THE SHORT ANSWER
Stop when the backlog runs beyond ten to twelve weeks, because you are buying enquiries you will disappoint. Stop when cost per job has exceeded a third of gross margin across two consecutive cohorts of sixty leads and the cause is internal. Stop when nobody can call new enquiries within an hour. Stop when referral and repeat work already fills the diary at better margins. And stop when you are buying leads to avoid fixing a sales process.
A pay-per-lead supplier is not the right answer for every contractor, and the businesses that stay on it longest are not always the ones it serves best. Some firms should never start, and a good number should stop for a quarter, a season, or permanently. Pretending otherwise would make everything else on this hub less useful.
There are five situations where continuing to buy destroys value. Four of them are temporary and one is structural. Knowing which one you are in matters, because the remedy for a capacity problem is a pause and the remedy for a structural mismatch is a different model entirely. Get the diagnosis wrong and you will change supplier when you should have changed process.
The numbers, at a glance
Backlog threshold: 10 to 12 weeks of committed work is the point at which new enquiries start damaging your reputation
Cost threshold: cost per job above a third of gross margin across 2 consecutive cohorts of 60 leads
Response threshold: if new enquiries routinely wait more than 1 hour, you are wasting roughly a third of what you buy
Referral threshold: if referrals and repeat work already fill more than 80 percent of capacity, purchased volume is displacing better work
Stop when the backlog is longer than the customer will wait
A homeowner with a leaking roof or a failed boiler will not wait eleven weeks, and a homeowner planning a kitchen may wait but will spend those eleven weeks talking to two other firms. Buying enquiries you cannot serve promptly converts marketing spend into negative reviews, which is the most expensive way to spend money in this industry.
Ten to twelve weeks of committed work is the practical ceiling for most trades, shorter for anything urgent. When you cross it, pause purchasing rather than reducing it, and use the pause to quote the backlog properly. Restart four to six weeks before the backlog is due to clear, allowing for the lag between purchase and signature.
Stop when the arithmetic has failed twice and the cause is yours
One bad cohort is noise. Two consecutive cohorts of sixty or more leads where cost per job exceeds a third of gross margin is a signal, and the important question is where the failure sits. If contact rate and quote rate are healthy and the win rate is poor, the problem is your pricing or your proposition, and more leads will simply cost you more.
This is the uncomfortable case, because the instinct is to change supplier. Changing supplier when the failure is internal reliably produces the same result eight weeks later at the cost of another sixty leads. Pause, fix the identified stage, and return with a measurable hypothesis rather than a hope.
Stop when the business cannot absorb what it buys
Nobody can call within an hour. Contact rates collapse and you are paying full price for perhaps two thirds of a product.
Quotations are running more than a week behind. The enquiry has been priced by someone else before you reply.
The owner is the only person who can sell. Purchased volume converts an owner constraint into an owner crisis, and it does so within weeks.
Peak season with every crew committed. A deliberate seasonal pause is good practice, not a failure, and it costs nothing if the supplier lets unused volume roll forward.
Stop permanently when the model does not fit the business
Some firms should not buy leads at all. A specialist doing heritage or highly technical work priced per project has too few relevant enquiries in any territory for a repeatable pipeline to exist. A firm whose diary is filled by architects, developers or a strong referral network is buying weaker enquiries at higher acquisition cost than the ones it already receives free.
Very small operations face a different version. A single-vehicle business needing four jobs a month does not have enough volume for any cohort to be measurable, which means every judgement is made on noise. For those firms the honest advice is to spend the same money on local search presence, reviews and a website that answers the questions homeowners actually ask, and to revisit purchased volume when a second crew appears.
There is also the case where purchased leads are working and should still be reduced. If an owned channel has matured and now supplies most of your requirement at lower cost, step purchased volume down to a floor rather than eliminating it, and keep the relationship alive. That is a reduction, not a stop, and it is the ending this article recommends most often.
Decide honestly whether to pause, stop or reduce
Measure the current backlog in weeks and compare it against what your customers will tolerate.
Check cost per job against gross margin across the last two cohorts of sixty leads.
Identify which stage is failing before blaming the source, using contact, quote and win rates separately.
Ask whether referral and repeat work would fill the same capacity at a better margin.
If you pause, set the restart date now, allowing for the lag between purchase and signature.
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
Is pausing better than reducing volume?
For a capacity problem, pause cleanly and restart with a date. Reducing volume keeps a trickle of enquiries arriving into a business that cannot serve them, which produces the reputational damage without the commercial benefit. For a cost problem, reduce and diagnose instead.
How long should a pause last?
Long enough to clear the cause, plus the purchase-to-signature lag. For a backlog problem that is usually six to ten weeks. For a sales process problem, allow a full cycle after the fix so you can measure the change rather than assume it.
What happens to volume I have already paid for?
It depends entirely on the supplier. Ask before you buy whether unused volume rolls forward, expires or is forfeited, because the answer determines how freely you can pause when you need to. Forfeiture on prepaid volume is worth negotiating out at the start.
Should I tell the supplier why I am stopping?
Yes, specifically. A supplier told that close rate fell on a particular criterion can fix it; a supplier told the leads were poor cannot. It also preserves the relationship, which matters when your backlog clears and you need volume at short notice.
NEED A CLEARER PLAN?
Let’s turn your next move into momentum.
Talk to us →