Hey.

Work through it in this order. First, sharpen the offer and who it is for, because a specific offer for a specific person beats a generic one. Second, test new creative every week and cut what loses fast, since creative is the biggest lever on most ad platforms. Third, remove friction from the form or landing page: fewer fields, a fast page and one clear action. Fourth, stop making daily edits so campaigns can exit the learning phase. Fifth, avoid over-narrow targeting and let the platform find buyers. Then check cost per client, because cheaper leads that never close are not cheaper.

The usual causes are creative fatigue (the same ads running too long while frequency climbs), seasonal competition in the auction, a change that reset learning (a big budget jump, new targeting or edited ads), tracking that stopped firing so the platform is optimizing blind, or a slow or broken landing page. Check ad frequency, tracking and any recent changes first; one of them is almost always the answer.

Start with a budget you can hold steady for at least 30 days without panic-cutting, and enough to get real data. Meta's learning phase looks for about 50 results per ad set in a week, so very small budgets spread across many ad sets rarely stabilize. Concentrate spend in fewer campaigns, learn what works, then scale it.

No. Very cheap leads often come from vague offers, giveaways or audiences with low intent, and they cost more in wasted calls. Judge campaigns by cost per opportunity and cost per new client. A lead that costs twice as much but closes three times as often is the better buy.