What Cost Per Lead Actually Did Across 22 Google Ads Accounts

The actual spread

Most articles about Google Ads cost per lead quote an industry average. We are not going to, because we do not have one worth quoting and neither does anyone else who has not seen your account. What we do have is twenty-two engagements with the documents published, so here is what those cost.

A ten-fold range across six accounts in the same channel should tell you how little a single industry figure is worth. The construction company at $12 a lead and the fire protection contractor starting at $118 were both run competently. They sell different things to different buyers with different urgency, and the price of attention follows that.¹²

Cost per lead is a curve, not a number

The single most consistent pattern across these accounts is that the interesting figure is the direction, not the level. Three of the engagements above are quoted as a movement rather than a point, and that is deliberate.

• Trailer dealership, nine monthsCost per lead fell from $81 to $41 while lead volume rose from 162 to 288. Both halves matter: a falling cost per lead with falling volume usually means the account has retreated to its easiest traffic.
• Roof and exterior cleaningLeads climbed from 21 to 85 per period while cost per lead fell from $172 to $61. This is the shape you want and it is not common.
• Fire protection, two years$118 in year one to $68 in year two, with leads up 182%. A year is a long time to hold a client's nerve, and it is the honest timescale for a considered-purchase account.

What actually moved it

Reading the strategy sections of these twenty-two write-ups next to each other, the same three levers appear again and again, and none of them is a bidding setting.

1. 01Cutting the queries you were never going to winMost inherited accounts are paying for breadth. The early work in these engagements is subtractive far more often than it is additive.
2. 02Matching the landing page to the queryA lead that arrives on a page answering a different question than the one they typed is expensive twice: once for the click and once for the sales time.
3. 03Letting the account gather enough conversions to learnAutomated bidding needs volume before it is better than a careful manual bid. On the $300 and $500 accounts here, that took months, which is the real argument for patience on a small budget.

Where we would not use this channel

A point of view is only worth having if it rules something out. We would not lead with Google Ads where there is no existing search demand to capture: a genuinely new category, or a product people do not know to look for. Search advertising harvests intent that already exists. Creating that intent is a different job, and paid social does it better.

The honest tell is in the keyword research: if the terms that describe what you sell have almost no volume, the channel is not broken, it is the wrong channel. We would rather say that at the proposal stage than bill for six months of proving it.

Related