How to Build a Trading Platform? Expert Advice.
Updated: Aug 11
Velmie, a Fintech software provider, has partnered with Exberry, the marketplace and exchange technology pioneer, to deliver advanced and comprehensive technology for cryptocurrency exchange and trading platforms. Exberry core technology, a cloud-based matching engine, is suited to support every type of marketplace from traditional assets to digital assets including security tokens or tokenized assets, alternative assets as well as gaming and real estate.
We talked to Magnus Almqvist and Ronen Nachmias about best practices when launching a trading solution, technology decisions, pitfalls and mistakes to avoid.
What knowledge/experience should the founder of a trading platform possess?
Magnus: There are three areas I would say are crucial for a trading platform to succeed.
Firstly, an intimate understanding of the trading and trading participants of the specific assets that are made available for trading on the platform. With this knowledge, you will understand how to create market participants want to act on, and you will be able to drive innovation and improve market dynamics, thereby increasing liquidity and volume even further.
Secondly, the regulatory landscape. A strong understanding of the regulatory requirements and expectations will help to have a constructive discussion with the relevant capital markets authorities to ensure they support your operations and market innovations that you are planning.
Thirdly, a good understanding of the requirements and expectations of the technology, platform operations, and market operations. This is crucial to building efficient and effective processes and procedures, with clean interfaces between the different functions. This is of course even more important if significant functions such as the technology and platform operations are outsourced to one or even several vendors.
What is the difference between trading and an exchange platform in terms of technology?
Magnus: A trading platform connects to an exchange platform in order to place orders and receive trade details when there is a match on the exchange platform. The exchange platform maintains a central limit order book and matching engine, where buy and sell orders are managed, and trades executed when prices cross.
The crucial difference is that the trading platform needs the functionality to analyze and set prices on orders before they are submitted to the exchange or marketplace, and a smart order router to manage order flows. The exchange platform, on the other hand, facilitates price discovery and continuously communicates market evaluation in the form of the best bid offer, and a last traded price. It does not itself set the price but helps market buyers and sellers meet at a price they both accept as fair.
What are the typical pitfalls that arise on the way to launching a trading solution? Why are those important? How to avoid them?
Magnus: Liquidity is your largest challenge and the most common reason a new marketplace fails. How do you attract buyers and sellers to your new market? Derived from this, you see that onboarding of clients and integration of trading systems needs to be as simple and straightforward as possible to attract users. If trading is done via an app or website, these need to be easy to access and to use.
“Liquidity is your largest challenge and the most common reason a new marketplace fails.”
Fee schedules need to be easy to understand and attractive so that traders understand the cost of participating in the market.
Of course, the technology needs to be stable and reliable, so that technical onboarding and use of the platform feels robust. This is crucial to building trust in the markets.
Depending on the nature of the market, monitoring for potential market abuse and enforcing market rules are very important to build trust in the market. If participants cannot trust the prices they see, or if they are taken advantage of when placing an order, they will reduce their interactions with the market, and liquidity and volumes will suffer.
What are the most often mistakes owners of trading platforms make while planning and launching the product?
Magnus: Underestimating the regulatory steps and processes required to receive the necessary licenses and credentials for launch.
Not appreciating how important liquidity providers or market makers are. It’s important that a very strong partnership is built before launch with key liquidity providers who will support the initiative and want it to succeed in the long run. Owners often underestimate the time and effort required to build sustainable liquidity where volumes start being self-supporting and trading is attracted to the market simply by the activity on the market.
Buy vs Build dilemma: what to choose?
Magnus: Building gives you more control, but increases risk, cost, and time to market. Buy means compromise in that you will need to rely on available functionality and architecture. On balance, buy is a very sensible choice to reduce risk, keep costs under control, and dramatically reduce time to market. Why reinvent the wheel when you can buy reliable, proven, light to deploy SaaS platforms that are easily integrated into wider ecosystems and reduce operational burden on your own organization?
What are the benefits of a cloud-managed trading service vs self-hosted trading software?
Time-to-market - with exberry managed exchange as a service it is possible to get to market much faster reducing the barrier to entry by integrating via a well-documented simply to use API of choice such as websockets/json, TCP/FIX, binary/UDP. The price discovery engine built as a cloud-native solution can be launched in a matter of minutes at any region and availability zone.
Market-optimized - The service is built in a way that it can be optimized for performance or cost and can scale by on the fly changing hardware profiles to meet any required SLA and can horizontally scale to optimize to specific demands such as having unlimited amounts of instruments or managing billions of orders per second.
Total-cost-of-ownership - Get the hardware you need when you actually need it. With exberry managed solution exchange operations do not need to worry about maintaining the software or owning the hardware and as demand grows operators can on-demand meet the expected loads. The service is elastic and can be optimized and sensitive to the fluctuations of the current business needs in terms of the total cost of ownership and required SLA.
Multi-tenant service and exchange operators can launch exchanges in dedicated environments and shared environments managing more than one exchange type or exchange environment. So exchange operators can now iterate and innovate much faster than ever before. Launching new types of markets faster.
Cloud infrastructure gives the advantage of having on-demand hardware resources anywhere in the world. Exberry has integration with the cloud providers and knows to orchestrate it in order to run any operation at any size at any given time in a matter of minutes.
Sandbox developers can have an available environment to POC with a matching engine as a service without the need to install or maintain the service. Instead, they can focus directly on the business and workflows of their business. As such exberry is a solution like no other as developers just need to establish a session with the exchange and get up and running in a matter of a week.
Best-of-breed technology and microservices architecture combining the requirements of traditional platforms performance with up-to-date practices. It is agnostic to the infrastructure and the cloud. As such it can run on bare metal and on-premise as well and still be operated as a managed service in an on-premise environment exberry treats the on-premise as a private cloud and deploy, run and operate the environment with 24/7 support SLA.
Vendor-lock-in - There is no higher vendor lock-in than installing and operating and/or building and matching engine yourself when using exberry API you have the matching engine decoupled by an Application Programmable Interface and its set of functionality. as such exberry reduces the vendor locking.
High availability and disaster recovery - Except Exberry matching engine being highly available deterministic and replicated state machine - the availability of the cloud is unparalleled to owning on-premise hardware this introduces new opportunities to deploy and run exchange operations in several modes that fit the market needs:
Warm standby active-active is when one replica-set is deployed over 3 availability zones and data is synchronously synced in real-time. the advantage of zero downtime and cost-effectiveness with a tradeoff latency due to sync replication.
Warm standby active-passive is when the replica set is deployed over 2 datacenters deployment and keeping 2 replica sets. One set per site. The replica sets are asynchronously replicated. With the advantage of an ultra lower latency and the need to mitigate minimal data loss when switching to the DR recovery site.
Does your solution also come with market-making algorithms?
Magnus: Exberry is easily integrated with market-making algorithms. If the platform is a broker-driven market, the broker order management system and quoting engine can be closely integrated with the exchange platform to ensure reliable and very low latency.
Magnus Almqvist is Head of Exchange Development at Exberry. In his role, Mr. Almqvist builds and nourishes contacts with exchange pioneers around the globe with over 20 years of experience behind him. With a background as Head of Sales, Aquis Exchange, Senior Business Development Manager at FIS (SunGard) and prior to that heading up Smarts Group’s (part of NASDAQ OMX) London office, and as the CTO of the London Stock Exchange Group Company EDX, he has a solid understanding of capital markets exchanges, their ecosystem, and their evolving universe.
Ronen Nachmias is Chief Technology Officer at Exberry. He is a visionary with over 20 years on the bleeding edge of the software industry. Ronen is a Technological expert CTO, R&D Director Founder and Project Lead of scalecube.io an Open Source for reactive-low latency microservices. He holds vast experience in low-latency internet-scale mission-critical, reactive, distributed systems and has a track record implementing online cloud-based Microservices architectures at scale. In 2005 he visioned the first serverless platform and co-founded Prima-Grid LTD later he acted as Chief-Architect at Playtech a low-latency mission-critical gaming platform running over 500K money transactions per second at consistent load over a global bare-metal operation of 300 sites.
Contact us to get the details on how Velmie and Exberry can help you build a trading solution.