
Pay-Per-Lead vs Retainer: Who Owns Follow-Up?
Pay per lead vs retainer is argued as a risk question, but the client's sales team decides whether a lead turns out good. The four clauses to sign first.
Pay per lead vs retainer looks like a question about risk: who pays when the leads are bad. For a Malaysian agency running Meta or TikTok lead ads for an MSME client, it is really a question about a number nobody has written down, which is how fast the client's own sales team replies. A pay-per-lead deal pays the agency for a lead and then lets the client's salesperson decide, by calling in four minutes or in four days, whether that lead turns out "good".
The position this post takes: performance pricing is not wrong, it is unfinished. Take it only when the deal defines a billable lead, sets the client's response time, limits what can be rejected and why, and lets the agency see what happened after the handover. Without those four, a retainer is the safer deal, and the four clauses still belong in it.
A note on our interest: Raion sells a CRM, Raion HUB, to agencies and their clients, and our prices do not depend on how you bill your client. The fourth clause below does need a shared view of the leads, which any CRM the client's sales team works in can give, ours included.
Pay per lead and a monthly retainer fail at the same place: the gap between the lead arriving and the client's salesperson replying. Under pay per lead the agency feels it every month as an invoice dispute ("half these leads are bad"), and under a retainer it feels it once, at the renewal meeting. The fix is the same for both: write the billable lead definition, a response time on the client's side, a rejection window with a closed list of reasons, and the agency's view of the pipeline into the deal before the first ringgit of ad spend goes out.
What is the difference between pay per lead and a retainer?
A retainer pays the agency a fixed monthly fee for the work: strategy, creative, campaign management and reporting. Pay per lead pays a fixed amount for each lead that meets an agreed definition, and nothing for the rest. A hybrid sits between them, with a smaller base fee for the work plus a per-lead or per-appointment amount on top.
The usual comparison stops at who carries the risk. Under a retainer the client carries it, because they pay the same in a bad month. Under pay per lead the agency carries it, because a bad month is a small invoice. That framing is correct, and it misses the more useful question: what each model judges the agency on, and who controls that number.
| What changes | Monthly retainer | Pay per lead | Base fee plus per lead |
|---|---|---|---|
| The client pays for | The agency's work, every month | Each lead that meets the definition | The work, plus each lead or appointment |
| Who carries a bad month | The client | The agency | Shared |
| What the agency is judged on | Whether the client feels the spend was worth it | Whether the client accepts each lead | Both |
| Where it breaks | At renewal, when the client cannot see results | Every month, in the invoice dispute | In the per-lead half, for the same reason |
| Who decides whether a lead becomes a sale | The client's sales team | The client's sales team | The client's sales team |
Read the last row. All three models end in the same place, with a salesperson the agency has never met deciding when to pick up the phone. The pricing model only decides when the agency gets blamed for it.
Why do pay-per-lead deals break down?
They break down because the agency is paid on lead quality, and the client judges lead quality after their own salesperson has, or has not, followed up. A lead that gets a reply in five minutes and a lead that gets a call two days later are the same lead on the invoice. By the time the client decides whether to pay for it, they are two different leads.
The research on this is old and nothing since has overturned it. When three researchers audited 2,241 US companies by sending each one a web enquiry, 37% replied within an hour, 23% never replied at all, and the average response among companies that replied within 30 days was 42 hours. Their analysis of lead data found that firms trying to contact a lead within an hour were nearly seven times as likely to qualify it as firms that waited one hour more, and more than 60 times as likely as firms that waited 24 hours or longer (Oldroyd, McElheran and Elkington, "The Short Life of Online Sales Leads", Harvard Business Review, March 2011).
Now read that from the agency's side. If the client's reply time can move the qualify rate that far, then the variable that most decides whether a pay-per-lead lead is "good" is not the targeting, the creative or the form. It is the client's sales floor. And paid leads are the ones that need the fastest reply: in what we have seen, organic leads talk about the problem they have, while ad leads arrive colder and need more of a push.
What we have seen inside 20-plus-agent property agencies makes it sharper. About three in ten leads that died had been assigned to an agent who simply did not act on them, and nothing happened to that agent. Some agents also glance at a lead's name, decide the odds are low, and quietly skip it. Every one of those leads goes back to a pay-per-lead agency labelled "bad quality".
Here is how it plays out. Say a five-person performance agency in Petaling Jaya agrees RM40 per lead with a 25-agent property agency in Puchong, for a new-launch campaign on Meta instant forms. Month one: 300 leads, a RM12,000 invoice, a happy client. Month two: the sales manager returns 130 of 310 leads as invalid. The reasons column says "no answer", "not interested", "says never enquired" and "wrong person". The agency cannot check any of it, because it does not know which leads were called, when, or by whom. It accepts a credit to keep the account, and in month three it quietly loosens the targeting to win the volume back. That is the moment the deal stops working for either side.
Look at the reasons. "No answer" and "not interested" describe the sales conversation, not the lead. "Says never enquired" is usually a forgotten tap on a pre-filled form, which we unpacked in the lead who says they never filled your form, and it gets worse with every hour the first message waits. Only "wrong person" might be a real defect, and even that needs evidence.
Does a monthly retainer avoid the problem?
No. A retainer moves the same problem from the invoice to the renewal meeting. The retainer agency never has a lead returned, but when the client cannot see what the spend turned into, the retainer gets cut, and the reason given is the same one: the leads were not good enough. We wrote about the timing of that in why agency clients leave around month four even when the campaigns are working.
Here is the contrarian part. On this one point, pay per lead is the more honest of the two models: it forces the lead-quality argument every month, in writing, with a list of names. A retainer lets the same argument build quietly until it ends the account. The agency that prefers a retainer only because "nobody disputes the invoice" has not removed the dispute. It has moved it to a meeting it cannot win, months later, with no list to argue over. A retainer's real strengths lie elsewhere: it pays for the work the agency controls, and it does not make the agency underwrite a sales team it cannot see.
So the useful question is not which model. It is what the deal says about the half of the funnel the agency does not run.
What should a pay-per-lead agreement define before you sign?
Four things: what a billable lead is, how fast the client must make first contact, what can be rejected and within what window, and what the agency can see after the handover. Each one closes a door that the month-two dispute walks through.
None of this is new. In the US lead-buying industry, where insurance, mortgage and legal enquiries are commonly sold per lead, contract guidance puts return windows at roughly 3 to 14 days depending on the vertical. It also treats "no answer or voicemail" and "not interested" as invalid reasons to return a lead: a person who did not pick up still exists, and "not interested" is a sales objection, not a defect in the lead (LeadGen Economy, "Lead Contract Templates: Key Terms and Protections", updated March 2026). A per-lead deal agreed in one afternoon on WhatsApp often has a single line, "RM40 per lead", and every question below gets answered later by whoever is holding the list.
| Clause | What to write | What it stops |
|---|---|---|
| Billable lead | A submitted form with a reachable Malaysian mobile number, inside the agreed area, and not a duplicate of a lead from the last 30 days. A qualifying answer counts only if the form asked for it. | The client redefining 'lead' after the invoice arrives |
| Client response time | First contact within 15 minutes during sales hours, and by 10am the next working day for leads that land after hours. A lead contacted late is billable whatever happens next. | The agency paying for a salesperson who called on Thursday |
| Rejection window and reasons | Five working days to reject, from a closed list: unreachable number, duplicate, outside the area, fake or test details. Evidence for each. 'No answer' and 'not interested' are not on the list. | The month-end return of a third of the leads |
| Pipeline visibility | Read access to the leads the agency delivered: who each was assigned to, when it was first contacted, its current stage and its outcome. | Arguing from memory and screenshots |
The response-time clause is the one clients push back on, because it is the only one that asks something of them. Trade for it. Offer a lower per-lead rate, or a first week free, in exchange for the 15-minute commitment. A client who will not promise to contact their own leads quickly is telling you, before you sign, exactly how month two will go.
The billable lead clause is the one agencies write too loosely. "Genuine interest" and "serious buyer" are judged after the sales call, so they hand the decision back to the salesperson whose speed you were trying to take out of the equation. Write only facts that can be checked the moment the lead lands.
Frequently Asked Questions
How to set up a pay-per-lead deal that survives month two
The order matters. Most agencies set the price first and argue about definitions later. Do it the other way round: watch the leads travel through the client's sales team for a month, write the clauses from what you saw, and only then put a number on a lead.
How to Structure a Pay-Per-Lead Agreement With a Client
Step one is the step agencies skip, because a month on retainer feels like giving up the upside. It is the cheapest month of the whole relationship. It shows you the client's real response time before you have bet your margin on it.
How does an agency see the client's follow-up without asking for screenshots?
By putting the lead somewhere both sides work: the CRM the client's sales team already uses, with read access for the agency to the leads it delivered, or a shared system set up for the account. The lead form should feed it directly, so nobody copies leads across by hand. Whichever tool it is, check that it records four things on each lead, each with a time and a name: who the lead was assigned to, when the first message or call went out, every stage change, and the outcome. Those timestamps are the evidence for the four clauses, and they are what a rejection dispute actually turns on.
For the billing side, add two dropdown fields: "Billable" and "Rejection reason", with the closed list from the contract as the options. The salesperson picks a reason on the record while the lead is fresh, not in a spreadsheet at month end, and the delivered leads are exported for the invoice.
One thing to check before you rely on any tool: it has to keep the history, not only each lead's current stage. A record that says "Lost" without who changed it and when will not settle a dispute. The same history is the data behind reporting pipeline instead of clicks, which keeps a retainer alive for the same reason these clauses keep a pay-per-lead deal alive.
When should an agency stay on a retainer?
Stay on a retainer when the client cannot or will not sign the response-time clause. Pay per lead without that clause means the agency is underwriting the client's sales team, and the agency has no way to manage that risk from outside.
The same goes for a client with one salesperson who calls leads "when free", a sales cycle long enough that a lead's value shows up months later (a property launch, a B2B contract), or a client who will not let the agency see what happens to the leads. In each of those, the per-lead price is a bet on a number the agency cannot see. A retainer at least prices the work you can control.
Performance pricing becomes a good deal when the client is ready for it:
Tick all five and pay per lead aligns both sides, because the only thing left for the agency to answer for is the lead itself. Miss one and the hybrid is the honest middle: a base fee that covers the work, and a per-lead amount that switches on once the client has kept the response-time clause for a full month.
The bottom line
Pay per lead vs retainer is the wrong argument to have first. Both models break at the client's reply time, which decides whether a lead turns out good and which the agency does not control. Write the billable lead definition, the client's response time, a closed rejection list with a window, and the agency's view of the pipeline into the deal, then choose the pricing model. Without those four, stay on a retainer; with them, performance pricing becomes a deal both sides can check.


