Creating great financial software – part 1
Over the last 15 years I have worked at a number of leading organisations creating, developing, deploying and managing global trading platforms. I thought that as this is my first blog post in my capacity as a mentor for SBC Fintech, it would be useful to share some insights in a series of two posts (just to keep things digestible).
Seemingly dull businesses can be very profitable.
Near the start of my career I worked for a small software company that built connectivity applications for banks to various fixed income marketplaces. At the time there were a lot of these marketplaces being created and the banks were all too keen to outsource this connectivity function whilst their in-house developers focused on developing risk and trading features which were considered to be where money could be made. This company built a great reputation very quickly for developing connectivity to these various markets as they emerged. The great thing here was that they could build a connection to a market for one client and then sell the same link to all the other banks that had a similar need. The software was not necessarily that complicated or mentally challenging in an academic sense but it was providing vital plumbing for other applications to sit on top of. Forward to today and that company has grown from 50 people when I worked there to over 1000 and is involved in all aspects of the trade lifecycle. It was an excellent model for the company to choose at the time and I learnt a great deal. The lesson here for someone considering a fintech start-up is not to overlook the functions that can seem less interesting, they are normally the ones that have not evolved over a long time and are ripe for disruption.
Understand your customer, but don’t get hung up on pleasing them all.
Customers don’t always know exactly what they want, they will have an idea but it will be down to you to figure out the details. Prototyping helps a great deal as a way to hone in on the exact need, always get as much feedback as possible but do not listen to all of it, be selective and try to find repeating common patterns between your customers. Remember you are building for the 80% and if you get lucky you might also please some of the other 20% but sometimes that can just be a bonus.
Think about integration first not last.
It is very rare for a software application to operate in isolation inside a financial institution. For example, a trader’s pricing sheet will integrate into a bank’s middleware platform. An e-trading platform may connect to that middleware platform and distribute prices to client facing electronic screens. Customers will attempt to trade on those screens but will require risk limits in place set by separate risk departments with their own technology. Any trades generated will need to be booked to downstream middle and back office systems. Those trades might need to leave the organisation completely for confirmation matching by 3rd party companies. You get the idea. This presents two considerations worth noting for any fintech start-up.
Firstly, you will likely be selling an application to a customer that has to integrate this into an existing ecosystem. You should offer an application protocol interface (API) from the very start to allow them to do this from day-one. One approach I endorse where it makes sense is to create your API first. Then build your interface on top of your own API. Aside from giving customers confidence in your API, as you are eating your own dog food, it sets the right mindset inside your company from the start.
Secondly remember that a change in any part of the workflow could have an unforeseen impact elsewhere. Whilst you will be eager to get out as much functionality as possible remember that your clients will typically be more risk-averse than other industries, don’t forsake quality for functionality as in the long run it will not pay back. Also assist your customers where possible in end-to-end testing when you release new features, it is a great way for you to learn about other parts of their business and can create opportunities for new products down the road.
Embrace the open.
Look for open standards in the particular vertical of fintech you happen to be targeting. For example FIX and FpML are both used extensively by banks, asset managers and hedge funds in the securities market. If you find there are no standards in your niche then that could be an opportunity for you to work with market participants to create some for the benefit of the entire community. This altruistic idea might seem like a hard sell to your investors, however it can have great payback if you are seen as pioneers and are thought leaders by your future customers.
That’s it for this post, happy to take questions and comments and keep and keep an eye out of the second instalment.
(All of the opinions above are solely my own and not of my company)