different promises and different creatives. Uzbek-language ads often perform better precisely because fewer advertisers bother with them.
Production. Working with your existing photos and videos is cheaper than a film crew, and existing material is often enough for the first two or three weeks.
Tracking and lead handling. If there is no pixel and leads land on a manager's personal phone, part of the first month goes into infrastructure. That is not wasted money: without it you can't tell what worked.
Work backwards. Not "what can we spare", but "what can we pay for a customer and still make money".
An example with illustrative numbers: average order $200, margin $80, and two out of ten leads become customers. That means a lead can cost at most $16 for ads to pay back on the first purchase. If you want 40 new customers a month you need about 200 leads — roughly $3,200 in ad spend at $16 per lead.
The test period then checks this maths against reality. For a sense of what results look like over time: on the Team Avia travel project we generated 1,649 leads in seven months at an average of $3 per lead. In another niche the figure will differ — it depends on the product, season and competition.
Month one is testing: audiences, offers, creative formats. Conclusions you can rely on appear in month two. Month three shows whether results hold or the first good numbers were luck.
You can spend a budget in two weeks, but you can't learn whether a channel works. Treat promises of results "within a week" with caution.
Keep ad spend and fees separate. Plan for at least $500 a month on the platform. Derive your target cost per lead from your margin, not from someone else's case study. And plan for three months, not a weekend test.
Want to know what advertising would cost for your business? Get in touch — we'll work it out from your numbers and tell you straight if it's too early to launch.