Ads that bring in customers.

By moving the things the auction actually prices: relevance between search and page, the conversion rate after the click, and the quality of the signal you feed back. Raising or lowering the budget changes volume, not efficiency, which is why an unprofitable account stays unprofitable at every spend level. Most of the available saving sits on the landing page rather than in the account.

Enough to produce roughly 30 conversions a month, because below that the numbers cannot separate signal from noise and neither you nor the platform can learn anything. Work back from your cost per enquiry to find that figure rather than starting from a round number, and if the maths does not reach it, spend on one channel properly instead of three thinly.

Search produces enquiries within days and a stable cost per acquisition in six to eight weeks; social takes longer because the creative has to be tested first. The honest number is one full sales cycle, because a campaign judged before then is judged on cost rather than on revenue. Set the payback window before launch, not when the first invoice arrives.

You can, and it is usually the most expensive decision in the account. A homepage answers every question except the one the searcher asked, so it converts at a fraction of a matched page and doubles or triples your cost per enquiry. Building the page normally costs less than the spend wasted in a single month without it.

Search, if anybody is already searching for what you sell; social first if they are not. A company nobody knows has no demand to harvest, so paying search to capture it produces very little at high cost, while social creates the demand that brings the cost of search down later. The order follows from whether the demand exists, not from channel preference.

When it has had enough conversions to be judged and is still above your allowable cost with no lever left unpulled. Switching off inside the learning period destroys the data you paid for and restarts it at the same price, which is the most common way budgets are wasted twice. Pause on evidence, not on a bad fortnight.

Once the account is stable and someone owns it daily, in-house is often the right answer and we will say so. What rarely works is part-time ownership, because paid accounts drift within weeks and the drift is invisible until the month closes. The work that genuinely needs outside help is the setup, the measurement and the landing pages rather than the daily management.