Total Cost Of Ownership Of A Marketplace App – A Case Study With Swapit

In some conversations with partners and investors, I get asked what our TCO (Total Cost of Ownership) is. For those who are not sure what TCO means, here is how Gartner defines it (Gartner popularized the term in the 1980s):

Gartner defines total cost of ownership (TCO) a comprehensive assessment of information technology (IT) or other costs across enterprise boundaries over time. For IT, TCO includes hardware and software acquisition, management and support, communications, end-user expenses and the opportunity cost of downtime, training and other productivity losses.

Source: http://www.gartner.com/it-glossary/total-cost-of-ownership-tco/

Total Cost of Ownership at Swapit

swapit_marketing-dollarIn the case of Swapit, we are investing a large amount of resources into our product creation, marketing and customer retention. However, in the long run and with increasing scale, some very few factors are essential to ensure our profitability and high margin compared to our competitors. While this might feel a bit oversimplified, but from a long term perspective, our real important cost factor will be hosting charges. Hosting in terms of: bandwidth, storage, computing power. Yes, there will also be HR costs in terms of customer support etc., but with the right product, those can be minimized tremendously (just look at how Google optimized their products). So the more items we have on the Swapit platform, the higher such total costs for those will be, while our mobile app development costs will not increase on a corresponding level.

Therefore, it was essential for us to keep a very close eye on our hosting costs in the long run. We’ve built Swapit to handle billions of transactions in very short periods of time, but definitely not to cost billions of Dollars at the same time.

As of the writing of this blog post, Swapit has 111,500 registered traders, about $10M GMV posted, handled tens of thousands of comments, chat message, likes; processed hundreds of thousands of spacial queries and database operations.

Yet, our last monthly bill from Google’s Cloud Platform Service was precisely $6.82 USD.

To give you a quick overview and to show that we are on a very affordable road, here are our cloud hosting charges over the past year in USD:swapit-blog_cloud-plateform

  • May 2016: $6,82
  • April 2016: $7.73
  • March 2016: $16.29
  • February 2016: -$39.14
  • January 2016: $8.80
  • December 2015: $16.23
  • November 2015: $15.14
  • October 2015: $0.60
  • September 2015: $0.23
  • August 2015: $0.06
  • July 2015: $0.01

The numbers above are a bit fluctuating, because Google did actually overcharge us a bit last year, due to some technical issue on their side. So they actually reimbursed us in February this year.

Furthermore, we have been adding new features and integrated some optimizations to our data models, which allow us to actually lower our cost on a database size that’s rapidly increasing. To be honest, we didn’t look at the monetary costs much when integration new optimizations at our backend. Those costs were just too low to justify work on that. However, we did most optimizations because we always look for new ways to improve our user experience. Part of that user experience improvement is to process queries much much faster. Often, this comes with data model optimization or just smarter queries, which lead to much faster loading times … and more often than not, also to lower processing cost, which in turn results in lower monetary cost for hosting.swapit_marketing-chart-grow

So in the end, optimizing your own backend is good in terms of costs for short term and margin in the long-term, but also to provide a superior user experience. We believe this is one very important piece of our puzzle to attract more users than our competition.

Cost Projections

Given we are at a roughly $7 cost at 111,500 users right now, which is about $0.000063 per user per month. Let’s assume in a year’s time we managed to increase our user base to 1 million users, we’d still be well below $100 of cost per month on our platform operation side. In fact, we believe in the long run there is a need for using Swapit in all urban areas in the world. Right now more than 50% of the world’s population are living in an urban area and that trend is accelerating. So let’s be absolutely crazy and assume Swapit can achieve such a global reach (which we definitely work towards) and we manage to engage all of our users at least once a month, we’d be looking at 3.6 billion users. At that point, our cost will sum at roughly $226,800 USD per month or $2.72 million per year. Given the incredible reach we would have, that’s a very affordable cost we’d be happy to cover every month.

In short: It was an absolutely great choice to use the Google Cloud Platform for our backend.

P.S.: In the past 11 years I have been using other hosting services and I am absolutely certain, they would all have been 10-100x more expensive than our current choice.

P.P.S.: Obviously, there are a lot more costs involved to running a business like Swapit. Investing in creating a liquid marketplace and salaries are actually our largest cost positions right now and will remain to be there for the foreseeable future. However, as mentioned above, in the long their share in terms of overall costs will come down significantly.

 

Have you tried Swapit yet? It’s free. You can download it here: http://get.swapit.la

Advertisement