Pittsburgh: Potential areas for setting up an Office Space

Made by adobriya

Created: October 29th, 2017

0

Motivation

What if an Entrepreneur (Say me) and wanted to set up a small working space in Pittsburgh for a startup? As its a small firm, employees would have to use outside resources for food and places to hang out during breaks. What would be a good location for me to set up my office? I try to find this out by looking at the following data.

Raw data-set used

Four Square data for Pittsburgh, listing places to shop and restaurants with ratings, check-ins. Foursquare also points out trends and this could be a tool as well. Twitter data about how people feel about a certain place or what are their 'moods' when they are at a certain place would also help determine a balanced choice.

Objective

Using the data about restaurants and shopping from Foursquare, hotspots would be created at locations around the city and a list could be generated of possible spots for the office space. Based on walkability to amenities, a preference list would be created for the top 6 picks (1st, 2nd, etc.). Twitter data can be used to verify and validate this information as twitter users are known to be direct and unbiased. This data, when combined with property cost data from earlier research could then be used to find a perfect balance between utility & costs and choose a suitable location.

Rationale for approach

It is very important to look at the availability of offices and their accessibility. Also, Setting up a new office would mean that the employees will have to rely on outdoor amenities for food and leisure and hence these were looked into. Other factors such as shopping and recreation were also considered as employees would prefer is they could use these once they are done with office and their proximity would definitely help.

Overlapping these maps gave a bigger picture combining all the considerations and hence added value to the analysis by the virtue of ‘A whole is greater than the sum of its parts’. A walk-able service area seemed to be a good tool to gauge the distances highlighted by the overlapping hotspots.


Data Set Assembled

An overlay map was prepared by overlapping each of the following layers on the map of Pittsburgh in the form of hotspots:

•Professional Spaces – Co-work spaces, office buildings, parking, event space, corporate amenities
•Rental housing – Apartment/Condo, assisted living
•Transportation – Bus Stops, Bus stations, Bikeshare
•Food – Places in Pittsburgh with at least a thousand check-ins (only popular places)
•Nightlife - Places in Pittsburgh with at least five hundred check-ins (only popular places)
•Shops and Services - Places in Pittsburgh with at least two hundred check-ins (only popular places)
•Recreation and Parks – Gym, Gardens, Parks, Fitness Centers
•Hotel & Extended Stay – Hotels, Hostels, Bed & Breakfast


Each of these datasets was filtered based on the above said contents and imported as a new csv into carto as a separate layer. Each of this layer was made transparent and then overlapped to create a combined hotspot map. This map was then used to draw service areas of 0.5 mile radius from the chosen spaces.

Aspect of Focus: Amenities within 0.5 mile (roughly 5 minute walk) radius

The next slide shows a sample screenshot of the data set used for the layer 'Food'. The following two slides show the different layers to be overlapped. The graphic is recreated in black and white to show contrast and highlight the hotspots.


0

The Overlapped Map

0

The below Pittsburgh Neighborhoods Map can be used to empirically identify areas of high probability of setting up the office in. Some of these areas can be identified as Central Business District, Strip District, Southside Flats , Northshore, Oakland and East Liberty Corridor.

0

These empirically chosen locations were then used to identify Co-work office spaces in each of these areas. Only locations with a minimum of two co-work spaces were chosen. These spaces would be used as reference points to map the 0.5 mile service area onto the visual using carto.

The spaces chosen were:

•Alloy 26
•Space
•StartUptown
•Revv
•Alphalab
•Work Hard Pittsburgh
0

These were then georeferenced and the dataset was added to carto to produce a map as shown above.

Using layer analysis ‘ Create Areas of Influence’ , service area of 0.5 mile outer radius was created and the following slide shows these service areas overlapped onto the illustration created earlier.

0

This illustration below was created to identify the most suitable areas for setting up an office space. The illustration following this shows areas in the order of preference based on visual insights.

0

As Seen Below : Final illustration with areas in order of preference

0

Visual Insights Gained

As per the above illustration, the areas in order of preference are:

1.Central Business District (Space)
2.North Shore (Alloy 26)
3.Oakland (Revv)
4.East Liberty Corridor (Alpha Lab)
5.Uptown (StartUptown)
6.Southside (Work Hard Pittsburgh)


General Insights Gained

1.The Downtown Business District is the most sensible choice with office spaces, great transportation and good number of options for food.
2.North Shore & Uptown make to the list due to the proximity to downtown. Although Northshore has good office spaces, it lacks other amenities.
3.Even though areas such as strip district and squirrel hill also show up as hotspots, they do no have a good number of office spaces and are therefore not considered in the final list.
4.Oakland and east liberty have the highest number of co-work spaces. This could be due to the presence of professional schools in the area or could also be an indication of a new trend where new corridor developments incorporate more shared work spaces.


Next Steps

This puts us a step ahead in deciding on a place for the office. However, certain other things scan be done to validate our choices even further.

1.Twitter data can be used to identify commonly used adjectives in order to establish the general sentiment in an area. This could also be used to eliminate areas with a detrimental sense of sentiments.
2.Data from WPRDC can be used to map median property costs in the area and these can be used to eliminate any spaces which are too expensive to be used. However, the scope of this proposal does not deal with the financial analysis of the office space. Therefore, currently it can be assumed that finances are taken for granted and any space can be afforded if need be.
3.Apart from creating a service area of 0.5 mile radius, walkability in terms of time taken (5min walk, 10min walk etc.) were also looked into. This analysis however, could not be completed due to technical difficulties. This would form the next step.


Additional Insights Gained via Twitter

Top 5 adjectives within tweets from the areas shortlisted in the previous section:

1.Central Business District - 15222 – great, fit, practical, seasonal, ready
2.North Shore - 15212 – great, best, good, happy, big
3.Oakland - 15213 – great, natural, closed, hot, beautiful
4.East Liberty Corridor - 15206 – great, latest, drinking, sunny, fit
5.Uptown - 15219 – like, good, great, happy, right
6.Southside - 15203 – drinking, great, good, exact, pretty

The most common adjective across these zip codes is ‘great’ followed by ‘good’. There is not one negative sounding adjective in the top five across these areas. However, one might make a case for ‘hot’ being a negative word but it does not convey any common sentiments.

As can be seen from the data above, there seems to be no adjectives pointing at detrimental or negative sentiment in any of the six neighborhoods chosen. Hence, there isn’t a need to eliminate any of the chosen areas based on this data. This data validates to some extent the analysis performed in the previous sections from the point of view of Twitter, where people do not shy away from expressing their feelings and emotions.

x