A pay per call campaign test is a small buy with the rules already written, not a large buy you plan to 'clean up' if it feels wrong. One product. One geography you can staff. Hours your people actually sit. A cap that matches the chairs, not the budget you wish you had. You are trying to learn whether the definition of a qualified call survives contact with real callers. You are not trying to prove a close rate in four days.
If the invalid list is still a conversation, you are early. Write it before the test, using billable versus unqualified calls. Geo, hours, product, and exclusivity go in the same packet. That list is pay per call filters. A test without those lines tells you that phones rang. You already knew phones could ring.
One product and a floor that is present
Do not test 'insurance' or 'home services.' Test final expense in the age band you write, or HVAC repair in the ZIPs you roll, or auto in the states on the license. A mixed pile gives you a mixed report, and the post-mortem becomes a debate about which calls 'should have' been in the test. Final expense and life insurance are the usual pair buyers mash together. They are different talk tracks. Split them or do not call it a test.
Staff the hours you are buying. If the test runs while the desk is in a meeting, you measured your calendar, not the calls. Answer rate is the first number that matters, and it is yours. A missed transfer is not evidence the media was bad. Cap concurrency to the people who can pick up after you account for who is already talking. The roster math is the page for that. It is not a headcount we publish, because we do not have yours.
Decide exclusive or shared for the test and leave it there. Changing midweek makes the second half a different product. Shared will get shopped. Exclusive will cost more and give your agent the only shot on that delivery. Either can be a fair test. Both at once is two tests wearing one name.
What to read in the first days
Read calls, not a single ratio. A handful of recordings will show whether the caller asked for the product, whether the state matched, and whether your opener followed the brief. Pull them from more than one source if more than one source is on. One clean path can hide next to a noisy one in an average.
Dispositions should be usable on day one. Wrong product. Wrong geo. Outside hours. Duplicate. Decision-maker absent. Qualified, no sale. If the only buttons are 'sold' and 'other,' you will not know what to change. Call quality beyond duration is the reason the disposition exists. Duration is the floor. The button is the reason.
Do not kill a source because the first day did not close. A qualified call is not a sale. Do kill a source, or pause it, when the recordings show the rule failing on purpose: wrong product coached into a yes, a geo you excluded, a handoff that pretends a digital call was a television live transfer. That is not 'the test learning.' That is a path that does not match the order.
When to raise it, and when to stop
Raise one thing. More hours you already staff, or a neighboring geo you can actually serve, or a slightly higher cap after answer rate and handle time say the floor kept up. Do not raise product, geo, and cap on the same morning and then wonder which change filled the board with the wrong calls.
Stop, or do not expand, when the floor cannot answer, when the invalid list you wrote is showing up as most of the volume, or when you are disputing calls that met the rule because the close was weak. The third one is an intake problem. Fix the talk track or the hours. Buying more will not repair it. How to file a real reject, when you have one, is pay per call disputes.
A test can also tell you that you wanted a lead record, not a live call. If nobody can catch a transfer and the team dials a queue in the morning, CPL may fit better than cost per call. That is a useful result. It is not a failed test. Cost per call versus cost per lead is the comparison. Pick the event your team will work this month.
RidgeRise will run a short test on CPL or cost per call when the brief has a product, a map, hours, and a definition of qualified. Volume comes from campaigns we buy and from publishers screened to the same rules. We will not hand you a sample close rate, because that number would be fiction on a page like this.
Bring the one-product brief to buyers and start a test campaign. If you want the sheet checked before anyone turns a call on, send it through contact. The test is the rules plus a week of listening. It is not a slogan about scaling.
