Westpac’s Proprietary Channel Held at 3.9 Days Through Its Busiest Half While Third-Party Decisions Hit 6.1. RG 273.77 Says Your Turnaround Claim Has to Be Evidence-Based

What Westpac actually disclosed

Two documents came out on 2 November: the 2026 Full Year Financial Result and the Full Year Results Presentation & Investor Discussion Pack. Between them they set out, in the bank’s own figures, how far its proprietary mortgage push has travelled.

On the Financial Result’s own wording, Australian housing growth of 7% was “supported by an improvement in proprietary lending which increased from 33% to 36% of new lending” across the year, and from 34% to 37% between the first and second halves. The investor pack gives the same series to one decimal place: proprietary origination was 32.6% of flow in the second half of FY25, 34.0% in the first half of FY26 and 37.2% in the second half. Flow, in the pack’s footnote, means new mortgages settled in the six-month period.

Chief Executive Officer Anthony Miller put it plainly in his statement accompanying the result: “We also increased first party mortgage origination and this remains a priority.” The Financial Result is more specific about why the number moved, describing it as “responding to a targeted strategy to improve performance of this channel”, and attributing it in the Consumer division commentary to “initiatives to improve service, including book a banker, additional bankers, and process improvements”. The pack reports more than 100,000 Book a Banker appointments since the tool launched in February 2026.

None of that is new in direction. Westpac has said for several halves that first-party origination is a priority, and it is not the only major bank saying so. What is new is the service-level detail sitting alongside it.

The turnaround table

On the Consumer home loans page of the investor pack, Westpac discloses median time to decision in days, split between proprietary and third party, for three consecutive halves.

Read the last two columns together. The first half of FY26 was the heaviest settlement half of the three, at $67.4 billion. In that half the third-party median blew out by a full day, from 5.1 to 6.1. The proprietary median did not move at all — it improved to 3.9 days and stayed there.

It would be unfair to read the table as a story of deterioration. By the second half, third-party decisions had come back to 4.8 days — faster than the 5.1 they ran at a year earlier, and the fastest of the three periods. On Westpac’s numbers the broker channel got better, not worse. The structural point is narrower and more durable: the two channels are now managed to visibly different service outcomes inside the same bank, and the gap in the latest half is about nine-tenths of a day.

The pack lists what the bank is working on in that channel, under the heading “Proprietary focus areas”: increase brand consideration, differentiate value proposition, optimise lead quality using Westpac Intelligence, increase lender effectiveness, streamline application requirements, and reduce products. “Streamline application requirements” is the one to sit with. If document and verification requirements are simplified for one channel first, the turnaround gap is a symptom rather than the thing itself.

The number the headlines skipped

Most of the coverage on Monday and Tuesday led with the proprietary share going up. That is accurate, and it is also the least interesting half of the disclosure. The investor pack also publishes mortgage balances split by origination channel, and those move the other way.

The total book reached $531.9 billion, up 7%, or 1.1 times APRA housing system growth. Of the $34.9 billion it added, roughly three-quarters sat on balances the broker channel originated. On the same disclosures the proprietary share of the portfolio fell over the year, from 42.7% to 41.6% — which means the broker-originated share of the book rose, from about 57.3% to about 58.4%.

Both things are true at once because share of flow and share of book are different measurements moving at different speeds. Even at 62.8% of a $58.7 billion half, the broker channel put roughly $36.9 billion of new settlements on this bank’s book in six months, against roughly $21.8 billion through its own bankers — those two are our arithmetic on Westpac’s disclosed flow and share figures, not disclosed numbers. A three-point shift in flow share does not change who is carrying the portfolio. It will take many halves at this pace before it does.

Where the push is concentrated

The pack reports first-party flow twice — once for Westpac Group and once for the Westpac brand alone. The Westpac brand reached 42.4% in the second half, from 38.2% and 34.8% in the two prior halves. The group figure for the same halves was 37.2%, 34.0% and 32.6%.

The Consumer division, per the Financial Result, operates “a portfolio of brands comprising Westpac, St.George, BankSA and Bank of Melbourne”. If the Westpac brand is running five points above the group, the remaining brands must sit below it. The pack does not break them out, so no precise figure is available for any individual brand — but the direction is arithmetic, not inference: the proprietary build is concentrated in the Westpac-branded network, and the regional brands on your panel are, on these numbers, proportionally more broker-reliant than the group average suggests.

For brokers who place heavily with St.George or Bank of Melbourne, that is a more useful read than the group headline. It also raises an obvious question for your BDM: whether the service investment described in the pack — additional bankers, Book a Banker, streamlined application requirements — is being applied across the brand portfolio or inside one brand.

Business lending, and New Zealand

The same strategy appears in the Business & Wealth division, a half-step further along. The Financial Result states that in business lending “the proportion of new lending originated through the proprietary channel improved from 53% to 55%”, with the division’s loans up 12% to $129.0 billion. The pack puts the third-party share of new business lending flows at 45% in FY26, from 47% in FY25.

If you have diversified into commercial or SME lending partly to reduce dependence on residential volumes, note that the channel you diversified into is already the minority channel at this lender, and moving. That does not make it a poor place to write business — 45% of a $129 billion book is a serious amount of lending — but it should inform how much of your diversification strategy rests on a single bank.

New Zealand runs the other way, and the contrast is instructive. In the pack’s New Zealand credit quality pages, the broker-introduced share of the mortgage portfolio rose to 62.3% at September 2026, from 61.0% and 59.5% in the two prior periods, with proprietary falling to 37.7%. Westpac notes the Sep-25 and Mar-26 channel ratios were restated to align with current presentation. Same group, same brand discipline, opposite trajectory — a reminder that channel mix is a market-by-market commercial choice rather than a one-way structural trend.

What RG 273 asks of you

This is where a lender publishing its own service data becomes directly useful, and where it also creates an obligation to be careful. The following is general information about ASIC’s guidance, not advice on your own obligations.

ASIC’s Regulatory Guide 273, Mortgage brokers: Best interests duty, accepts that factors other than cost can matter. At RG 273.71 it lists “matters that are more difficult to quantify, such as the service levels and policies of the credit provider” among the features a consumer may derive value from, and at RG 273.73(d) it includes “the credit policy and risk appetite of the credit provider” among the factors a broker could take into account.

RG 273.77 is the paragraph that bears on turnaround specifically:

In some situations, features which do not relate to the cost of the product will be highly relevant. For example, some consumers will prioritise approval time and compromise on cost because of a time-sensitive transaction, such as an impending settlement date. However, we expect that any claims of this nature you make will be evidence-based and able to be substantiated.
ASIC Regulatory Guide 273, RG 273.77

Immediately after it, ASIC sets out a worked example squarely on this point. In Example 3 a broker tells consumers he has a strong relationship with a credit provider’s business development manager and can get applications processed “much faster”. ASIC’s commentary expects that broker to consider whether timely processing is relevant and valuable for that consumer, to weigh timeliness above cost only where the individual’s circumstances indicate it is of significant importance, to “be able to substantiate his claims about processing times”, and to keep a record of the process.

Three practical consequences follow from reading those together with Westpac’s disclosure.

First, published channel data is better evidence than a relationship. “My BDM gets it through quickly” is the claim ASIC’s example is built to question. “The lender’s FY26 investor pack reports a 4.8-day median third-party decision time for the half to September 2026” is a different kind of statement, and it is one you can put in a file note with a source.

Second, a median is not a promise. It is the midpoint of a six-month distribution, which means roughly half of third-party files took longer than 4.8 days. If a client is working to a settlement date, the useful framing is a range and a source, not a figure presented as what will happen on their file.

Third, it dates. This figure describes the half to 30 September 2026. On Westpac’s own series it moved by more than two days across three halves and worsened sharply in the heaviest volume half. A turnaround note written on last year’s data is not substantiated; it is recalled.

RG 273.165(g) expects your records to include “the options and ultimate recommendation you gave and the reasons why (including a detailed description of your decision-making process)”, and RG 273.169 notes that contemporaneous notes “may help you to accurately capture your thinking and reasoning at each relevant point in time”. If turnaround carried weight on a file, the period and source of the figure you relied on belong in the note at the time, not reconstructed later.

What to review this week

A short, concrete pass over your own process, using what is now public:

1. Pull your own turnaround data. Your CRM or aggregator reporting knows your median submission-to-approval time by lender over the last six months. Compare it with what the lenders publish. Where your experience is materially worse than a lender’s disclosed median, that is a specific, evidenced question for your BDM rather than a general complaint.
2. Check how turnaround appears in your file notes. Search a sample of recent files for the words “quick”, “fast”, “turnaround” and “service”. If the note asserts speed without a source, a period or a reason it mattered to that client, that is the gap RG 273.77 and Example 3 are describing.
3. Separate time-critical files from the rest. Timeliness is a legitimate factor where the client’s circumstances make it one — an auction, a finance clause, a bridging settlement. Make sure the file records the circumstance, not just the preference.
4. Re-read your lender panel by brand, not by group. If the proprietary build is concentrated in one brand of a multi-brand group, your service experience may differ across brands you currently treat as one relationship.
5. Diarise the next data point. Westpac’s next half-year result, covering the six months to March 2027, will refresh both the channel shares and the decision-time series. Any claim you are making from today’s figures has a shelf life that ends there.
6. Raise it where it counts. Aggregator lender forums and MFAA or FBAA channel feedback are the mechanisms that put a documented service gap in front of a lender. A published number is far harder to wave away than a sentiment survey.

What to watch next

Three things will tell you whether Monday’s numbers are a trend or a half. Whether the Westpac brand’s first-party flow keeps climbing past 42.4% in the first half of FY27, or plateaus as the easier gains are taken. Whether the third-party decision time holds near 4.8 days if volumes recover — the 1H26 reading suggests the broker channel absorbs the volume shock, and the next busy half is the test. And whether “streamline application requirements” arrives in the third-party channel at the same time as the first.

The wider context is a cash rate that moved again on Tuesday and a Consumer division reporting mortgage 90-plus-day delinquencies up 4 basis points to 0.67%. Lenders under margin pressure look hard at cost of acquisition, and a banker-originated loan with 3.3 products attached — against 2.8 for a third-party customer, on the pack’s own figures — is the economic case for the strategy in one line. Brokers who understand that case are better placed to argue against the parts of it that cost their clients, and to stop arguing about the parts that are simply commercial reality.

For now, the practical value of this result is narrower than the strategy debate. A major lender has published, in its own documents, what it takes to get a decision through your channel and through its own. That is a fact you can source, date and put on a file — which is more than most turnaround claims in this industry can manage.