Finextra Research: The Pursuit of API-ness08-Dec-2014
APIs seem to be a hot topic on Finextra right now. Many companies are making their APIs ‘Open’ so they can be accessed by a variety of software developers to use it to power their own services. An example of this might be Google Maps embedded within TripAdvisor.
This is particularly interesting in the Payments space as it opens up opportunities across a multitude of opportunities including accepting digital payments more easily and securely as well as collecting valuable data.
I recently carried out a project to get an understanding of what developers, and in particular those in disruptive start-ups, are looking for from an API provider.
Let’s bring that to life with a good example:
If you’ve made an online payment recently with a start-up, there’s a good chance you did it with Stripe, although you probably don’t know it.
Stripe, effectively a start-up themselves (albeit a hugely successful one given a recently valuation of $1.7bn) are perceived to understand other start-ups and so have cleaned up a lot of the market. Compare this to players such as Authorize.net and to some extent PayPal of old and you can see why they have been successful.
And it’s not just about accepting a payment.
Kount have a collection of third party APIs designed to improve the security of online payments, a particular challenge in the US where as many as 14% of transactions have the potential to be fraudulent. PayPal / Braintree package this up as part of their SDK for online merchants.
I recently worked on a project gathering rigorous developer insight by speaking o 94 developers around the world in the USA, UK, China, India, Russia, Israel and Brazil to understand their needs and attitudes regarding payment APIs.
I also carried out an ‘Immersion Day’ where the whole client team spent an hour with 8 developers around New York City to gain a deeper understanding of their needs and see their working environments.
Confirming some hypotheses
An existing hypothesis coming into the project was that in an online start-up the developer and the business owner are either close, or one and the same, whereas this may change as the business grows.
Start-up merchants also just want to be able to accept a payment so their primary concern is setting up this functionality quickly and easily. This is why Stripe and Braintree have succeeded vs. more complex solutions.
Meanwhile non-merchant start-ups where accepting payments is not such a core feature - think many mobile or web applications here – they could be interested in payments data. However this is pretty niche and may not be on the immediate road map for these start-ups.
I’m not sure if it’s surprising or not but despite the international nature of our research we found very common needs across our developer sample – a common experience was expected by all regardless of segment or geography and effectively boiled down to 5 key areas.
Documentation: Despite many companies thinking this is a hygiene factor, great documentation is a rarity. Even the likes of Braintree identify precedents such as Twilio as ‘gold standards’ their documentation should replicate.
Sandbox: Developers expect to use a sandbox that allows them to test all elements of the API functionality.
Speed: A day lost to integration, is a day the developer is not building a new feature. Start-ups want to get into market fast!
Support: If you’re API is good then fine but developers sometimes need to speak to someone to ask a question. Don’t make them work it out for themselves.
Dashboard: The real users of the API are business people, developers hate explaining things to non-technical people – create an output that you don’t need to know code to understand.
And finally, how do developers find out about an API?
Well … Google. Or if you’re in China – Baidu. Or if you’re in Russia – Yandex.
Developers also have collections of trusted sources that may vary depending on their industry or coding language. It’s important for API providers to be featured here as that helps them lend trust and included them as part of the final product. Being present at events and running hackathons to find new uses for APIs is also beneficial – PayPal Battlehack was continuously mentioned by interviewees.
So what next?
While these are just a highlight of some of our findings, the approach was crucial in getting to understand the hearts and minds of start-up developers. In the middle of this project we saw a couple of major developments. Apple Pay was born and PayPal announced they would spin out of eBay. Payments and API are a very hot sector right now….
….watch this space.