Which cloud computing model is best for business messaging

I’m old enough to remember a room in the first company I worked for that was full of racks of blinking machines, like the control panel of the Starship Enterprise. They were servers we were told, and we weren’t supposed to go in there. Although the door wasn’t locked and we all had a peek at some point, especially on very hot days when the industrial cooling system offered welcome relief.

These days, having all your customer and transactional data and the machines running the applications that drive your business all in one place just seems bizarre. What would happen if the air conditioner failed during a summer heatwave?

Off-site data centers were the next logical step, but now we just accept that our data is ‘in the cloud’ and assume that it is secure and backed up and that our applications will always be available. But how many of us actually understand what that entails, even though almost every enterprise business uses some form of cloud computing (the figure was 98% for the finance industry in 2023)

In this blog we start by explaining the basics of cloud computing and then dive into the specifics for business messaging aimed at helping brands to select a cloud deployment model that offers the perfect balance of security, customer experience, and cost.

Source: Oracle’s Security in the Age of AI Report

Some people use the word ‘cloud’ to describe any computing model where data and applications are hosted in another location and are accessed by users via the internet. That’s not the cloud, that is a data center. Even if the data center is in another country and owned by someone else, we still know where it is and could theoretically travel there and plug in a flash drive to retrieve our data. Bad actors could potentially do the same, or a meteor strike could wipe out the digital footprint of the entire company.

In its truest sense the cloud is an interconnected network of data centers, and the system of automated frameworks used to support an elastic and serverless deployment model. Of course, there are still servers involved, but we don’t know or care where they are as long as our applications keep running and our data is secure.

Of course, sometimes we do have to care where they are. For example, GDPR dictates that data has to be stored and processed within the European Union, and many government and military contracts specify where cloud servers are allowed to be physically located, but you get my point.

Once we have agreed on a definition of cloud computing that doesn’t trigger the technical folk, we can go a step further and classify the main types of cloud deployment model. We can then decide which is best suited to the technical and compliance complexities of global business messaging.

When most of us think of the cloud it is the public variety. These are computing environments provided by third-party vendors and accessed over the internet. Household names like Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform, all offer resources that can be accessed for a fee by (almost) any organization from (just about) anywhere. (If you are a hacking organization based in North Korea you may have to look elsewhere for cloud storage.)

There are several benefits of this model, the main one being convenience. Need a new virtual machine to use for AB testing? Just spin up a new instance. Need to warm up some sending IPs prior to the shopping season, add some virtual machines.

Your organization doesn’t have to worry about purchasing and maintaining the infrastructure, someone else is responsible for the security and compliance side, and with resources shared between tens of thousands of users, the costs are generally very reasonable.

In the context of business messaging, what you lose out on is control and flexibility.

If control and flexibility is a high priority, then you may wish to consider a private cloud model. As the name implies, this is a cloud computing environment that is used exclusively by one organization. The network of data centers may be owned and managed by the organization, or by a third-party service provider.

This means that you get exclusive use of all resources and more importantly, complete control over the configuration, performance, and security measures in place. Private clouds enable organizations to create bespoke computing environments for every use case, considering storage and performance requirements, and any integrations with legacy or third-party systems.

For global companies dealing with regional and industry specific rules and regulations, the private cloud model also allows them to easily tweak environments for compliance purposes.

As you might expect, the main drawbacks of the private model include the overheads and costs associated with managing a network of data centers. In the context of business messaging where there are predictable peaks and troughs in messaging volumes during the year, you may be paying for infrastructure and resources that are underused for part of the year. For example, there may be a sharp increase in marketing messages in the run up to Black Friday and the shopping season that your data centers need to cope with, and then a slowdown in all messaging during the summer holiday period.

Once again, the clue is in the name. With this model organizations can combine the best elements of private and public clouds to give themselves the flexibility that they need but also optimize their resource usage and reduce costs.

For example:

Another advantage of the hybrid model that is often overlooked is that it supports rapid innovation by allowing businesses to more easily adopt and benefit from new technology. A great example of this is in the field of business messaging where technology like Google’s Vertex AI can be used to automate messaging on channels like RCS.

Overall, the hybrid cloud model leans on the strengths of both public and private clouds but there are still the overheads of managing private data centers and the human resources required to manage two sets of integrations.

This is where businesses may want to evaluate the CPaaS model for delivering their business messaging use cases.

You really have to admire the no-nonsense naming strategy in the cloud industry. A multi-cloud strategy refers to using multiple cloud computing services from different providers to benefit from greater geographic and use case coverage.

Under this model different clouds can be used for different tasks, for example, one provider might offer cheaper or more secure data storage, while another may provide more computing power that results in faster customer transactions and message delivery times.

Using multiple vendors also spreads risk and reduces the dependency on a single cloud provider.

Source: Cloud Worldwide Service, Forbes.

CPaaS (Communication platform as a service) and cloud deployment models are not two distinct things as the CPaaS model is based on cloud infrastructure.

Both aim to solve many of the same business problems and can be used effectively in tandem to:

While with the cloud model organizations run their own messaging applications and store their data in the cloud, with CPaaS the business will work with a provider to add messaging and communication services into their own applications. These might include any combination of SMS, email, Voice, chat apps like WhatsApp, Messenger, and many more.

This helps organizations to easily scale the volume of their communications to deal with seasonal variations in messaging traffic – without having to make expensive hardware purchases to cover peak periods.

The CPaaS provider publishes an Application Programming Interface (API) that software developers working for an organization can use to access the messaging tools and associated services.

The actual applications could be running in a Private or Public cloud, in a data center, or even on site in that mysterious server room that we mentioned earlier. It doesn’t matter, as long as there is an internet connection.

For example, a financial institution or bank would partner with a CPaaS provider to be able to send customers alerts when they were about to reach their spending limit or when there was a potentially fraudulent transaction on their account. The channel used to send the alert is flexible and could depend on the use case, the region, or even the preferences of the individual customer.

The financial organization doesn’t have to worry about warming up sending IPs for email or about configuring a backup channel when an SMS isn’t delivered. All that is taken care of by the CPaaS provider The secure API link between their operational systems and the CPaaS provider’s communication services ensures that important messages will be delivered to their customers.

Whether an organization has fully embraced the cloud computing model or is looking for help with geographic expansion or scaling their business communication, the right CPaaS provider will be able to help in many areas.

Channel coverage: A leading CPaaS provider will be able to support a business with the adoption of all the channels that they need to use to reach customers across all regions. Some countries use specific messaging channels that have low adoption rates everywhere else. For example, if you wish to do business in Vietnam, then you should consider using Zalo, a messaging app with an 87% usage rate, which is far ahead of any other app in the market. 

Geographic reach: The right vendor will enable the business to reach recipients in all the markets they operate in both reliably, lawfully, and at a reasonable price.

Compliance: The CPaaS vendor will ensure that all communication complies with legislation in all the countries that recipients reside in. This helps brands to avoid fines and block listing, which could have a major impact on their business.

Volume: The right CPaaS vendor will be able to support all the communication that you send throughout the year, including during peak times like the run up to the holiday season when other businesses are also sending their highest volume of messages.     

Failover mechanism: For business-critical messages, CPaaS supports failover messages when the recipient can’t be reached on the primary channel. This process will be automated and tightly monitored so that you always know when a recipient hasn’t received an important message.

Management overhead: The provider will take care of all onboarding, number provisioning and ongoing management of their clients’ communication. This is a significant factor when you don’t have the technical resources to do this yourself.

Fraud protection: The vendor should provide the highest levels of fraud protection, with automatic detection of fraudulent traffic, effective SMS firewalls, and continuous analysis of messaging traffic to protect your business and customers from messaging scams and fraud.