I wonder if P.T. Barnum would have agreed that “There is no such thing as bad publicity” if he worked for Southwest Airlines and woke up one morning to the Forbes headline “Southwest Airlines Under Federal Investigation For Possible ‘Deceptive’ Scheduling.” While we live in a world where all publicity may be good for celebrities, the same cannot be said for businesses. There is just no way to spin an unexpected net loss of $825 million for a quarter that was projected to “generate strong profits and margins in fourth quarter” just two months earlier.
Many companies find out the hard way that their carefully developed brand image can be destroyed in days, if not hours when their organization fails their customers en masse. There is no quicker way for a company to accomplish this misfortune than having their systems implode in front of the world. While it is easy to pick on Southwest Airlines today, there are many examples of such breakdowns, often related to security breaches or product safety failures.
What Happened at Southwest?
In the blog, “Failing to Keep Up With Technology Will Undermine Your Digital Transformation Strategy,” Paweł Chądzyński discussed several recent examples of companies painfully mitigating the impact of their systems falling behind. The case of Southwest Airlines was particularly interesting since the meltdown was “100% due to a purposeful delaying of technology upgrades.” In other words, totally preventable. The article "The Shameful Open Secret Behind Southwest's Failure," details the appalling situation where over 17,000 flights were canceled over a busy holiday week last year. According to the article, the scheduling systems desperately needed to be modernized due to the significant amount of technical debt in the code.
What is technical debt?
Typically, technical debt is the result of coding shortcuts that are created within software during system modifications. The technical debt remains in the code base until it can be reviewed and improved later. This does not necessarily mean technical debt are bugs in the code, but rather strategic limitations that make the system less resilient for future changes (although bugs can easily be introduced as a result of technical debt). While technical debt may not cause an issue today, there is a high likelihood of system issues in the future if they are not addressed.
Creating technical debt typically occurs when tight deadlines limit the time to implement system changes. Unfortunately, new priorities often interfere with the best intentions. Delaying the correction of the technical debt creates a ticking time bomb for the organization. In many cases, normal software upgrades become too difficult to implement, leaving antiquated versions of software in place for years. The fact that there had been smaller scale meltdowns at Southwest was indicative of a larger issue brewing. According to the article, the situation had escalated to the point the president of the Southwest flight attendants’ union said there were system breakdowns even during mild hiccups.
Enter Software-as-a-Service (SaaS)
There is a major transformation happening in the software industry. Almost every system, crossing all types of applications, is being built or retrofitted to leverage cloud infrastructure. With widespread cloud adoption, software vendors are also altering their approach to commercializing their products by offering them as software-as-a-service, or SaaS. By providing the same software solution to multiple customers and using shared technical and human resources to manage it, the economy of scale brings a lucrative business case to the solution provider.
A cornerstone of SaaS offerings is that the functionality works the same across all customer deployments. Without allowing the capability for each customer to customize the solution for their specific business requirements, support of the product can be streamlined across their customer base. Most SaaS offerings provide some minor configurability to users, but significant functionality remains constant.
This limitation does not create an issue for many types of SaaS offerings. For instance, most people would not consider customizing the code for Microsoft 365 or for their Netflix subscription (yes, Netflix can be considered a SaaS company providing on-demand videos). For more comprehensive enterprise solutions like Product Lifecycle Management (PLM), which provides digital thread capabilities across multiple systems to tens of thousands of users globally, working out-of-the-box (OOTB) with a standard solution is unrealistic. Companies have very different requirements to support their unique business processes and combination of systems. It takes more than simple configurations to deliver to this level of complexity.
How can a modern, customized Software-as-a-Service (SaaS) solution help?
For enterprise solutions like PLM that require customizations to provide business value to their customers, vendors must support several capabilities not typically available in current PLM SaaS offerings. The solution must be able to implement customizations strategically, upgrade without impact from the customizations, and execute this within a DevOps culture for efficient delivery of system changes.
- Strategic customizations – The word strategic is vital. Many SaaS solutions claim to offer the ability to customize their applications, but the devil is in the details – what they really mean is configuration. While configurations are beneficial, they cannot replace executing customizations to solve a company’s most complex issues. A SaaS offering needs low-code capabilities to model a company’s unique business processes directly into the platform. Otherwise, IT departments will be forced to create workarounds and extra integrations to other tools leading to – you guessed it – technical debt.
- Upgrades without issues – Upgradability is a massive benefit of SaaS offerings, ensuring the systems will remain current. Upgrading SaaS solutions is just like upgrading your smartphone; you go to bed at night and wake up in the morning to an upgraded phone with new capabilities. Historically, PLM systems have been notoriously difficult to upgrade due to high levels of customizations and technical debt. If a SaaS offering cannot support customizations without affecting the ability to execute regular upgrades, you just lost a significant benefit of a SaaS. Of course, this leads to the earlier statement on Southwest’s meltdown being “100% due to a purposeful delaying of technology upgrades.”
- A DevOps culture – The nature of SaaS and working in the cloud is to foster collaboration across many diverse and globally located user groups. To support the creation of customized functionality in the cloud, the fundamentals of building, testing, and implementing system changes should not work differently than multiple business users collaborating on a business application in the cloud. A DevOps culture streamlines and automates the continuous integration (CI) and continuous delivery (CD) processes across different functions and locations to ensure the delivery of system changes without impact to the user community.
When combining the ability to strategically customize applications, upgrade the system without impacts from customizations, and streamline the continuous integration and continuous delivery process with a DevOps culture, a modern SaaS offering on the cloud delivers a solution with the power of an on-premise solution. These components will eliminate the creation of technical debt in the code base and ensure system upgrades will be executed on schedule and the system will be modernized with every upgrade. These capabilities are some of the features necessary to be considered a “Generation 2 SaaS PLM” solution as CIMdata describes in their recent commentary.
Conclusion –- What’s your brand worth?
Maintaining and enhancing global, enterprise systems is expensive, but it is possible to mitigate the costs. If the platform is designed to evolve as the company changes, it will not only reduce the operational costs of building and implementing modifications, it will also eliminate the creation of technical debt as described in “Southwest’s Meltdown Will Cost Up To $825 Million.”
What happened at Southwest should be a warning to companies that ignore the impact of creating technical debt. These shortcuts tend to snowball until a system cannot be upgraded to keep up with new functionality from the vendor. As companies are making major system changes to adopt cloud technology and new SaaS offerings, now is the time to consider how they will support their unique requirements and stop assuming an “out-of-the-box” approach will work for them. Otherwise, their employees may wake up one morning to “Southwest Stock Plunges: Flight Cancellations Top 10,000 As Feds 'Examine' Airline’s Historic Winter Storm Woes.”