When visualisation becomes the lingua franca for data scientists, it is clear that we have lost the plot somewhere Image Credit: Pixabay

Real estate with its combination of the profound and the banal, the cerebral and the spurious, its autodynamic symmetries and nebulous structures, has obvious affinities with William Brinton.

He famously wrote more than a 100 years ago about the last-mile problem: “Time after time it often happens that some presumptuous or ignorant member of a committee of a member of the board of directors will upset the carefully thought-out plan of a man who knows the facts, simply because the man who knows the facts cannot present it readily enough to overcome the opposition. As the cathedral is to its foundation, so is the effective presentation of facts to the data”.

More than a 100 years later, we are still vexed by the same problem. The obvious output has been the output of “infographics” that strip the data of nuance that is inherent in the ecosystem. More than a 100 years later, we are still trying to understand how to build a bridge between people who know the facts and those that make the decisions.

Analysis fails the data

How is this possible? How did we get here? Why is it that despite so much progress in the field of data science, we are at a standstill in terms of analysis.

When visualisation becomes the lingua franca for data scientists, it is clear that we have lost the plot somewhere. This has become a deep-rooted problem in the field, and in Dubai, any data sets that highlight contrary narratives to those that the analysts are peddling get conveniently put aside, or are subject to the scorn of having agendas.

Inherent gaps

An examination of data sets on real estate reveal for example that not only is there a lag between sales and registrations, it also reveals that the lag period has increased. No one quite knows why this is happening, nor for instance why prices of certain units that appear to be identical are rising faster than others.

An easy fix to this problem would be to get the data scientists to interact with stakeholders on the ground in order to assess the “qualia” differentials that do not show up in quantitative measurements. Post-handover payments plans (which are now being dialled back after having reached their theoretical and arithmetical maximum limits) account for the single biggest reason for price differentials intra-community. But these are hardly commented on.

Make sense of the numbers

Moreover, when developers acquire a reputation for building superior (or inferior) buildings, the expectation that they will continue to do so gets baked in to their future market offerings, resulting in wide price disparities that are not obvious at first glance. All of this becomes part of the last-mile problem of data analysis.

And narratives that fail to take this into account, fail their clients and the broader market, who are left scratching their heads at their inability to comprehend what distinguishes the signals from the noise. The formation of the real estate committee has been seen as a welcome move by all and sundry. But even as it assesses the impact of various projects, it would be well served if it brings into the regulatory net, the output that data providers are releasing so as to arrest the contagious cynicism narrative that currently plagues the mindset.

The ladder of incomprehension is, at any rate, clear enough. Not being able to account for the lag between the sale and registration, not being able to account for the price difference in post-hand over payment plans, not being able to adjust for “build differentials” such as closed kitchens, we instead resort to blanket statements that the entire market or community is falling.

Even when underlying raw data continues to show a clear preference for a particular type of product and price paradigm. And even as inter-community analysis shows that prices have inched upwards (a fact that is only now being covered by some index providers).

To be sure, there is still plenty of data that is out there that suggests a plethora of narratives, and most of them are presented with arresting visuals. But sometimes, conclusions cannot be arrived at instantaneously, especially when the data set is not properly distilled for the right constraints.

In these times, if it is indescribable, don’t describe it. Just like it would be invidious to quote bad sentences from a book that has too many good ones, it is time that analysts noticed the important distinction between brilliance and dazzle.

Sameer Lakhani is Managing Director at Global Capital Partners.