Full Stack vs. Marketplace Model
One of the merits of working in a fast growing startup like Spinny is that I get to meet a lot of new people as a part of our hiring process. It is an opportunity to learn about experiences of others and to tell our story — a privilege that I truly enjoy. One question I often get asked in these interviews is why did Spinny choose a full stack model vs. an asset light marketplace model. Don’t institutional investors prefer asset light models that can be scaled fast and need lesser capital to grow? Aren’t some of the biggest and most popular success stories from startups that of asset light marketplaces — think Uber which owns no cabs, Airbnb which owns no houses or hotels, Zomato/Swiggy that own no restaurants (well almost) and don’t employ the delivery partners? They don’t take the inventory risk, don’t employ the service providers and need lesser capital to scale. Yes, we know and we have been there, done that (we pivoted to a full stack model in 2017). And, yet today, we take end to end ownership of every step in car buying or selling experience for each customer. From the test drive to financing options, from the inspections to car reconditioning — all processes are owned by us. So, why did we choose this path? Please read on to know why.
Marketplaces solve the problem of discovery and make the transaction process seamless by standardizing it across various service providers. For high frequency, low involvement transactions in industries where the supply is disaggregated, this model adds a great deal of value for customers. Let’s double-click on this.
You do one time processes of downloading the app, creating an account, setting up your address and payment information and post that you can transact seamlessly across a variety of service providers (think restaurants, cab drivers, B&B hosts). After all, you don’t want to eat from the same restaurant every time you order and you can’t be served by the same cab driver every time (even if you’d prefer that). And, doing this across each individual service provider for every transaction would be a lot of pain.
Secondly, marketplaces aggregate a wide variety of supply. In fact, the more disaggregated the supply is, the more value a marketplace adds. For ex: OTAs (Online Travel Agents) earn significantly higher commissions on hotels than flights (fewer airlines in each market vs. a large number of hotels). People crave for variety when it comes to food or their travels and if the same company was to run everything they would need to standardize it a lot more to scale and who wants that! Since the transaction volumes are high, user generated reviews help other customers in decision making amidst the wide variety of choices available.
Thirdly, the lifecycle of involvement with the service provider (restaurant, cab driver or B&B host) also ends soon after the transaction and the stakes are low. I have had my share of rude cab drivers and uninspiring food, but there have been many more pleasant experiences as well. I don’t remember most of these experiences today (barring stays at some AirBnBs) as the need they were solving was very transactional.
However, things change when you think of a car buying transaction that is high in value, infrequent and with more emotion involved. This is a high involvement purchase where customers are willing to spend more time exploring options and don’t want to rely on the ratings of other people to put down a large sum of money (which is also financed using a loan in many cases). The customer may not buy another car for 3–4 years or more (modern cars can run well for many more years) and hence, building comfort with a particular interface or having saved addresses or payment information is not a key priority. But, ensuring that someone has their back after the transaction is done is critical.
Similarly, the number of car models being bought and sold are much fewer and an ideal used car is as close as possible to how the OEM shipped it — customers don’t crave for variety here. Once you have a few units of a particular model and variant of a car on the platform, the marginal utility of having more of those is not as high. A customer would rather choose from a few high quality options of a particular car model than have to evaluate between a much larger number of cars of varying quality from varied sellers (individuals or used car dealers). In such a scenario, having control on all aspects of the transaction lifecycle and giving the maximum comfort and convenience for that one transaction is what matters the most. And, in our experience, till we can fully standardize all aspects of car evaluation, refurbishment, pricing and sales using technology and crisply defined SOPs, customers are best served through a full stack retail platform with fewer independent entitities that can make or break a customer’s experience with buying an owning a car.
At Spinny, we understand that each of the thousands of car transactions we do every month is very special for that customer. For many of our customers, it is their first car and for all of us who have owned a car, know the joy and memories it brings. We are committed to providing a seamless and delightful experience to our customers and we will do whatever it takes to do it. We want to establish the right standards and infrastructure in the market and set a new benchmark in the retail experience of used cars. Building a full stack retail platform is more intense, needs more work but it is the right thing to do for delivering the best customer experience.
If you are keen on exploring roles in Product Management, Engineering or Design, please reach out to us at techhr@spinny.com
