Buy, Build, or ‘Buybrid’?
With Francesco Lo Conte, Managing Director, Esprow, and Chris Kingsbury, CTO, TradingScreen
With regard to trading technology, what is the state of the buy-vs-build debate? What are the pros and cons of each, and which side is the industry trending towards?
Francesco Lo Conte, Esprow: What we understand from our clients is that neither model works very well anymore.
Those who opted to build a testing framework eventually could not secure budgets that stretched for the five to 10 years required for many projects. In today fast-moving markets, nobody is going to commit money for that long, so these projects eventually ran out of money. Their problem was compounded by normal employee churn, when subject matter experts who built the frameworks leave the company and leave projects to die a slow and costly death.
On the other hand, many who opted to buy a solution for FIX API testing ended up with a black-box system they could not extend and didn’t evolve with their requirements – if it evolved at all. Sometimes this happened because the vendor didn’t listen to their feedback, other times because the vendor charged them for feature requests, but most often because the vendor focused on other products in their portfolio and let the FIX testing solution become obsolete.
I think that both Build and Buy models are no longer suitable for the ultra-fast pace of change of our global markets. This is especially so for large enterprises that famously struggle to adapt quickly.
A third type of client instead opted for Esprow’s ‘buybrid’ model, which we call our Business Partner Program. In essence, Esprow acts as an extension of the client’s development team and wallet; Esprow commits to the long-term technology investment so they do not have to, and clients get to tell us what they want to see on the product roadmap. Since most features requested are generic and benefit our entire user base, they fit perfectly well into our normal development roadmap, give clients what they want, and keep our solutions productized. It is as simple as that. And I think that we will see this hybrid model increasingly adopted in niche technology areas like ours because it is simple and it gives huge benefits to clients. It solves the build vs buy dilemma.
When TradingScreen was looking to automate testing of its FIX infrastructure, what were the company’s goals, and why did it choose Esprow as its technology partner?
Chris Kingsbury, TradingScreen: Risk reduction is always one of the main goals for TradingScreen’s FIX infrastructure. Specifically, the TS FIX integration team has been working to expand its set of regression test scripts for our top buy-side clients so that we can regression-test their dedicated FIX sessions prior to a major system release. This would help to identify and correct any unexpected bugs before clients are impacted in production. In addition, test scripts for high-volume, high-risk broker FIX sessions are needed for the same purpose as the buy-side FIX sessions – to identify potential bugs prior to a major system release.
There was no buy vs. build discussion. TS was looking for a modern FIX testing tool that met our complex integration needs, backed by a company willing to work with us. A partner that would not only support and implement but show us where we can improve our processes.
What did Esprow offer, in terms of capabilities and functionality, that TradingScreen could not achieve by building, or by going with another vendor?
Chris Kingsbury, TradingScreen: The Esprow product offered functionality that was superior to other products we have experienced in the past. Adding custom FIX tags and repeating groups have always been time-consuming tasks that are much quicker now with the Esprow application. Another previously nearly impossible task was the creation and use of complex, multi-leg messages. Esprow has been able to show TS a way to do this with their product giving us testing abilities not previously available. There is also a template sharing ability that makes sharing complex test templates amongst the team easy and helps the team be more efficient.
Given how important ease of migration is to integrate new solutions into an existing IT framework, how was Esprow specifically able to deliver on this for TradingScreen?
Francesco Lo Conte, Esprow: API testing frameworks are just tools to create testing scripts that automate testing of trading systems. Good frameworks let you do this quickly and offer many features. ETP Studio for FIX, for example, can generate test scripts from log files and offers more than 100 test commands. However, a lot of the business value resides in the scripts themselves, since they imbue the knowledge of how to test the target system. It is crucial for new clients to minimize the cost and disruption associated with upgrading to a modern framework like ETP Studio. Some pains are difficult to alleviate – for example, the need to retrain users. For that, Esprow offers online videos, online training, and even on-site instructor-led classes. We also offer an automated test migrator that transforms legacy test scripts to ETP Studio’s own format. The process takes less than an hour and allows clients to hit the ground running. This has proved a blessing to some of our enterprise clients who needed to migrate tens of thousands legacy scripts.
For readers who might be technology-procurement decision makers, what are the main takeaways from the TradingScreen-Esprow success story?
Francesco Lo Conte, Esprow: In my opinion, the main takeaway is to remember that “People Buy from People”. It is difficult to describe how much our team at Esprow genuinely cares to see users of ETP Studio succeed. Even though we have a one-hour SLA response time, our team normally replies within 10 minutes — faster than getting yourself a coffee! They become literally an extension of clients’ teams – especially now that many of them are working from home.
As another example, when a procurement buyer in the past forgot to include some features in their purchasing order, we just added them afterwards at no additional cost to spare them having to go through the entire process again. This is what you want from a real partner – to be in the trenches with you when you need them. As part of our “Business Partner Program,” we schedule regular weekly meetings with their stakeholders, even if just for 15 minutes, to hear how they are doing, whether they have feature requests, or whether we can help with other projects. This is how we imagine a great relationship with a great vendor should be – about people helping other people succeed. It is part of our work ethic, it is in our company’s DNA, and I am convinced that it is also part of our success. Today, most of Esprow’s new business comes through referrals. As we continue to grow, it will remain our priority to be able to offer such personalized service.