All language subtitles for captions MapLav

Hi and welcome to this demo of MapLab.
MapLab empowers analysts of all levels as well as

business executives to map
and explore geographical data,

and create insights that they can share with
their organisation or with the world.

In this demo we will see how you can connect
live to a database, fetch data dynamically,

create useful data layers for analysis as well
as share this map and the output of your work

with your organization.
So today we are connected to a

Postgres / postgis database and we can see we
have lots of tables. The table I will connect

to today is five years of real estate transactions
in France and that’s millions of rows of data.

Let’s see how we can navigate this sea of data
and retrieve insights from it, with MapLab.

Today I wish to analyse transactions that
have taken place in Paris and its region.

So I just type Paris in the search bar and I can
fetch data. We can see that the map already has

data attached to it, and that’s because I can
create and enrich map background via MapBox

and load them directly into MapLab. In this
case, now that we have loaded the data points,

we can see that perhaps there’s little too much
in there. So, let’s see what we have available.

I can for example switch to a satellite map and
this will allow me to really understand the look

and type of buildings as well as whether
they’re located near parks, or busy areas.

But today I just want to load one of my

custom maps that I have added which doesn’t
have all that detail in the center of Paris.

Okay, after I typed Paris we can see that
MapLab loaded for me a very large amount of data

almost a hundred thousand transactions in fact, in
the Paris region and the suburban paris region.

Today I wish to analyse data in Paris but
not just any data, I want to understand the

hotspots for real estate, especially
for the luxury real estate market.

I’ll be playing the role of a luxury real
estate agency analyst and my agency works with

high-net-worth customers and want to really focus
on some key markets, to own those markets and have

high chance of conversions on visits. I want to
focus my excellent agent team on the highest yield

areas on the market. So I want to focus on where
the price is highest but also where there is as

many transactions as I can so that all together
I will make more commission per transaction

and I will also have higher chances of winning
transactions because there are more of them.

So we can see that by default, MapLab has
loaded my data as points in price per sqm.

I can open the layer legend and I
can see the detail to the right.

My layers are positioned to the left and I
could see here I have got points representing

each transaction via latitude and longitude
and the coloring is done via price per sqm.

So I can see that the orange parts are where the
price per square meter are highest and without

too much surprise this is the centre of Paris,
however I could already see that there is a spill

in the north west suburb of Paris so I may want to
look into this. But right now, with over a hundred

thousand of rows, there’s too much granular
information for me. I want to be able to aggregate

this data to be able to make sense of it.
Whatever data is connected, is really connected.

You have live control over it which means I can
hop to my filters and really select the criteria

that I want. The first thing is, I really want
to focus on big houses and not apartments because

they are too competitive. So I can select the
type of property and select ‘maison’ which means

house in French. Already I can see that logically
Paris has cleared out and now I am really focusing

on the suburbs. Now, again one thing I want
to make sure I do is focus on the big houses,

so I will add another filter for number of rooms
and I want houses that have at least five rooms.

Unfortunately, there is lots of transactions
happening in the Parisian suburbs and that is

still quite lot of data to analyse.
So, I will now aggregate my data.

Aggregating data is very
easy in MapLab. I will just

hide my first layer and add a new layer. I will
add it and call this ‘Aggregated Transactions’

and I have got multiple layers to do this job.

Here, I choose the type of layers
that I want to analyse my data with.

We just looked at points which
is pretty self-explanatory,

but I can analyse ‘Arc’ mapping from and
to data points, lines, grids, hexbins,

polygons and lot of data visualization
types. Here, let’s look at a heatmap.

I will select latitude and longitude and
MapLab will map this for me. Here, by default,

we are looking at concentration where most of
the transactions happen, but I want to filter

not based on the density but
based on the average price.

I can edit my colour scheme by clicking on
it and selecting a single hue or diverging.

I will decrease the radius so that we can
really see the key areas. So the north west

and the south west are very popular as well
as this area around the ‘Bois de Vincennes’

But, we can see that when I zoom out or
in, things change and that is because

heatmaps are dynamically aggregated.
But maybe I want to be even more formal

about it and define neighbourhoods,
of no more than 500 SQM and look at

them in terms of volume of transactions as
well as price of the average transaction.

This way I could really have a view of my two key
indicators for choosing a key market to focus on.

In my grid layer, I have changed the colour
and witnessed that the south-east region seems

to have the highest prices. But I will change the
radius of my layer to 0.5. It seems to confirm it

as we have very high concentration of red in
this part of the map. However, let’s now look

at our volume of transactions. For this, we
will use a ‘height’ dimension and that means

we need tot look at the map
slightly differently now.

Here, the highest bars represent the
highest number of transactions that

happened. Here I can see I might want
to reconsider the region of north-west.

There is quite high average price per transaction,
but also I can see that my average bars

are quite a bit higher than the south
west, maybe a new market to focus on.

Let’s make sure my intuition is right, we
will need to look at our points data again

to inspect the transactions that happened in this
region and make sure the prices have been high.

I can select filtering on a particular layer
and now we can really zoom in on the action.

I can also inspect the data and if I look
at the column ‘price’ which I can sort

I see that my prices have been consistently
just about or over one million euros

that’s very good, that is what
I want. I can sort this column

descending if I want to look at the best
addresses where the price was highest.

At any point in my analysis, it is
still easy to deactivate a layer

and zoom in on wherever you want to be.
I’m going to save this and send it to

the agency manager for review.
Maybe just before that I’ll add one last

dynamic filter so he can filter or replay the map
by the date at which the transaction happened.

On the graph, I can see that the number of
transactions is also increasing with time,

it’s definitely a great area for me to
invest more of our efforts in as an agency.

I can replay now the history of how
sales evolved in this particular region.

I can also click on ‘Y-axis’ here to change
the columns I want to use in my bar chart.

Okay, let’s share the map now,
I will select ‘Share Map URL’

and MapLab provides me with the option to get
a URL which I could directly send, but also

a little piece of code that I can insert directly
into an html page that could for instance

help me to publish my research on
our real estate agency’s public blog

and impress our prospect customers
with our knowledge of the market.