The most expensive advertising mistake is not a bad creative or a wrong bid. It is putting your budget on the wrong platform for the size of your budget. A channel that prints money at 15,000 EUR a month can quietly burn a 900 EUR budget, because it never gets enough data to learn. This guide maps ad platforms to budget bands, so you spend where your money can actually work.

Why budget, not preference, picks the platform

Every modern ad platform runs on machine learning that needs a minimum number of conversions per week to leave the learning phase and optimise. Spread a small budget across four platforms and none of them ever gets enough signal. The rule of thumb: concentrate budget until one channel is genuinely working, then expand. Depth beats breadth until you are past roughly 10,000 EUR a month.

The second factor is your margin and average order value. Platforms differ enormously in cost per click and cost per lead, so the same 40 EUR product is viable on one channel and hopeless on another. Judge each channel against your break-even, never against a generic benchmark.

Under 1,500 EUR / month: pick one channel

At this level you have one job: get a single channel past its learning threshold. For most e-commerce and local businesses that means Meta (Facebook and Instagram) or Google Search, and the choice comes down to intent.

  • Google Search if people already search for what you sell (a plumber, a SaaS category, replacement parts). You are harvesting existing demand, so even a small budget converts. Keep the account tight and cut wasted spend ruthlessly at this size.
  • Meta if you need to create demand or your product is visual and impulse-friendly. It reaches people who were not searching, which is most of e-commerce.

Skip LinkedIn and, usually, TikTok at this budget: LinkedIn clicks are too expensive to learn on a shoestring, and TikTok rewards a volume of creative you probably cannot produce yet.

1,500–10,000 EUR / month: one strong channel plus one test

Now you can run a primary channel properly and carve out 15–25% to test a second. Common pairings:

  • E-commerce: Meta as the engine, Google (Search plus Performance Max) to capture branded and high-intent queries. Or add TikTok if you can feed it native, hook-first video.
  • Lead-gen / services: Google Search as the engine, Meta for retargeting and top-of-funnel.
  • B2B: Google Search for high-intent terms, and a small LinkedIn test if your deal size justifies its premium clicks.

Over 10,000 EUR / month: a portfolio, allocated by role

Above this level, running multiple platforms is not diversification for its own sake — each earns a defined role: prospecting, retargeting, brand-demand capture, incremental reach. This is where a single cross-platform view stops being a nicety and becomes essential, because you are now moving budget between channels on blended economics, not per-platform ROAS.

Platform cheat sheet

  • Meta — broadest reach, best for demand creation and visual products; creative is the main lever.
  • Google Search — captures existing intent; the most reliable first euro for many businesses.
  • Google PMax / Shopping — strong for e-commerce with a clean product feed, but needs guardrails.
  • TikTok — cheap reach and fast growth if you produce native video at volume; weak if you cannot.
  • LinkedIn — unmatched B2B targeting, but the highest cost per click; only worth it with a high deal value.

Frequently asked questions

Should a beginner run Meta and Google at the same time?

Usually no. Under about 1,500 EUR a month, splitting the budget starves both. Prove one channel first, then expand.

Is TikTok worth it for a small budget?

Only if you can produce several native videos a week. TikTok rewards creative volume; without it, a small budget underperforms.

How much should I put on a new-channel test?

Ring-fence 15–25% of spend, commit to a fair test window, and judge it against your break-even, not the incumbent channel’s mature numbers.

Not sure which channel is actually earning right now? Run a free 3-minute profit audit — no card — and see where your money works hardest. Or try the demo.