Data Analysis

What KAPS Tickets Tell Us About Malls in Nairobi

If Nairobi is experiencing a glut in shopping malls, then KAPS would the first to know. The company enjoys a near monopoly in managing parking lots for malls in the country – and with that, collect tonnes of data on mall traffic. I spent the better part of the December holiday and early this year rummaging through mall trash bins to find this data.

The escapade started off in Nanyuki town at Nanyuki Mall, then to United Mall in Kisumu, through to Zion Mall in Eldoret and finally to eight malls in Nairobi. In the excursion, I came to realise that the ease of finding disposed tickets varied with each mall. However, a pattern started to build up – malls with outdoor activities had fewer disposed tickets than the ones without. So I started paying attention to how each mall is utilising its extra space.

If you are observant, you’ll notice that Sarit Center, Yaya Center and The Junction Mall recently expanded their parking space. These malls would have opted to use their extra space to host bouncing castles but they didn’t, which points to the priority given to parking. The Junction Mall just acquired Riara parking lot next to the mall to increase its parking capacity. On the flipside, when a mall has more outdoor activities it screams ‘we don’t have many cars coming in’. Next time you spot a mall with lots of space allocated to outdoor activities, do know car traffic isn’t doing good. Example of such malls include:

  • Greenspan Mall
  • Garden City
  • Prestige Plaza

On the way to collect these tickets at Yaya Centre, I found out that the mall does not use the services of KAPS rather a competitor known as Pay-N-Go. I was rather disappointed for the break in analysis consistency but I came upon an interesting piece of knowledge. I struck a conversation with someone privy to shop leasing at the mall and he told me that the management use cameras within the building to track foot traffic on each shop and use that to price the monthly rent. If foot traffic goes up, the rent goes up. I was impressed – it also shows if the management presents these numbers to the tenants then they are confident the mall is doing great.

The Data

parking_tickets

Once I had the data nicely entered into Excel I began to think what sort of analysis to conduct. The idea was inspired by an economic measure known as Price Elasticity of Demand –  it tracks how much the demand for a commodity changes for every unit change in price when all factors remain constant (ceteris paribus).  If you consider parking space in a mall as a commodity, you can measure how much is in demand by the amount charged for it and the effect on it utilization. From the data, two malls (Adlife Plaza and The Green House) increased their minimum charge to Kshs 100 for the first hour while The Junction, Prestige and Yaya charge Kshs  50 for the first hour.

In consumer theory availability of substitute goods (other malls) that are cheaper should drive people to other malls and reduce demand from the high priced ones. However, this is not the case, Adlife Plaza processes 1 car every 7 minutes on average. To extrapolate this phenomenon, part of the underlying reason could be there are unique brands outlets that can only be found at Adlife plaza. A good example is NewsCafe restaurant, if another mall would be built that would host NewsCafe, there would be loyal adherents who will follow the brand. Given Adlife Plaza doesn’t have space for outdoor activities, it thus caters for a different demographics.

The Behavior

arrival_time

On average, people spend 65 minutes at a mall with most mall entries happening between 3:00 pm and 6 pm, and another spike at 8:00 pm. These times can provide a proxy on what people do at malls. The high arrival times in the evening points coupled with the average time spent shows most people go to grocery and house shopping. Another interesting phenomenon is that 12 percent of the tickets had cars that spent less than 10 minutes in the malls – I formulated an opinion that these must be Uber pickups, perhaps from folks who have shopped and need to carry the heavy luggage home.

The other surge in arrival happens at 8:00 pm, a good hypothesis would be folks who partake in evening culinary delights. I suppose data from Java House and Art Cafe can corroborate this observation.  So far, it is right to say most malls in Kenya are supported by the anchor tenant and food outlets. Another factor that’s worth of interest connected to a mall’s performance is the denomination used to pay the parking fee.

In a concept known as Denomination Effects, a theory is put forth that states, people spend more when they break a large currency into the smaller denomination. Unofficially it is known as The New York Street Tax. From the KAPS data, people start paying the tickets with large bills from 3:00 pm as shown in the diagram below. This means more spontaneous spending in the evening. Thus a surviving mall must exhibit the above characteristics. Who’s not in the pattern?

ticket_change

Foursquare has data on the time with most popular check-ins. I pulled the data several malls to compare their peak hours. The Junction Mall pattern is similar to the KAPS data and Prestige Foursquare data matches The Junction which indicates similar use. However, The Green House mall has a different pattern, the peak hours are 9:00 am, 1:00 pm, and 4:00 pm. This suggests the mall isn’t attracting the same crowd – the 9:00 pm peak suggests office arrivals. It can be concluded then that The Green House has more office than recreational facilities. These include private doctor offices, pharmacies, lawyers et cetera.

photogrid_1486663576555

The Verdict 
There are 3 types of malls in Nairobi, the family outdoor friendly, the office block mall and recreational non-family mall. Do we have a saturation of them? – Not yet, we need to have at least three of each of these within the same neighbourhood to reach saturation. I asked an American entrepreneur whether we are likely to experience a mall glut similar to the US, his response was “the city is 15 years away from experiencing a glut”.

Access the data here: https://www.dropbox.com/s/3tulherclaevx1q/kaps.csv?dl=0

Cover image courtesy of: https://missgaceru.com

Advertisements

48 comments

  1. What if you just found a contact in KAPS to lend you the data? That would be faster.

    I don’t have data, but my gut feeling is like in the states, there are malls that will have customers and there are malls that will not have customers. I think we are reaching the place where about half of the malls are starting to struggle on income.

    Like

    1. I wanted to avoid the bureaucratic process of requesting for data. That would be make it much faster. The downside of sharing the would be the possibility of analyzing the financial position of a company and sharing it with competitors.

      Same situation in Kenya. Location still plays a major role in driving customers to a mall. The concept is to build different types of malls that serves local needs instead of copy-pasting models from affluent neighborhoods.

      Like

      1. Data privacy issues too are a critical factor. one cannot just walk into an office and request such data. Remember that data has personally identifying information (like car registration numbers which are tied to national IDs and thus all your life).

        Like

      2. The printout does not have the car reg but the database has the details. Check at the gates – the moment you get a parking card, the small screen already has your car reg details.

        Like

  2. Great Evaluation.

    I also note that KAPS uses a footfall counter at sarit and junction and some Bata shops to determine the number of people in and out of the mall.

    Like

  3. I struck a conversation with someone privy to shop leasing at the mall and he told me that the management use cameras within the building to track foot traffic on each shop and use that to price the monthly rent. If foot traffic goes up, the rent goes up. I was impressed – it also shows if the management presents these numbers to the tenants then they are confident the mall is doing great…????? not so sure about this. I stand to be corrected thou, most head of terms on lease agreements are based on SQ FT. Besides not all the customers who walk in and out of an outlet are buying, some are window shopping. But good observations on KAPs.

    Like

    1. Hi Kevin,

      Yes, terms of lease are indeed based on SQ FT, but traffic is another factor that affects the price per SQ FT. That’s why stalls in town cost more than out of CBD.

      Whether customers make a purchase or not it indicates visibility of a shop. In advertising, companies sell impression and not conversion.

      Like

  4. Most Analysts have been using data from selected malls using data from tenants…..
    The problem arises when you are using data from Junction to use for say Greenspan or Juja mall..
    This is the closest I have seen to an analysis of real and credible data.
    Well done

    Like

  5. Great work. Did you also talk to the mall tenants about the foot traffic in their stores as part of your analysis? Business is terrible in most of the malls you have mentioned in the article.

    Like

  6. brilliant read!! One small correction
    ‘The Green House mall has a different pattern, the peak hours are 9:00 am, 1:00 pm, and 4:00 pm. This suggests the mall isn’t attracting the same crowd – the 9:00 pm peak suggests office arrivals.’

    I think you meant 9:00 am not PM

    Like

  7. “I asked an American entrepreneur whether we are likely to experience a mall glut similar to the US, his response was “the city is 15 years away from experiencing a glut”.”

    I dislike the fact that some “random American entrepreneur” (excuse my choice of words), can be cited as an “authority” considering there are 51 different states with different demographics and different economic conditions.

    Secondly, who is to say we will follow the same patterns they followed. Their malls were most likely built back when you couldn’t order anything from anywhere right on your smart phone.

    Like

    1. I didn’t say he/she was an authority. I merely asked their opinion. We have lots of similarities as well as difference. If they are having a mall glut, it is worth finding why and avoid the same pitfall.

      Like

    2. Great observation Johnny.

      That ‘American entrepreneur’ ‘expert’ comment doesn’t quite wash. Having read the author’s response to your observation, this has made me certain that the ‘expert’ comment was slipped in as an ‘authoritative’ conclusion on the author’s analysis. The evidence of this is clearly seen in the concluding remarks “Do we have a saturation of them? – Not yet, we need to have at least three of each of these within the same neighbourhood to reach saturation” our author has furnished us with.

      If I may humbly suggest (considering I am certainly not an authority), Blackorwa, your research is great and I believe you’ve got something great going on here, especially in how you stratify your mall data into three categories. Going by your hunch, if your research was to focus on saturation, or draw an informed conclusion on this, then your analysis inferring mall selection based on preference, or target audience (the family outdoor friendly, the office block mall and recreational non-family mall) shouldn’t arise.

      P.S: I take it you do like American examples, bearing in mind that the 2008 housing market crash was based on mortgage defaults owing to bad loans (my apologies for the simplistic summary), it would do your saturation a great deal of help if you found out if the malls in Kenya (existing and those to be built or ongoing) are bank financed. The assumption being, if the malls exist as assets rather than investments (again, pardon my overly simplified drivel, it is clear I am no ‘American expert’)…

      Like

  8. Great analysis Orwa. Did you consider the number of parking slots, as that may influence decisions that people make on which mall to visit, if they are all located within a reasonable radius? Also, some malls – because of location – such as Greenspan, TRM and Garden city may have more non-parking spaces as they expect most of their customers are walk-in anyway?

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s