Why your cost per lead looks fine while your cost per sale climbs

Cheap leads and flat sales are the same problem

I get sent a lot of ad reports by machinery dealers and ag service businesses, and the first page nearly always looks great. Cost per lead down. Volume up. Someone has circled the number in green.

Then I ask what cost per sale did over the same period and the room goes quiet. Nobody knows. The sales live in a CRM, or a spreadsheet, or in the head of a bloke called Trev who has been selling headers for twenty years and does not fill anything in.

So here is my position, and I will defend it. Cost per lead is not a performance metric. It is a volume metric wearing a costume. You can push it down any afternoon by asking for less, targeting broader, and letting the platform find you the cheapest humans who will type something into a box. None of that has to move a single piece of agricultural machinery off the yard.

The two numbers move in opposite directions more often than people expect, because the cheapest leads and the most expensive customers are usually not the same people.

The leak sits between the form fill and the quote

Take a normal month for a dealer running Google Ads on machinery terms and Meta Ads for retargeting and local reach. A hundred leads land. That feels like a strong month and the report says so.

Then you follow them all the way through. A chunk never get reached inside three days, because the enquiry came in at 9pm and the callback happened Thursday. Of the ones you do reach, plenty are researching for next season, or chasing a part, or want a number to beat up their existing dealer with.

The share that reaches a quote is where your money actually lives. I have seen that number sit anywhere between 8 percent and 30 percent on near identical spend, and the difference between those two businesses is worth more than every bid adjustment you will make this year.

Halve your cost per lead and you feel clever. Double your quote rate and you make money. Only one of those shows up on the ad platform dashboard.

Rural buyers do not buy on your reporting month

A grower is not sitting in a city office with a quarterly budget and a procurement calendar. He is in a spray rig, or a header, or trying to get a crop off before the weather turns.

So the enquiry arrives in February and the deal closes in July. In between sits seeding, a conversation with the accountant about what the write off looks like this year, a partner who has opinions, and usually a field day where he puts hands on the machine.

Your ad platform knows none of that. Conversion windows commonly close at 30 or 90 days, so by the time the sale lands the click has aged out and the algorithm has already crowned a winner based on who filled in a form fastest. That is exactly how you end up optimising an entire account toward tyre kickers, one automated bid at a time.

Rural buyers are slow, not cold. Those are different things and your reporting has to tell them apart.

The plumbing that fixes it

None of this is exotic. It is five steps, and most businesses stop dead at step two.

The one that always gets skipped is step four, the value. If you push every won deal back at the same fixed number, or worse at sticker price, the bidding learns that a $28,000 side by side and a $410,000 self propelled sprayer are the same event. They are not, and neither is the margin on them.

The other thing people get wrong is cadence. A monthly upload means your bidding is permanently a month behind the season, which in agriculture is the difference between being in front of a buyer during the tax window and turning up after he has signed.

A lead, a sales-qualified lead and a sale are three different things

Your form cannot tell them apart. Your CRM can, if a human moves the stage.

I would take 22 leads a month where six are genuinely in market and the salesperson can name all six, over 90 leads where nobody can tell you a single thing by Friday. The second account looks better in every report and earns less.

The rule I hold agriculture marketing accounts to is simple. If a lead type cannot be traced to a dollar outcome inside twelve months, it does not get fed back to the bidding as a conversion. Track it, report on it, do not train the algorithm on it.

What I would do in the next fortnight

You do not need a new platform, a rebuild, or a six month project. You need the first three of these done properly before the next buying window opens.

Do the first three and within one season you will know whether your farm equipment leads are genuinely cheap or just cheap looking. Do all five and you can start pouring money into the campaigns that shift machinery instead of the ones that shift form fills.

Written by

Michael Nadalin

Founder, Market Lead & Harvest Lead

Michael is the founder of both Market Lead and Harvest Lead, building lead and sales systems for agriculture and machinery businesses across Australia and the USA.