THE SHORT ANSWER
Ask four questions and insist on written answers. Where does the demand originate, channel by channel? How is exclusivity enforced, and what happens if it is broken? What are the grounds for rejecting a lead, within what window, and does a rejection produce a credit or a replacement? And how many leads will the pilot contain? A ten-lead trial can only detect an obviously bad supplier; distinguishing a good one from a mediocre one takes sixty to a hundred.
Buying leads is buying a promise about strangers you have never met, made by a company whose incentives are only partly aligned with yours. That is not a reason to avoid it. It is a reason to run the same diligence you would run on a subcontractor who was going to represent you on a customer's roof.
Most of the audit can be done in one phone call and one email, before any money changes hands. The parts that cannot be answered in advance are answered by designing the pilot correctly, which is the section most contractors skip and the one that decides whether the trial tells them anything at all. Get that part wrong and you will spend real money buying an opinion.
The numbers, at a glance
Minimum pilot size for a usable signal: 60 leads, with 100 needed to separate a good supplier from a merely acceptable one
What a ten-lead trial can detect: gross failure only, since a 20 percent and a 40 percent close rate both plausibly produce two or three wins
Reasonable rejection window: 48 to 72 hours from delivery, with named grounds rather than discretion
Reasonable rejection rate: 5 to 15 percent of delivered volume; a supplier refusing any rejections and one accepting all of them are both warning signs
Question one: where does the demand actually come from?
There are only a handful of real sources: paid search, paid social, organic search and content, comparison sites, affiliates, outbound calling, and databases of people who filled in something months ago. They behave completely differently, and a supplier who will not tell you the mix is telling you the mix is embarrassing.
Paid search enquiries are people actively looking and they convert best. Paid social enquiries are interruptions, which means volume is easier and intent is softer. Affiliate and co-registration traffic is where most of the horror stories originate, because the person may have been offered a prize draw rather than a quotation. Ask for the percentage split and ask whether it changes when they scale you up, because that is precisely when cheap traffic gets blended in.
Ask one more thing: can you see the actual advert and the actual form the homeowner completed? A supplier confident in the source will send screenshots the same day. That single request filters out a surprising share of the market.
Question two: how is exclusivity enforced, not merely claimed?
Exclusive is the most abused word in this industry. There are at least four meanings in circulation and only one of them is what you think you are buying.
Sold once, ever. The genuine article: the enquiry goes to you and to nobody else, at any time.
Exclusive within a territory. The same enquiry can be sold to a firm in the next postcode, which is fine for a plumber and useless for a national installer.
Exclusive for a period. Yours for seventy-two hours, then resold. This is shared volume with a head start.
Exclusive to this supplier. The enquiry is sold once by them and separately by the affiliate who generated it. Legally accurate, commercially worthless.
Get the definition in the order document, not the sales email, and ask what the remedy is when it is breached. A supplier with no stated remedy has no enforcement mechanism, because nobody builds enforcement for a promise that costs nothing to break.
Question three: what does the rejection policy really pay?
Read the grounds before the percentage. A policy that accepts rejections for wrong number, wrong country, wrong trade and duplicate is a fraud filter, not a quality guarantee, and almost every supplier offers it. A policy that also accepts out-of-area, renter rather than owner, budget below a stated floor and timeline beyond a stated horizon is a quality guarantee.
Then check what a successful rejection produces. A credit against future volume keeps you buying; a replacement lead keeps you buying; a refund is rare and is the only one that involves the supplier actually losing money. None is wrong, but the difference matters if you intend to stop buying at some point, so ask what happens to unspent credit if you leave.
Question four: is the pilot big enough to prove anything?
This is the part contractors get wrong most expensively, because it feels prudent to start small. With ten leads and a true close rate of 25 percent, the range of plausible outcomes runs from zero wins to five. Zero wins would make you cancel a perfectly good supplier and five would make you scale a mediocre one. The test has almost no power.
At sixty leads the noise narrows enough to distinguish roughly 15 percent from roughly 30 percent, which is the distinction that actually matters commercially. If sixty leads is more than you can absorb in a quarter, run the pilot over two quarters rather than shrinking it, and hold the specification fixed for the whole period so the sample stays comparable.
Design the measurement before the first lead arrives: who calls, within how long, what gets logged, and on what date you will do the arithmetic. A pilot without those four decisions produces an opinion rather than a result.
The audit, in order
Ask for the channel mix in writing, with the advert and form the homeowner saw.
Get the exclusivity definition and the breach remedy into the order document.
List the rejection grounds, the window and whether the payout is credit, replacement or refund.
Size the pilot at sixty leads minimum and fix the specification for its whole duration.
Write down the response process and the review date before the first enquiry lands.
Want leads like this in your pipeline?
Flock runs the campaigns, screens the enquiries and hands you only the ones that match your service area, job size and capacity. You pay per lead, not per month.
Book a 15-minute fit check | See lead package pricing
Related answers
Frequently asked questions
Should I ask for references from other contractors?
Ask, but weight them lightly. Suppliers offer references who are succeeding, and success in another trade or city tells you little. A reference is most useful for one narrow question: how the supplier behaved when something went wrong, and how long it took them to answer the phone about it.
Is a supplier who refuses a small trial being unreasonable?
Not necessarily. Small trials are unprofitable to service and statistically meaningless, so a supplier declining one may simply understand the arithmetic. Judge them on whether they explain that clearly or just push for a bigger commitment. Ask instead to spread sixty leads across two months, which keeps the sample honest and the cash flow manageable.
What if the supplier will not name their traffic sources?
Treat commercial sensitivity as a legitimate reason to withhold specific campaigns and an illegitimate reason to withhold channel categories. Nobody is reverse-engineering a business from the sentence sixty percent paid search, forty percent organic. If categories are refused too, assume the mix leans on affiliate traffic and price that risk in.
How do I check exclusivity in practice?
Ask every homeowner during the first call how many companies have contacted them. Log the answer. Over sixty leads the pattern is unmistakable, and it is evidence you can put in front of the supplier rather than a suspicion. One extra question on every first call costs nothing and is the only proof anyone will accept.
NEED A CLEARER PLAN?
Let’s turn your next move into momentum.
Talk to us →