B2B Fintech Trends for 2021: Storage, Security, Cloud Migration, IaC

As CTO at , a large part of my work revolves around observing the forward-looking industry trends (including fintech trends) that affect how we create innovative products.  The fintech industry is experiencing a paradigm shift across multiple vertices of the technology spectrum, which has been accelerated by the Coronavirus pandemic. This has sparked a series of new fintech trends, which I have been observing throughout 2020. Most of which are based around financial corporations moving their operations from on-premises to the cloud. More and more financial corporations are shifting from on-premises legacy solutions to full cloud and hybrid infrastructures – for both their innovative and core/mission-critical products and services.Moving from on-premises to the cloud allows tasks that previously took a month and a half to complete to be completed in a couple of hours. This is not the only driver behind fintech moving operations to the cloud, however.This fintech trend also allows for 

• Ease of scaling
• Resilience
• Increased Security
• Compliance
• Decreased costs 
• Maintenance efficiency 

Let’s take a closer look at the main drivers behind the shift from on-premises to the cloud.  

Storage

Due to regulations and compliance, established finance companies usually store large datasets in physical data warehouses. If you decide to follow the fintech trend of moving everything to the cloud, it won’t be a quick task – it’ll take a couple of months. There are a lot of constraints such as network latency, security, audits and so on. Fortunately, cloud vendors are now offering hybrid solutions to assist with this matter. They’re able to set up a gateway which bridges to your company’s premises and starts moving data to the cloud in the background.This process will take a couple of days. During this time your applications will be consuming data as they were locally; bypassing the constraints of network speeds, latency and security.Throughout the backup and synchronisation process, it’s not possible to know if the data is being consumed from the cloud or locally. Once the synchronisation is complete you may experience a couple of hours of downtime. However, it is possible to complete synchronisation without experiencing downtime.

After you migrate to the cloud, you’ll be able to sunset your on-premises data centre. This will allow you to cut costs and prepare for scale-up or relocation with almost no maintenance or downtimes.

There is another benefit of storing your data on the cloud. It’s a lot easier to reach different markets across the globe. This is achieved by leveraging a replication layer that cloud vendors offer – giving customers a consistently superior experience regardless of their location.Conversely, moving your data from one physical warehouse to another requires a long audit process with a lot of bureaucracy.By placing data on the cloud, vendors can ship you all the required documentation easily and quickly. Moreover, they will work with regulators to help facilitate the move. To assist with your transition, the cloud vendor will usually provide a dedicated team to liaise with the company’s internal IT team.

Network

The reason one of the main fintech trends is cloud migration is that (regardless of whether a company is B2B or B2C) two of the most important metrics, that all stakeholders want, are speed and resilience. To meet these metrics cloud vendors provide edge locations in most countries to allow for even faster access – we have one in Lisbon, for example. Meaning, if you’re a US-based company creating a partnership in Europe, your partners will experience the same quality of service as they would in the US. 

Payments Applications

By default, fintech corporations operating payment applications run on legacy application environments (on-premises physical computers running virtual machines).The problem with this, and the reason the fintech trends have evolved, is it’s not easy to scale or move the machines between availability zones or regions.Banks, who have been established for over two decades, are now moving their mission-critical payment services to containerised architectures on the cloud. This means that scaling up, scaling down and deployments are a breeze.I think this is one of the main drivers for companies which operate payments or obligations. They’re starting to move towards the concept of containerization because it allows them to develop easily without any downtimes and without having to spend money setting up additional physical capacity resources. This is a change in the, usually observed, conservative pattern of innovation in fintech companies where only less critical services are evolved. There are some innovation adopting banks which haven’t moved entirely, however they are running containerized architecture using, for example, OpenShift from Redhat, which allows them to run hybrid solutions between on-premises and the cloud (hybrid cloud). In this case, the cloud vendor sets up an on-premises cluster together with a container registry. There are a lot of disadvantages of having legacy virtual machines on-premises. Namely, adding capacity takes a long time, up to a few weeks or months. This is due to the 

• Application of more hardware
• Installation of the hardware
• Establishment of middleware to configure the systems and work on the security layers.

Usually with fintech companies, if there are downtimes or severe degradation of service, fines and sanctions are applied by the regulators.

By containerising your infrastructure, adding more capacity can happen in minutes or hours, not weeks. Meaning your applications can scale up quickly when peaks of traffic happen. Additionally, the downtime is close to zero and the degradation of service is limited or non-existent.If your IT team is not familiar with containerisation and the new pattern of deploying apps, cloud vendors can provide on-going support. They will work with your operations engineers to make the process as easy as possible. In addition to the above, there is a trend of companies offering DevOps as a service. These are highly qualified cloud engineers that can help you with the process.To explain further, normally, a bank has a workforce of operations engineers maintaining their infrastructure. By moving to the cloud they need far fewer. This results in an obvious reduction in IT management and, now redundant, legacy management personnel. This allows the banks to move engineers to other, more productive parts of IT applications and services. In summary, it’s possible to cut down on IT management whilst gaining cloud vendor teams to assist with the transition. 

Security

Moving from on-premises to the cloud brings more opportunities in terms of security. This is another reason why cloud migration is becoming one of the most adopted fintech trends. 

There are a lot of vendors and external services available to help you take advantage of the security advantages cloud migration offers.

One of these opportunities is automated threat detection – which runs machine learning models against your centralised logs. In seconds it can run and detect millions of threats leveraging the tools enabled by cloud migration.These threats contain data sets: Malicious IPs, new modern attack techniques; all of which are updated by the cloud vendors on a daily basis. Not only does this give your financial company peace of mind, but it also reduces your cost. If you’re already using managed services from a cloud provider’s ecosystem they will offer threat detection at an extremely affordable price.

Legal & Fines for Service-Level Agreements (SLA) 

As you’re aware, your fintech company requires protocols of resilience and . As I mentioned in a previous section, when it comes to cloud migration, vendors are able to prepare the required documentation ready for you to hand-over to regulators. This makes the legal process quicker than implementing a new on-premises machine – which can take a couple of weeks. This is critical to maintaining your SLA and avoiding sanctions if your SLA goes off threshold.With cloud migration, the capacity increase/decrease burden disappears when and also makes higher levels of SLA uptime achievable.

Infrastructure as a Code (IaaC)

As a fintech, your IT Department has a lot of certifications and protocols to process and comply with. To make things easier for their IT operations, corporations are following a pattern of having Infrastructure as a code (IaaC). Meaning all the protocols, tasks and tooling used to deploy the application environment are now written in code. Using this, your IT operators can simply run the code with the cloud vendor and the infrastructure will be provisioned in a few hours.This is a big advantage when compared to the other solution – following a 50 page PDF of procedures and protocols step-by-step.This also simplifies the process for audits and compliance. Most major cloud vendors provide IaaC. Each provider gives it a different name, but at its core, it’s a YAML file. You can move this YAML file between cloud providers to avoid vendor lock-in (with some minor adjustments). This flexibility has become a key reason company’s are adopting cloud migration. 

Disaster Recovery

Another problem that I’ve been noticing is having services redundancy across multiple server regions. The fintech trends are evolving to solve this problem too. With more people on the internet, everyday traffic is increasing. This has been exacerbated by the recent pandemic. This has resulted in some regions’ internet connectivity being overwhelmed, increasing the chances of shortages.The solution is to run your services in multiple regions and availability zones to prevent intermittent downtime – as traffic can be routed to healthy and responsive servers.Under extreme circumstances, in a disaster scenario where regions are compromised, it’s feasible to move everything from one country to another within a couple of hours.In a worst-case scenario, your financial services will go down for a few hours. Conversely, the compromised regions may be unavailable for a much more significant amount of time. This is why, together with cloud migration, companies are adopting rapid disaster recovery patterns. They can incur fewer fines, or none, due to a higher SLA uptime – whilst making their customers and partners happy.It is mandatory to have a disaster recovery plan. Cloud vendors will usually manage this and provide you with all of the necessary paperwork completed. Then you just need to be forward the paperwork to the appropriate regulatory authority.  

Managed Blockchain

Quick note: This section covers blockchain-based distributed databases to store different types of data, such as financial operations logs and signature management – one of the bleeding edge fintech trends. This does not cover cryptocurrency. There has been some friction from established companies in adopting blockchain networks and they are not yet the standard.Those who are adopting it are using Hyperledger which is an open-source framework for setting up and running a blockchain network.These adopters end up facing issues with scalability, maintenance and service degradation. The main reason behind this is the overall lack of expertise in the IT departments on this matter and the complexity of the operation.To create an example, let’s take an internally managed blockchain network. There, we start placing non-mission-critical information.  Problems start to occur when part of the network is down, or a network node becomes available. The IT operations team don’t know how to fix or upgrade because this version is already deprecated. This is just one hypothetical, a plenitude of problems may happen on these “playgrounds”. Which, despite looking promising at first sight, end up becoming a burden in terms of operations –even though they are only running non-critical applications.

Cloud vendors noticed this as one of the niche market fintech trends, and now offer managed blockchains.

Meaning, your fintech’s IT team can set up a blockchain network that is managed by the cloud provider– not the internal team. Now, if you require more data the cloud vendor can add more nodes. It is even possible to implement scaling policies with the cloud vendor – in which they will implement more nodes automatically if reading and writing operations are becoming slower, for example. Currently, the majority of cloud vendors are already compliant with regulators. Meaning, in a highly regulated industry, such as the financial one, a cloud vendor can provide certification after they deploy a blockchain database. You can then forward that certification onto the regulatory authority. We’ve started seeing fintech companies using blockchain, for storing different types of data. Not crypto-currencies, because most of the financial companies still operate on Fiat currencies. We’ve also started seeing more and more managed blockchain networks because they’ve become easy to set up and operate. Therefore, companies don’t lose innovative momentum. Most financial companies want to be on the edge – so it makes sense the earlier they bring these types of technology in-house for non-critical applications; the smoother the transition.By operating a managed blockchain it’s easy and secure to onboard partners to the loop of innovation. By simply whitelisting IP addresses, your partners can connect to the blockchain network. Meaning they can all start writing and consuming on this distributed database – without any network constraints or security issues. 

Automated Identity & Document Verification

In financial industries, this has always been a problematic process. All the documentation, for example:

• Printed statements
• Identity cards
• Passports 

The manual, or semi-manual, certification of these paper artefacts is becoming a thing of the past thanks to AI by SaaS.The fintech trends are changing, and now companies are starting to use automated identity verification, and not going through the manual process – because it’s less time consuming and it’s cheaper to operate. Passport service is standard; a standard form of bank statement is also feasible. However, there are special artefacts – such as a land acquisition paper from a specific country which must follow a specific format. For those use cases, cloud vendors offer I.T. departments, high-level tools which allows the re-training of a machine learning model to understand this new type of document. Within one or two days, it’s possible to start using this new type of artefact, like the registration of offline property owners – for example.I believe it won’t be long until this fintech trend will become de facto and we will see the end of manual verification.

Remote Working

By default, banks and financial companies need their employees to use a computer in the office because of security and compliance. Usually, employees can’t take their work computer home or install software on personal devices.

Due to the pandemic, we saw a shift in fintech trends where most companies switched from on-premise to remote overnight.

Cloud vendors rapidly created a solution for this. It enabled banks to switch the machines their employees use to virtual machines that run on a private network within the cloud. This enables employees to use a personal computer without installing company software. Instead, they install a piece of software to link to the private network. Then, they simply use their keyboard and mouse to operate the virtual, cloud-based computer remotely. Theoretically, this means a financial startup can now launch without a building or premises without security or compliance issues. This could change the way financial employees work in both fintech and banking as employees can work on tools which belong to the bank remotely.

Serverless

Normally, an application is deployed and run on the server. If it’s containerised it’s deployed on the cluster, it scales it up and down and the application is replicated – then if high traffic occurs the servers can call more machines to handle it. But that has a cost. If we keep adding machines to our cluster, our costs will just skyrocket. As a result, we’ve seen that financial companies adopt one of the latest fintech trends. Migrating some of their application environments to serverless. This allows them to take the pressure off of their existing services. If someone were to upload a PDF it triggers a serverless function; so it’s theoretically not running on any server. A serverless function is triggered once a passport is uploaded and the file becomes available in the bucket. This analyses the passport and tells another service whether or not it is valid. Another common use for serverless functions is to serve a landing page. With a serverless landing page is there can be more at the same time and it will be fine. A final example of the benefits of serverless pertains to alerts. Threat detection can trigger a function that will run a mitigation procedure, notifying the responsible person on the phone. Serverless allows the creation of complex scenarios which can be overwhelming to carry out with APIs. With serverless, it’s easier and faster to achieve results. 

Wrapping Up

These are just a few of the fintech trends I’ve observed this year. As we moved closer towards 2021, what fintech trends do you think we will see emerging in the financial landscape?

This article was also published by Claudio Teixeira.