Three months of paid search and I finally believe my cost per conversation

Does $220 a conversation hold if you triple the spend? My experience with anything keyword driven is that the cheap intent is a small pool and the marginal dollar buys worse traffic. You were at maybe $900 a month in media. What happens at $3,000.

One closing is not a cost per deal, it's an anecdote with a receipt attached. You know that, I know you know that, but somebody is going to read this and budget off it.

The 19 conversations is the number with any weight, and even that is thin. I'd want two more quarters at the same settings before I let the agency raise the retainer on the strength of it. Still, the geo fix is a real finding and it cost you five weeks, which is cheap for a finding.

How fast are you calling them back? On the management side we track this obsessively for maintenance calls and the falloff after about ten minutes is brutal. A land seller who filled out a form at 9pm has probably filled out two others. If you're getting 19 real conversations out of 61 forms, some of that gap is speed, not quality.

$1,500 management on roughly $900 of media is a 165% ratio, which is fine at your size and awful at any other size. Most shops I know quote a floor like $1,200 to $2,000 and then switch to 15% to 20% of spend once you're past five or ten thousand a month. Ask them now what the schedule looks like at $5,000 so you're not renegotiating from inside a working campaign.

Also, two person shop with two other land clients in adjacent markets is worth a direct question. Are you bidding against their other client's keywords.

Dumb question while I'm still studying for the exam. Did you send them to a standalone landing page or to your regular site? I can't tell from your post whether the form was on a one page thing built for the ad or on a normal about-us type site.

The county keyword rebuild is the whole result, and it's worth being precise about why, because the mechanism generalizes.

Land is the one asset where the searcher's body and the asset are routinely in different states. Google's location settings distinguish between people physically in a place and people showing interest in a place, and the default has moved around over the years, so check what your account is actually set to rather than assuming. Anchor, your first five weeks bought you people standing in the county, which for absentee owned parcels is close to the wrong list. Putting the county in the keyword and again in the form field moves the qualifier from geography to stated intent, which is what you were paying for the whole time.

On the number to run the business on: don't operate off cost per contract with n=2. The interval around that is embarrassing. Cost per qualified conversation stabilizes much sooner, and at 19 it's shaky but usable. Keep two conversion rates separately, conversation to accepted offer and accepted offer to closing, because those move for different reasons and blending them hides which one broke when your cost per deal jumps.

What I would hold the agency to going forward is the leading metric, share of form fills that name a target county and own the parcel. If that percentage falls while spend rises, you're buying volume off worse intent and you'll see it two months before it shows up in closings. Put that number in the monthly report as a line item, in writing, before you increase the budget.

Three wholesalers fishing in 23 forms sounds about right honestly. They search the same terms you buy.

Weekly negative keyword review you actually attend. That phrase told me more about whether I want to do this than the rest of the post. Is this a job now, or does it settle down after the setup?

Did you ever test Facebook against this, or was search the only channel from the start? For land I'd assume nobody is scrolling and deciding to sell 20 acres, but I've heard the opposite from people who run it, that the absentee owner audience is easier to build there than to buy on keywords. Anchor, if you had a second $4,000 would it go to more search or a Facebook test?

What is call tracking exactly? I've now read it three times in this thread. Is it just a different phone number for each ad so you know which one made the phone ring, or is there more to it?