How do mortgage brokers get clients, ranked by cost?

The pages that rank for this question are mostly written by firms that sell mortgage leads or marketing to UK brokers. In the search results we could read, the cost per lead those pages print differs several-fold from one to the next, and we saw no source for any of the ranges. We have not opened those pages, so this is an observation about the results and not a finding about the firms. It is enough to explain why this article quotes no market price. A managing partner who plans spend from a range nobody can trace is planning from a guess, and a sceptical reader is right to discount it.

Why cost per lead is the wrong number to rank by

A lead is an enquiry, and a brokerage earns fee income only from a case that completes. Between the two sit the applicants who do not proceed and the weeks a purchase takes to finish. A route with a low price per lead can cost far more per completed case than a route that looks dear, if few of its leads get that far. The reverse is also true, which is why the ranking has to be built from the brokerage's own records and not borrowed.

The second hidden cost is the partner's time. A coffee with an estate agent costs nothing on the invoice and an afternoon in the diary. The worksheet below prices that afternoon at an hourly figure so a free route and a paid one sit on the same scale.

The worksheet: six routes, one quarter, invented figures

Every number in this table is made up to show the method, and none is a market rate. The brokerage is imaginary, a UK firm with a managing partner who values an hour of their own time at £80. Replace each figure with the real one for the last quarter. The last column divides the cash spent plus the hours at £80 by the completed cases.

Reading the ranking without over-reading it

The top two rows win on this sheet because they cost hours and no cash, not because referrals are better. A different brokerage with a thinner network would see fewer referrals for the same hours, and the order would change. The sheet ranks what happened to one imaginary firm in one quarter. It does not rank the routes in general, and no table could.

The bottom row needs separate handling. Writing for the website costs hours now and pays over many quarters, so one quarter's cost per case is the wrong test of an asset that stays. Give it four quarters of the same hours before judging it, and compare the fourth quarter's figure, not the first.

Referral routes also carry a risk the table cannot show. If most completed cases come from two sources, the brokerage's fee income depends on those two relationships. A trade press report we saw in search results describes an online estate agency offering brokers a referral fee, so paid arrangements exist alongside the unpaid ones (Mortgage Solutions, 2019; seen in a search summary and not opened). Whether a given arrangement is allowed, and what must be disclosed, is a question for the brokerage's compliance function and not one this article answers.

The one figure that reorders the sheet

The same spend can rank second or last depending on how many leads reach completion. The table below holds the cost of the shared bought leads and the paid search route fixed, using the totals from the worksheet (£6,000 and £3,200 once the hours are priced), and changes only the number of completed cases.

Read the table the way a partner would. If shared leads complete at three cases a quarter the route is the dearest paid option on the sheet. If a firm that answers within minutes gets nine, it is not. Nothing here says which is true for a given brokerage. It says the count has to be measured before more money goes into either route.

What the sheet leaves out

A mortgage case completes weeks after the first enquiry, so a quarter's spend and a quarter's completions are not the same quarter. Keep a cohort for each period: note which route each enquiry came from on the day it arrives, then count how many of that cohort complete, whenever they do. Judge a route on a cohort that has had time to finish, and be wary of a quarter that looks poor only because its cases are still open.

A bought lead or an introducer is also someone else's marketing sending enquiries under the brokerage's name. A law firm's summary describes a 2016 FCA alert on authorised firms accepting business from unauthorised introducers and lead generators (Regulation Tomorrow summary). We have opened neither that summary nor the alert, so treat it as a pointer for the compliance function and not as a statement of what a regulated brokerage must do.

Filling in the sheet with your own figures

An afternoon with the diary and the enquiry log is enough for a first version. The method has four steps, and the first is the one that gets skipped.

1. List every enquiry from the last two quarters and tag its first source, from the firm's own records or by asking the client.
2. Add up the cash spent per route and the partner hours, using the diary and the invoices.
3. Count the cases from each source that completed, and mark the ones still open.
4. Divide and rank, then run the sensitivity table on the two routes closest together.

The result is a ranking the brokerage can defend to its own partners, which a lead seller's price page cannot give it. Our article on how advisers get clients without cold calling applies a similar partner-time view to a regulated advice firm, and the brokerage page sets out what we check before we introduce an agency to a firm like this.

What to do with the ranking once it exists

Put the next hour or pound into the cheapest route that can still absorb it. A referral route saturates: a second lunch with the same estate agent rarely doubles the introductions, so the cheapest line on the sheet may already be full. The route in second or third place is then the honest home for extra effort, and the sensitivity table tells the partner how many completions it must deliver to earn it.

Set a ceiling before the quarter starts. Decide what a completed case may cost the brokerage against what a completion brings in, and drop any route that stays above the line for two cohorts in a row. Keep the sheet and rerun it each quarter, because a route that ranks fifth now can rank second after a faster call-back process.

What an agency can change on the sheet, and what it cannot

An agency can move the enquiry count or the cash cost of an enquiry, and it cannot move the completion rate the firm's advisers produce. Paid search and the brokerage's own website are the two rows an agency can work on directly. The completion count depends on how fast the firm calls back and what the advisers do with the call, which is the firm's part of the bargain. Our article on how much partner time an agency engagement needs covers the hours the firm should expect to give, and the one on what a broker should look for in an agency covers the questions to ask before anyone is appointed.

When not to hire an agency yet

Do not hire an agency to win clients until the firm can say how many enquiries from each route completed last quarter. Without that figure nobody can tell whether the agency helped, and the reporting turns into a conversation about clicks. Two other conditions make it premature. Nobody at the firm returns an enquiry within the working day, so extra enquiries would go stale. And the partners have not agreed what a completed case is worth to the firm, so there is no number the agency's cost can be set against. In each case the first job is internal and costs an afternoon and a spreadsheet.