DSR3 Has Landed – Inside Our New Property Research Platform

Why DSR3 Was Built From the Ground Up

Jeremy summarises DSR3's core promise in one line: higher capital growth, sustained for longer. The algorithm follows the original DSR (2010) and DSR Plus (2015), and rather than incrementally improving DSR Plus, the team rebuilt it entirely, incorporating a new methodology, a larger set of underlying metrics, and better use of that data overall. Jeremy explains starting fresh allowed them to move past accumulated inconsistencies in the older approach, and gave a clear reason (a genuinely new algorithm, on a new platform) for existing users to transition across.

On the scale of improvement, Jeremy notes the step up from the original DSR to DSR Plus was already significant, but the step up to DSR3 is larger still. Based on back-testing, he estimates DSR3 tends to roughly double the national growth rate on average, though the exact multiple varies by era: in slower periods (say, 3% national growth), a DSR3-selected market might achieve 8–9%, comfortably more than double; but in a genuine boom (say, 15% national growth), doubling that figure becomes much harder to sustain. He notes this outperformance has historically held for over a decade on the current model.

Why Not Just Keep Using DSR Plus?

Jeremy acknowledges a mistake with the earlier DSR Data website: hosting both the original DSR and DSR Plus side by side created confusion about which to use, rather than making clear that DSR Plus was simply the newer, better version. He confirms he personally stopped using the original DSR once DSR Plus was available, and has now moved entirely to DSR3, no longer referring to DSR Plus at all. The older DSR Data platform will remain available for a period for users who prefer its interface, but the underlying algorithm guidance is clear: use the newest version available.

A New Pay-Per-Use Pricing Model

Unlike the subscription-based pricing of DSR Data and most competing platforms, the new platform uses a pay-per-use model. Jeremy explains this responds to years of user requests (people wanting to use the site for a single day or week, or only needing a couple of specific metrics rather than a full subscription), and that offering only two flat pricing tiers previously (as DSR Data did with its "Light" and "Pro" plans) never suited everyone's actual usage patterns. Damien notes heavier users (larger buyers agent firms, for instance) may find a subscription still better value for their volume of use, but lighter, more sporadic users are likely to prefer paying only for what they need.

Platform Tour: Getting Started

Damien and Jeremy recommend starting with the "Getting Started" tutorials on the main site before diving in independently. The platform offers five core research tools: suburb search, heat maps (new, not previously on DSR Data), market metrics, historical charts, and context rulers, alongside a trial mode using mock data so new users can learn the interface without incurring any cost or risk of making a real decision based on fake data.

Suburb Search: Setting Up a Real Search

Using a live example (data as of January 2024), Jeremy explains the suburb search tool starts with defining hard limits (budget, and any geographic restrictions, for example, needing to avoid an existing state due to land tax exposure, or requiring proximity to home for a hands-on renovation project) before applying softer sort preferences (like sorting by yield if cash flow matters, without using it as a strict cutoff).

They demonstrate searching at the SA3 level first (a statistical area roughly equivalent to a local government area) rather than jumping straight to individual suburbs, since smaller suburbs can have limited sales data and therefore less reliable figures. Jeremy explains the ABS geographic hierarchy briefly: SA1 (smaller than a suburb) rolls up into SA2 (roughly postcode-sized), which rolls up into SA3 (roughly LGA-sized), which rolls up into SA4 (of which there are just over 100 nationally).

Statistical Reliability as a Key Filter

Jeremy reiterates that statistical reliability (a score out of 100) reflects how much confidence can be placed in a suburb's DSR score, since even a suburb with very little underlying data will still receive a DSR score, just a less trustworthy one. He recommends checking the context ruler for any unfamiliar metric to understand what counts as a strong or weak value before applying it as a filter.

A Live Example: January 2024, Filtering for Perth

Setting a budget of up to $700,000 and DSR3 sorted from highest to lowest, the search results are dominated by Perth-area SA3s (Gosnells, Wanneroo, Armadale), with a scattering of Darwin, all showing strong statistical reliability. Jeremy notes seeing a top-10 list dominated by a single city (rather than scattered, isolated results) is itself a good confidence signal, a genuine cluster rather than a one-off anomaly. Damien highlights the strong yields present in the data, alongside a comparatively low long-term (10-year) growth average visible at the time, since that historical figure hadn't yet caught up with Perth's more recent, rapid growth.

The Cost of a Real Search

Running this live search costs $12.78 in the demonstration, illustrating the pay-per-use pricing in practice.

The Most Common Mistake: Over-Filtering

Jeremy reiterates a now-familiar warning: adding too many filters (particularly a strict vacancy rate cutoff, commonly set around 2%) can eliminate genuinely excellent markets purely due to small-sample volatility. Using the same example as in prior episodes (a 200-house suburb with 40 rental properties, where a single vacancy already equals 2.5%), he stresses that vacancy rate is already factored into the DSR score itself, so manually filtering by it separately only risks excluding good opportunities rather than genuinely improving results. His recommendation: keep filters limited to genuine hard constraints (budget, geography) and let the DSR score itself do the heavy lifting, sorted from highest to lowest.

Historical Charts: Testing Real Outcomes

Using Gosnells (Perth) as an example, Damien and Jeremy pull up 18 months of historical growth data: a typical value increase of around 25.5%, or roughly $152,000 on a property purchased around $598,000 in January 2024. Extending the same chart back four years shows 74% growth, from roughly $432,000 to a $319,000 dollar increase over that period. By contrast, running the same query for Melbourne over the same four-year window shows just 2% growth, illustrating a real-dollar opportunity cost of well over a quarter of a million dollars on a comparable-sized property, depending on which market an investor had chosen four years earlier.

Jeremy explains a further use for historical charts: identifying markets that have been flat for an extended period but are just beginning to show early signs of an upswing, useful for timing entry, as well as for visually smoothing out short-term volatility in sensitive metrics like vacancy rate, checking the general trend line rather than reacting to any single month's figure. He notes DSR3 already incorporates trend analysis internally, so most users won't need to do this manually, though some may still want to for their own research purposes.

Comparing Multiple Metrics on One Chart

Using Sydney's growth from mid-2015 to around early 2022 (roughly 89% total growth) as an example, Jeremy shows how a market cycle timing (MCT) chart overlaid on the same period visually explains why Sydney's MCT score dropped, since it had already had its major run, while Melbourne's MCT remained comparatively higher, reflecting its longer period of stagnation (only around 76% cumulative growth from 2015 to present, still below the national average over that timeframe). Jeremy notes that a long period of underperformance can actually set a market up for above-average performance later, and that the ideal entry point is generally once early signs of upswing become visible, rather than trying to enter right at the very bottom of a flat period.

Regional and Smaller Markets: A Different Kind of Trade-Off

Both note that smaller regional markets (currently including places like Darwin) can offer strong short-to-medium-term opportunities for a "get in, get out" style investor, but caution they're less suited to a genuinely long hold, since they can experience extended flat periods (potentially up to around 15 years) in a way that larger significant urban areas are somewhat less prone to, though even large markets aren't immune (citing Perth's extended downturn following the earlier resources sector decline as an example).

Jeremy also reiterates his consistent view on yield: the small amount of extra yield gained by heavily restricting a search (say, chasing an extra 1.5%) isn't itself what holds an investor back, it's the much larger pool of potentially higher-growth markets that gets excluded in the process of chasing that yield.

Heat Maps: Visualising Clusters

Demonstrating the heat map tool on Sydney units, Jeremy explains its purpose: revealing whether a promising suburb is a genuine, confidence-inspiring cluster of strong-scoring neighbours, or an isolated one-off result. Running a live search (costing $109 in this case, reflecting a broad set of metrics requested across a large city) reveals a concentration of stronger scores in Sydney's north-west. Jeremy immediately flags the need to filter by statistical reliability in this specific area, since some of the apparent "heat" turns out to be driven by very limited underlying data (one example, Yellow Rock in the Blue Mountains, showed a DSR3 score of 64 but a statistical reliability of just 17, indicating the algorithm was working with very little real information for that specific pocket).

Once filtered for reliability and a stronger DSR3 threshold, the field narrows considerably, showing a cluster around Kellyville and Rouse Hill, though yields for units in that pocket remain relatively modest (around 4.8%). Damien demonstrates a second practical use case: an owner-occupier restricted to a specific city (Sydney) and budget (around $600,000–$700,000) can use the same tool to see how far that budget stretches once DSR score is factored in alongside price, in this case, pointing toward areas like Penrith, St Marys, Werrington, and Kingswood as realistic options if growth potential is a genuine consideration, rather than budget alone.

What's Coming Next

Jeremy notes the current platform is version one, with more statistics, additional data visualisations, and general usability refinements planned as they continue to gather user feedback, encouraging users to submit feedback directly via the platform's chat/feedback icon, or seek help via the same channel if they get stuck. He also teases an upcoming new content series without giving further detail, timed loosely around the podcast's upcoming 50th episode.

Closing Thoughts

Damien and Jeremy close by directing listeners to the new suburbdata.com.au platform, noting sign-up and trial-mode use is free, with charges only applying once a user pulls real data. They encourage feedback via YouTube or Spotify comments (noting Apple Podcasts doesn't support comments), likes, and subscriptions for further tutorial content.