How We Prevented App Performance Degradation from Sudden Ride Demand Spikes? : Taxi Sharing business tips
When the factors responsible for particular service change, it effectively affects the demand of the service. During the summers, there is a surge in the demand for AC servicemen. Similarly, when it pours, the demand for on-demand cab services increase. There are certain set factors that produce the demand. Your app needs to modify itself to match the surge in demand.
However, we have often seen how during a peak sale season, the ecommerce stores tend to receive maximum footfall but, they are not able to match the increasing demand. The servers fail and in certain cases the malware hits them badly, thus leading to poor experience.
Grab, a startup offering unmatched on-demand services, recently experienced a surge in demand when the rains poured in SouthEast Asia. This led to an increase in the on-demand taxi services in the particular location. The engineers, handling the backend of the cab service, seemed to have a tough day managing the peak hour traffic during this time, which lead to experience issues. Eventually, Grab found a way to manage the demand, and understood that they need to understand the factors responsible for performance of the app as well as increase in demand.
Here, we will take you through the steps Grab took to manage the demand, and how the developers enhanced the app performance.
The Challenges Grab Faced
- The company had to note all the events that led to an increase in the demand. For instance, there were local events during which the public transport, especially cab services were in demand. Secondly, in case the local transport broke down, people preferred using the cab. The company had to keep tab on all the conditions that led to the increase
- During surge, there was a lack of available drivers, which led to issues managing the traffic. The increased load and the lack of availability to manage the load led to situations that affected the app performance.
What Grab taxi booking app did to Manage the Demand Spikes?
The company was clear they wanted to manage the increase in demand without affecting the app performance, and they did it.
- Worked on a concept: The initially worked on a concept that was similar to circuit breakers. The concepts were short circuiting and adverse conditions
- In case of short circuiting, the app would perform minimal processing on a particular booking made by the customer, which would eventually reduce the load on the system
- Adverse conditions would involve a large number of unfulfilled requests for a particular service within a small period of time for a particular location.
- Designing the Filter: Based on the above concept, the designed a Spampede filter, which helped them manage the traffic spikes
- The first aim of the company was to identify the requests that have not been allocated by the app. This could be owing to unavailability of the drivers in the particular location or because of overload on the server
- The app backend then tries to go precise with the location using particular algorithm, already fed into the system. The app is not using any smart technology or sensor for the particular algorithm. They have utilized ways in which CPU efficiency can be increased and the memory usage reduced. The aim is to make an algorithm that is efficient and fast
- The next is to calculate the time for the requests. For instance, the requests are ever-evolving and the factors affecting them like traffic, driver availability and passenger needs change. That’s why the app needed an effective way to manage the unallocated requests and identify the time stamp for the same. They used Unix Timestamp, as it is efficient and saves a lot of time and CPU processing power.
- Once the timestamp and location have been identified, using Redis key the app keeps a track of the unallocated requests
- The idea is to allow the app to go live again when there is capability to allocate the requests. The app will keep studying all the factors and parameters, comparing them to the Spampede parameters. Once the maximum limit has been reached, the processing will be stopped. Till then, the allocation for the requests will continue.
Lessons to Learn from Grab for Taxi booking app development
Grab has evolved and has learnt to manage the growing traffic. There are terrific lessons that you can gain from this algorithm that they built to manage the traffic.
- Keep Evolving: Once you have defined an app solution, it doesn’t mean you have done it all. You will keep learning on-the-go, while your app is being used by the customers. That’s precisely why you should not stagnate with app development. As a company offering on-demand services, you will feel challenged multiple times. You will learn from your experience, and you need to incorporate the learning to devise the new app version.
- Design for Experience: The users of your app need to stay engaged and should enjoy a good experience, if you want them to stay on your app longer. However, if you are not able to manage the demand or, you are unable to offer them the experience they are looking for, then you will not be able to retain them longer. Always, design for the experience of the users. Understand how they use the app, what type of usability factors are important for their engagement, and accordingly design the mobile app solution
- Enhance Performance: The key to user engagement and growth of your business is performance. Make sure you have minimal to no downtime with your app. The users should be able to access and avail the app services at all times. This will help you increase customers and conversions for your business.
Grab is one of the leading on-demand mobile app solutions and Taxi booking app solutions because they keep looking for gaps in their service, request for customer feedback and work on them to deliver impressive experiences. They changed the algorithm and instead of increasing the CPU processing, they improved their efficiency.
If you want to deliver on the engagement needs of your customers, then you need to keep upgrading your app solution and manage the increasing demand and evolving needs. Coruscate is an ideal technology partner, as we offer capable solutions with our expertise and experience. With in-depth industry expertise and research on the usability needs, we offer engaging apps.
If you have an on-demand app idea or thought to start ride sharing business, connect with us for capable solutions.