5 things every AU mortgage broker dashboard already gets wrong (and how to fix two of them with public data)

> Up front: ausdata.io doesn't sell to mortgage brokers. Quickli, Lendi, and the lender-aggregator dashboards own that segment, and rightly so, broker workflow needs serviceability calculators across 30+ lenders, borrower CRM, AFCA-grade compliance trails, and live LMI lookups, and we don't ship any of that. This post is for the broker-tech engineer who's curious where public data could help around the edges.

I've talked to four broker-tech founders this year. They all said the same thing: the dashboard space is brutal. APRA changes serviceability buffer guidance, the RBA moves the cash rate, the major lenders update their assessment policies on different cadences, and every broker still has 6-8 lender portals open in tabs at any given time.

Quickli and similar tools have done genuinely impressive work consolidating that mess. This post isn't a "we're better than them" post, we're not in that game. It's a list of five things their dashboards understandably get wrong (because broker-tech is hard) and two narrow fixes a public-data layer can plug in.

<!-- IMG: broker-dashboard-public-data-overlay.png -->

What we're not arguing

We're not arguing that public-data APIs replace:

• Quickli's serviceability calculator across 30+ lender policies
• Lendi's borrower CRM
• The Iress / Salestrekker style end-to-end workflow tools
• LMI premium calculators, postcode lender appetite matrices, or anything specific to commission tracking

Those are deep verticals with hundreds of lender-specific rules. A horizontal data API can't and shouldn't try to compete there.

The 5 things broker dashboards get wrong

1. Serviceability buffer assumption is stale

APRA's serviceability buffer guidance is 3% above the loan rate (since October 2021, raised from 2.5%). Most broker dashboards I've poked at hardcode loan_rate + 0.03. Fine, until APRA changes it. The September 2025 APRA media release confirmed the 3% buffer stays, but the next change is just a media release away.

A live dashboard should pull the current buffer guidance from APRA's published policy, not hardcode it. ausdata.io serves APRA's published statistics; the policy text itself is at apra.gov.au. The fix is a tiny daily check.

2. Cash-rate-tracker latency

Several broker dashboards I've seen pull the cash rate from a third-party RSS feed of "RBA decisions" that updates inside 5 minutes of the board release, but a few I've seen still pull from a daily-refreshed source. On board day at 2:30pm, "daily" is six hours wrong.

import requests
r = requests.get(
    "https://api.ausdata.io/v1/real-rate-regime",
    headers={"X-API-Key": "ak_..."}
).json()
print(r["data"]["cash_rate_pct"], r["data"]["cash_rate_period"])
# 4.35 2026-05-06

ausdata.io's cash-rate cache TTL is 15 minutes on latest. On board day you'll see the new rate inside 15 minutes of the RBA HTML changing. That's not "live-live" but it's close enough that nobody's reading a stale rate at 4pm.

3. CPI and unemployment numbers stale on release days

ABS releases land on Wednesdays usually 11:30am AEST. A broker writing up a serviceability assessment at 1pm wants the new number reflected in the dashboard, not yesterday's. Same fix as #2, pull from a 15-min-cached macro endpoint:

r = requests.get(
    "https://api.ausdata.io/v1/economic-dashboard",
    headers={"X-API-Key": "ak_..."}
).json()["data"]

print(f"CPI annual:    {r['cpi_annual_pct']}%")
print(f"Unemployment:  {r['unemployment_rate_pct']}%")
print(f"Real wages YoY:{r['real_wages_yoy_pct']:+.1f}%")

This is the first of the two things public data fixes well. Macro-context numbers that change predictably (CPI, unemployment, real wages, real cash rate) shouldn't live in lender-policy code. They should be cached, cited, refreshed automatically.

4. Borrower benchmarks against national averages, not state

A lot of dashboards show "your borrower's household expense is X% above the HEM benchmark". The HEM benchmark is national. Sydney living costs run materially higher than Adelaide. Showing borrowers a national benchmark hides the actual variance.

ausdata.io's /v1/cost-of-living exposes ABS Selected Living Cost Indexes for six household types, but it's national. We don't have sub-state cost-of-living data because ABS doesn't publish it at that granularity. So this gap is real and we can't close it. Flagging it so the next broker-tech engineer reading this doesn't go looking for SA4-level living-cost data that doesn't exist.

5. Rate-cut tracking buried in news scrapes

When the cash rate moves, brokers want to know which lenders have already passed the change through. Most dashboards track this via news-monitoring services or broker chat groups.

This is the second thing public data can help with, partially. /v1/real-rate-regime gives the official rate; the actual "which lender moved on which day" is private data you still need a scrape pipeline for. But the anchor is structured and citable:

# Two consecutive cash-rate observations let you derive direction
r = requests.get(
    "https://api.ausdata.io/v1/data/rba/F1?limit=2",
    headers={"X-API-Key": "ak_..."}
).json()["data"]

latest, prior = r[0], r[1]
direction = "cut" if latest["value"] < prior["value"] else (
            "hike" if latest["value"] > prior["value"] else "hold")
print(f"Latest move: {direction} of {abs(latest['value'] - prior['value']):.2f} pp"
      f" on {latest['period']}")

Pair that with your existing lender-pass-through scrape and you've got the full picture.

So what fits where

| Gap | Public-data fix? | Verdict |

|---|---|---|

| Serviceability buffer hardcoded | Partial, pull APRA policy text via a daily check | Useful |

| Cash-rate latency | Yes, ausdata.io /v1/real-rate-regime | Strong fit |

| Macro number staleness (CPI/unemp) | Yes, /v1/economic-dashboard | Strong fit |

| State-level cost-of-living | No, ABS doesn't publish at that grain | Honest gap |

| Lender pass-through tracking | Anchor only, RBA rate is the only public piece | Partial |

Two strong fits, two partial, one we can't help with. That's the honest accounting.

What this isn't

This post isn't:

• A pitch for brokers to switch off Quickli / Lendi / Salestrekker / their AFCA workflow tool. Those exist for good reasons.
• A promise we'll ship a broker-vertical product. We won't. Public data is horizontal; verticalising into broker workflow needs domain experts we don't have.
• A take on which lenders are good or bad. Public macro data has nothing to say about that.
• A live LMI / serviceability calculator. We don't have lender-policy data.

It's a "here are two narrow, defensible places public-data tooling can plug into a dashboard you're already building" post. That's it.

Pricing

If you're a broker-tech engineer who wants to plug live macro numbers into a dashboard:

• Free: 500 calls/month, fine for a personal prototype.
• Analyst: $29/mo, 10k calls, fine for an internal-tool refresh every 15 min for one brokerage.
• Pro: $99/mo, 100k calls, webhooks, signed payloads, for shipping AU data inside a broker-facing SaaS product.

Free key at ausdata.io.

Sources

• APRA, Authorised deposit-taking institutions: serviceability buffer guidance
• RBA cash-rate target
• ABS Selected Living Cost Indexes, Australia
• ABS Consumer Price Index, Australia
• ABS Labour Force, Australia