Two questions, one answer. You can't set a budget until you know what a lead is worth to you.

Ignore the "industry average" CPL. Cost per lead for equipment rental swings wildly — by category, region, season, and how strict you are about what counts as a qualified lead. A benchmark you read in a blog is someone else's account, with someone else's tracking and someone else's definition of "lead." The only number that matters is yours: your average rental value, your margin, and your close rate. Everything works backward from those.

Work backward, not forward. We put this question to a rental operator we work with, for a compact track loader: median order around $600 (average $850, pulled up by bigger machines and longer rentals).

Margin is where rental surprises people. Rental is a fixed-cost business — the building, the staff, the insurance cost the same whether a unit rents or sits. So the marginal cost of one more rental is small: on a healthy, already-running operation, 60–80% of an incremental rental drops to the bottom line. Call it roughly $400 on that $600 order. (The honest caveat: if you've got one loader that never rents and a 10,000 sq ft building supporting it, there's no profitability at all. This math is about adding rentals to a working business — which is exactly what ads do.)

Now the close rate. Rental calls are high-intent — someone calling about a loader needs one, usually within days. Be conservative and say you close 1 in 4. That makes a qualified call worth about $100 to you. That's your ceiling on cost per call, at the floor of sales performance. If you're closing worse than 1 in 4, the problem usually isn't the ads — it's the counter. When targeting and sales are both right, the numbers get remarkable: the operator we spoke with closes most orders on the first call, and one of our clients says he books nearly every call that comes in. Good targeting hands your sales team layups — it doesn't take the shots for them.

And that ceiling still understates it: the customer renting a loader today rents other equipment, several times a year. The first order is the smallest thing about that lead.

Here's the catch: that $100 ceiling belongs to one product. Run the same math on a $100 aerator rental and the ceiling collapses to a few dollars a call. Same phone, same account, completely different economics.

You cannot rely on Lifetime Value (LTV) to rescue the aerator math, either. The equipment itself acts as a natural intent filter. A skid steer selects for high-LTV contractors who will rent repeatedly, whereas an aerator attracts a high volume of one-and-done homeowners. With that B2C traffic, there is rarely a second rental to offset your initial acquisition cost.

Which is why a blended account can't answer the budget question at all. If skid steers, aerators, trenchers and pressure washers all live in one campaign, your dashboard shows you one averaged cost per call — say $50 — and that number is simultaneously excellent and terrible. $50 for a skid steer call: great, push harder. $50 for an aerator call: you're burning money on every ring. Averaged together, you can't see either. You don't have one cost per call; you have a dozen, and the blend hides all of them.

Segmented by product, the picture inverts. Each category gets its own campaign, its own target, its own verdict: the loader campaign earns more budget because its calls are worth $100+; the aerator terms get capped or cut because their math never works. That's not a reporting nicety — it's the difference between knowing where to push and guessing.

The trap: spending more when more won't help. Google reports two kinds of "lost impression share" — the slice of searches where your ad could have shown but didn't: budget-lost (you ran out of money) and rank-lost (your Ad Rank wasn't high enough). They call for opposite fixes. Picture an account where budget-lost is basically zero but rank-lost runs 60–80%: that account is never short on money — it's losing auctions on bid and quality. More budget does nothing there. And here's the same segmentation rule again: check this per campaign. Your loader campaign might be budget-capped (feed it) while your general-tools campaign is rank-limited (fix it, don't fund it). The account-level number is another blend that hides both.

Your dashboard number overstates your profit. Google sees the booking, not what happens after — the cancellation, the no-show, the refunded deposit, the unit swapped for a cheaper one. If you collect on about 80% of what looks "booked," a job that reads as $600 is closer to $480 in your pocket. Bid to the money you keep, not the number on the screen. If you optimize to ROAS, same rule: want a true 6x while collecting on 80%? Your reported target needs to sit nearer 7.5x.

Do this now. Pick your top three product categories and run the math separately for each: median order, marginal margin, close rate → cost-per-call ceiling. If your account structure can't tell you cost per call per product, that's the first thing to fix — because until it can, every number on your dashboard is an average of things that should never have been averaged.