HtAG Platform Overview Video Pt.1

This video introduces property investors & home buyers to the HtAG Analytics platform. We explain how to apply HtAG proprietary set of tools to perform property market research, covering most of the current features available. Every feature will be explained in detail via the screen capture, which is the format of this video series.

This is the first video in the series, in which we explain the following:

  • How to use LGA map of Australia to measure council area capital growth rates
  • LGA/Suburb Ranking table and the use of filters / optional column toggles
  • Confidence levels and their significance
  • Research considerations when using the ranking table (with an example)
  • Introduction to GRC (Growth Rate Cycle)

1 thought on “HtAG Platform Overview Video Pt.1”

  1. Video Transcript
    For the purposes of this demo we have logged into the portal using a professional subscription. There are three different tiers of our subscription, which we will delve deeper into later. But for the time being, the reason we’ve logged into HtAG on the professional tier is because it provides us with all the options and all the data available on our Web site, which will make it easier for us to contextualize our services and provide the background information in its’ entirety.

    The first thing you see after logging is the map of Australia. The coloured shapes represent LGA (Local Government Areas) or council areas for which we have collected property market data. Each colour represents different rate of growth. The map in itself can be used for various purposes which will be covered in detail separately.

    But, to give you a brief overview of this section of the homepage – you can easily switch between different map views for houses and units. As you can see, there’s much less data collected for units because houses are a dominant dwelling type in Australia. You can click on the blue icon to see LGA information for the median price, capital growth, number of sales metrics. The data is presented for the current calendar quarter. So on and so forth.

    Maps are a useful tool to evince current hotspots country-wide. The colour scale ranges from dark red to dark green with the highest capital growth rate areas positioned in the far-right spectrum of the scale.

    You can zoom in and out on the map in a number of different ways.

    One way to zoom in is by clicking on the plus and minus in the top left-hand corner. Or you can zoom in on an area of interest in, let’s say in New South Wales, by hovering over the number 9. This number represents the 9 council areas inside the connecting shape. Because we are zoomed out it is simply not possible to show all the intricacies of the council geographies at the current zoom level. When we click on number nine, it enlarges the area and it provides you with a more detailed delineation of all the councils within that particular region of the map.

    In this section of New South Wales, we still have some numbers on the map. 32 here, which looks like Sydney. If I click on it, it actually zooms in more, so on and so forth. The little pin that stands out and says “click me” provides you with the name of the council, and the property metrics for current quarter, which is quarter two of the year 2020 at the moment. Year on year growth in Ku-ring-gai is shown, which means that from 2019 quarter 2 to 2020 quarter 2 in those respective years, there’s been a negative growth of -1.27%.

    And then we have the number of sales as well for the current quarter. The map provides you of an overview or the current snapshot of the market. However, the main benefit of all this information comes through when we scroll down and are presented with the ranking table. So essentially professional subscription allows you unlock data for every single council and suburb that we collect data for.

    The data in the table is arranged based on pre-defined boundaries that you can change based on your needs. At the top, there are several filters, which can be used to shuffle through all of the data and single out particular council areas or suburbs. You can filter through by median price in a number of thousands, by median yield and confidence.

    Just to delve a bit deeper about confidence, essentially HtAG provides three different confidence levels, high confidence, median confidence and low confidence.

    Confidence levels relate to the algorithm that we’ve implemented to calculate median values as well as growth rates and cycles. The high confidence areas have an error rate of up to 5% or less, median confidence of up to 10% and low confidence of up to 15%.

    The confidence column is very important when it comes to deciding high quality investment areas and later we will contextualise the use of this column with examples. To put it in perspective, if you click on the confidence heading in the column, it sorts the data by the values in this column. And you can see there is a fourth – very low confidence level.

    Very low confidence signifies that there’s simply not enough sales in the area for reliable median value and forecast calculations. However, since some data is available it is shown despite the fact that the error is not measurable. We are transparent about our data and chose to list very low confidence areas so that their data can be contextualised against other areas in the table.

    We list areas with at least 30 historical sales recorded. In the past we chose to only list areas with significantly higher number of sales but decided to reduce this threshold after we received feedback from our customers that when they use the site search, some council areas and suburbs are not found. The new threshold was introduced on the 20th of June 2020 which resulted in approximately 1,500 more suburbs appearing on the site.

    So to ease the searching process, we’ve included nearly all of the council areas and suburbs irrespective of the fact whether they have a significant number of sales, whether they’re good investments or not. At the moment, we report property market data for 6,180 suburbs.

    So, essentially there’s been a dramatic increase. And even though, majority of the added suburbs have low and very low confidence, it’s good to have them to benchmark different areas against each other. A lot of investors buy in areas because they live close to these areas or they heard that certain suburbs are good to invest in for whatever reason. But when you put it in perspective and you see that the confidence level for a particular area is very low, you would need to vet your own thinking process, because very low confidence areas are frequently poor investment choices.

    And when you compare that against areas which could be in close proximity or in the same state or on the other side of the continent, but have higher confidence you can make a more informed decision based on reliable data on areas with more active property markets.

    Going back to the ranking table, there are several filters here that you can use to input information, which will change the output in the table. You can filter by maximum median price, minimum desired yield, confidence, and the number of bedrooms. On the other side, you can also switch between council data, which is the default view, and suburb data.

    And you get all of the suburbs within Australia and all of the associated metrics. And finally, you can switch between houses and units in the same way.

    There are 15,000 or so suburbs and localities in Australia. As you can see, we’re looking at units now. There are fewer suburbs on the list now and the reason for that, is that units are commonly only sold in highly populated urban areas, whereas houses are also sold in both urban and regional areas. So there’s more data for houses than there is for units.

    [00:10:36.630]You can see that the summary information fur number of rows is provided underneath the table. If we go to houses, you’ll see that the number of entries is actually significantly higher than that seen when the units toggle was active. Note that it can take a couple of seconds for the data to be refreshed in the table when toggling between councils and suburbs due to large volumes of data.
    We can see that it filters from 9,676 because by default it shows you all bedrooms. We know the total number of dwellings when we add the number of dwellings from each filter, that is, from one to two, there, four and five bedrooms combined. This means that the filter enables one to see the median price per number of bedrooms but also a median price for all dwelling types combined. This becomes pertinent when one assesses the investment potential of a single property in a particular area as they are in a position to compare the median price of a particular dwelling type with the property in question where these characteristics match. As such, if we look at the number of entries retrieved when filtering for 2-bedroom houses, we can see that there are three hundred and seventy-six entries with statistical significance for two-bedroom houses.

    On the other hand, if you look at the three bedrooms filter, you’ll notice that there’s a lot more and so on and so on. The ‘all’ filter has the most entries because it combines all dwelling types and can be used as a suitable guide for suburbs that do not have a statistically significant number of entries. As some suburbs and some dwelling types don’t have enough sales, the only median price available to gage the suburbs potential is the median price for all number of bedrooms combined.

    So going back to how we can use all of this. You use HtAG by just browsing through the map, trying to look for green areas, which suggest a high level of year on year growth. The red is an indication of a decline or negative year on year growth. As such, the amp provides a snap shot which can be used to ascertain growth clusters and growth corridors. This information can then be used to enquire further by probing into particular localities with additional HtAG features.

    But you can also come to HtAG with a preconceived idea of which section of Australia suburb or council you would like to look at. Because let’s say you were either provided information from a buyer’s agent or a property professional, which you would like to vet by using our platform. We think the best way to explain all of the services offered is to use an example.

    To make things simple lest use Camden Council as a case study. However, it is important to highlight how one could have come to take interest in a particular council or suburb: In this instance, I might want to invest in this area due to its proximity to my other investments that have performed well.

    Or I might want to invest in a suburb in this council because of the advice from a buyer’s agent, whose rationale I would like to vet by doing my own due diligence using the HtAG platform. Essentially, Camden Council, or any other area for that matter, should also be considered in terms of my investment strategy, and/or existing portfolio composition. I could also choose a particular council by using different filters, which best suits my investment criteria as an individual.

    This means that it is beneficial to have intimate knowledge of our existing portfolio which will guide our decision making. For example, let’s say that we have a portfolio which is more geared towards cash flow, and as a result we are looking to buy a house in an area which will experience higher capital growth just to balance the cash flow geared portfolio. As such, one can use the information provided in the heat map, the table as well as the toggles underneath the table to retrieve area which best suits our investment needs. The information provided in the tables can also be used to understand the entry point of buying in a particular suburbs which is not only used in respect to the demands of our portfolio but also in respect to our current and future financial circumstances as, for example, higher entry points required larger deposits and greater financial capacity to service and maintain the property.

    On top of all of this, one can also be interested in countercyclical investing, which means that he or she would only be looking for suburbs which are at the bottom of the cycle because they are positioned for imminent growth. Essentially, when a suburb is at the bottom off the bottom its growth is positioned to move in an upward trajectory. We will go into more detail about the relationality between the property clock and HtAG’s growth rate cycle feature but juts as a quick example, by introducing the growth rate cycle feature, one is not only able to make a decision on which locality to invest in based on its existing and forecasted growth but also based on the position of the locality in the cycle of growth which assists customers to time their purchases and maximised their returns.

    For example, if we are only interested in areas which are at the bottom of the cycle, we can generate a list of those areas and then compare them based on other filters and statistics such as year on year growth, forecasted growth, year on year yield and forecasted yield as well as by using HtAG’s confidence feature to be able to make the most appropriate decision possible. We will return to this later but essentially what we are trying to highlight is that you can make use of many features combined so that the suburb you settle on matches best your financial circumstances and investment strategy. If it is your first purchase, HtAG’s platform can also be used as an educational tool in such a way that the customer becomes familiar with all considerations that need to be made prior to making a decision by simply understanding the different outputs retrieved in respect to filters chosen.

    In addition to strategy specific filters, we can also filter based on many different locality variations. You can filter by city, by state, by council and by suburb.

    Let’s say we interested in Sydney. When we choose Sydney, you can see that the number of entries has changed. So, if we choose Sydney and introduce the growth wide cycle filter for example, we can then shift through all Sydney areas which are at the bottom of the growth rate cycle. And then focus on those which best matches what we need by drilling deeper by way of additional HtAG pages.

    Going back to our case study of Camden Council. We will come back to the suburbs and other level searches later but for now, let’s take interest in Camden. Camden Council is located in South West Sydney.

    To simplify things and address every option and filter individually in reference to the case study area, we will remove the growth rate cycle toggle total because we will address each and one of them progressively. So, this is what we get when we search for Camden.

    The search result returns a table which contains property statistics of Camden Council. The information provided highlights average statistics for all types of dwelling, irrespective of the number of bedrooms. There has been 19 number of sales in quarter two for Camden Council.

    The median price $807K. Its year on year percentage change has been -1.5%. As you can see, when we hover over the year on year percentage change, we can see what that means. It is the percentage change in the median price from quarter three in 2019 to quarter three, that is, current quarter in 2020. The gross yield is 3.3 %.

    Just for information purposes, gross yield is derived by dividing the total annual weekly rent with the price of the property. In this instance, the total annual median rent with the total annual median value of property came in Camden Council. As such, to calculate gross yield we would have to know what the median rent is for Camden Council in order to be able to understand how the provided statistic has been retrieved.

    This is where we come to the optional toggles here that I have referred to before. The toggles are available at the bottom of the table, which when chosen or clicked on, introduce additional columns to the default statistics table for Camden Council. As we can see, there are one, two, three, four, five additional toggles in total.

    Returning to the gross yield elaboration, if we wanted to do a sanity check, we can click on the median rent toggle and introduce a column to the default table that highlights median rent for Camden Council being five hundred fifteen dollars at quarter three 2020. We can also introduce additional columns such as median rent you and you change. When we do this, we can see that median rent has decreased by 2.85% from this quarter to last year. Should we be particularly concerned with all rent related statistics, we can go further and introduce two additional rent related toggles such as ‘projected median rent’ and ‘projected median rent change’.

    This is where we come to the second part of the table, one delineated by the blue background and associated with HtAG’s forecasting algorithm. Information under ‘Current Values 2020 Q3’ heading contains current and past statistics for Camden Council. On the other hand, information under the ‘Projected Values 2022 Q3’ heading contains forecasted statistics associated with the median price and median rent. So, as we can see, the projected median price for Camden Council in quarter 3, 2022 is 799K. This means that in two years Camden Council median price will experience a decline of 8K.

    Which is -1.06% negative capital growth. By clicking on the ‘projected median rent’ toggle we can see that forecasted rent will be $493 which is a -4.27% decline in comparison to current median rent.

    To see the negative growth in rent we have to introduce another column by clicking on the ‘projected median rent change’ toggle.

    By looking at these statistics, we can actually anticipate that Camden Council either at the bottom of its growth cycle or is at the decreasing stage about to reach the bottom. This is evident when we compare the year on year percentage change which is -1.5% and the forecasted capital growth which is -1.06% suggesting that the negative rate of growth will be slowing down for Camden between quarter three 2020 and quarter three 2022. In order to affirm this conclusion, we can introduce the ‘growth rate cycle’ toggle which supports our interpretation and shows that Camden Council is indeed at the bottom of its growth cycle. Now that we have introduced all toggles, we are provided with a broad snapshot of Camden Council investment statistics. Finally, we can also see that the confidence in the provided statistics is marked as high which suggests that there has been lots of purchasing activity in Camden Council which should urge any investor not to discount Camden Council as a potential investment locality.

    We can see that in quarter three alone, there’s been 19 sales. We will take this opportunity to explain what the confidence feature stands for and what information it can provide to customers. The usually questions we receive from customers is how the confidence level is generated and whether it is solely based on the number of recorded sales.

    The answer is that it is primarily connected to the number of sales. Obviously, the more sales there are, the more data there is which enables our model to predict with higher degrees of accuracy. As a rule of thumb, we commonly need around 30 sales per Council (also referred to as LGA) for high confidence forecast or 10 and above for suburb level forecasts.

    However, the model does not only consider current sales—It also looks at historical sales including performance in all previous quarters. So, for example, when you look at sales data for Camden, we can see that Camden has historically experienced quite a large volume of transactions. As such, although there are only 19 sales in the current quarter, the fact that we are only at the beginning of the quarter suggests that based on current and past performance Camden will experience further increases in activity making the projections one of high confidence.

    Medium confidence data can be relied on quite safely. There is a slightly higher error rate associated with it, so it may be prudent to perform a sanity check on the relevant council area or suburb. For example, you’d want to ensure there are no sudden rises and falls in historical median values on the median price charts. Low confidence data is for those with higher risk appetite. The most common reason for low confidence data is poor sales performance in the given suburb or council area, so there is not enough data for us to provide good enough median values and forecasts.

    Confidence can be best described as a measure of error. And calculation of error is best explained with an example. Let’s say we have a suburb with a current median price of 300,000 dollars. And our two-year forecast indicates that two years from now, this price will be 320,000 dollars, which is equal to 6.67 percent growth.

    Two years from now, we measure the actual price and see that it is in fact at 324,000 dollars, which equates to 7.41% growth. It’s obviously 4,000 more than what we forecasted. The way to calculate the error is to subtract the forecast percentage growth from actual percentage growth. So, in this example, when you subtract these 2 figures from one another, you get a difference of 0.74%. Therefore, the error is 0.74%.

    Since the error is under the 5% threshold, this data along with the forecast can be classified as a high confidence forecast.

    As such, although high confidence localities equip clients with an enhanced capacity to anticipate the locality’s future median price movements and thus improve decision making, it is important to note that one should not discard medium and low confidence localities. This is because the confidence data should not only be used to determine which areas to invest in, but also in case of an inverse scenario—to determine which area not to invest in: a measuring stick so to speak defined by our risk profile. For example, one will come across localities that have experienced 20% year on year growth however when considered against the low confidence feature, they will come to a realisation that the locality’s sales have been rather sporadic and minimal per quarter making it a high-risk investment locality irrespective of its recorded year on year high growth. On the other hand, implications of the three-tier confidence feature can also vary depending on the median value of the property for a particular locality as well as our initial contribution or deposit. For example, minimal deposits for higher median value localities marked with medium confidence can still have a profound impact on our return on investment irrespective of the plus or minus 10% error rate.

    Most importantly, however, as we collect data in near-real-time, that is, as our data is updated forthrightly, the confidence interval for a particular locality can change from one fortnight to another or from one quarter to the next. This means that a low confidence suburb or Council can suddenly be marked as a medium confidence locality in the subsequent time interval son on and so forth. This is what makes HtAG’s service offering so exciting and what differentiates it from other service providers in the sector—it gives fluidity to the statistical representations of different localities which inadvertently makes the data more relevant and real so to speak. This is not the case with the report representational model that the property advisory space uses predominantly, that although useful, provides only a snapshot of a particular moment in time insights from which can diminish in relevance as time goes by.

    This leads us to another very important point—no feature should be used in isolation but should be considered in relation to other features and subsequent statistics for a particular locality of interest. This means that one should not only focus on the confidence feature although it represents an integral part of our forecasting model but should also consider the confidence feature in relation to the growth rate cycle feature for example. As such, in the same way that high year on year and forecasted growth feature is essentially meaningless if not considered in relation to the confidence feature, the high year on year growth, the high forecasted growth and the high confidence statistics combined can still render an area as being of low investment quality if the area is at or near the peak of its growth rate cycle for example. This would suggest that such an area is positioned to enter the decline stage of the growth cycle reducing the customer’s ability to leverage capital gain into additional purchases.

    So, all features should be considered in unison. Also, each feature can carry different meaning and relevance depending on one’s personal circumstance and risk appetite. As such, the information one considers important will affect not only whether one invest in Camden Council for example but also how much one is prepared to outlay to enter the Camden Council Market. Using the previous example of the arrangement of features, the fact that Camden is forecasting negative growth could potentially not mean anything to an investor whose decision making is driven by area’s position in its growth rate cycle, that is, a person who invests in areas that are at the bottom of the growth rate cycle aiming to take advantage of subsequent growth.

Leave a comment