Selling used machinery online when every unit is a one-off listing

Every listing is a shop with one product in it

A used tractor is not a SKU. There is one of them, it has 4,180 hours on it, a cab that smells like the last bloke who owned it, and a price that only makes sense to about eleven people in the country. Sell it and the page is dead.

That breaks almost everything people carry over from normal ecommerce. You cannot build up historical data on a listing that lives six weeks. You cannot scale a winner. You cannot run a proper A/B test on a single unit and get a read before it is gone.

So I stopped optimising the listing and started optimising the category. The individual header, the 2014 JD 6150M with 4,180 hours, is a landing page. The thing that gets budget and learning is the layer above it: mid-size row crop tractors, 120 to 180hp, under $200k. That layer survives stock turning over. The listing does not.

The two searches a rural buyer actually makes

Watch enough search term reports on agricultural machinery accounts and you see the same two behaviours, and they need different treatment.

The first is the model hunter. They type "used case ih puma 165 for sale nsw" because they have already decided what they want. That search is gold and it is cheap because almost nobody bids properly on it. The second is the job hunter. They type "second hand tractor for sale victoria" or "used self propelled boom spray australia". They have a problem, not a model number.

Most yards blend the two into one campaign, let Google smart bid across it, and then wonder why cost per lead sits at $180. Split them. My typical bands on farm equipment leads are $35 to $90 for model hunters and $90 to $200 for job hunters, and the second group needs a lot more hand holding before it turns into a sales-qualified lead.

How I run Google Ads over inventory that changes weekly

The mistake I made early on was building a campaign per unit. It works beautifully for about three weeks and then you are managing 90 paused campaigns and a graveyard of 404s.

Now the structure is fixed and the stock flows through it. Category campaigns stay live permanently, funded by category demand. Individual units get exposure through a feed, not through hand built campaigns. When a unit sells, the feed drops it and the ads stop by themselves.

The unglamorous part that pays for itself is the redirect rule. A sold unit should 301 to its category page with a banner that says this one is gone, here are four similar. I have seen yards recover a meaningful share of their organic and paid traffic doing nothing more than that, because those old listing URLs keep pulling clicks for months after the machine left the yard.

Meta Ads is for the machine nobody is searching for

Every yard has them. The unit that has sat for four months. Odd configuration, wrong colour, priced above what the market thinks it is worth. Nobody is typing that model into Google, so search cannot save it.

That is the job Meta Ads does properly. You are creating demand instead of catching it. A walkaround video shot on a phone, the operator talking about what the machine actually does well, thirty to ninety seconds, targeted at farmers and contractors within a sensible freight radius. I have watched aged stock move off the back of a $600 spend on video views and traffic when four months of listing it did nothing.

What Meta Ads is not good at is being your main source of farm equipment leads. Lead forms on machinery pull a lot of curiosity. Expect $15 to $45 per raw lead and expect a good chunk of it to be people who like tractors. Rural buyers are on Facebook constantly, that part is true, but liking a machine and having finance approved are very different states.

Count units out the gate, not forms filled in

This is where most agriculture marketing goes wrong and it is not a tracking technicality, it is a money question. A machinery enquiry and a machinery sale are separated by weeks, a phone call, an inspection, a trade-in valuation and often a finance approval. If your ad platform only ever sees the form fill, it optimises toward whatever produces the most forms. Which is usually the cheapest, least serious traffic you have.

The fix is not complicated. Every enquiry gets a source stamped on it and gets pushed into the CRM or even a shared sheet, and when a unit sells you mark which enquiry bought it and what the machine sold for. Then push that back into Google as an offline conversion. Now the bidding is chasing sales instead of curiosity, and you finally know that model hunter traffic is worth four times what the generic stuff is worth.

Below is roughly what I see on a yard doing this properly. The shape matters more than my exact numbers. If you cannot draw this for your own business right now, that is the first job, before another dollar goes into ads.

The buying year decides your budget, not your calendar

Machinery money follows the season and the tax year, and no amount of clever bidding argues with that. Push hard while a header is in the paddock and you are talking to nobody. The bloke is working eighteen hour days.

I plan spend against this rhythm, and I plan content against it too. The enquiry that lands in April usually did its looking in February. If you only turn the ads on when you want the sale, you have already missed the part where the decision got made.

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.