Third article of a blog series: Build vs Buy Software
In the first blog post in this series, “Three Critical Strategy Questions”, we explored some fundamental considerations for companies contemplating the development of an enterprise-scale MES (Manufacturing Execution System). The key issues included:
• Am I a Software Company or a Products Company (or What Do I Do for a Living)?
• Am I Prepared to Invest in Software Company Infrastructure?
• What is the Relative ROI of Build or Buy?
The second blog post in the series, “Seven Critical Questions—What Should We Build” covers a framework for defining the final product. It addresses key categories such as functionality, standards and compliance, integrations with other systems, architecture and deployment, scalability, management of underlying components, and maintenance.
If you and your company want to embark on this journey—aligning it with your manufacturing data strategy and overall, IT strategy–we offer some guideposts for analyzing system and project cost. Here are three suggestions for the cost categories you should consider, beginning with near-term expenses and the associated cash requirements.
Category #1: Near-Term Expenses and Cash Needs
You will need to replicate all the functions of a software development company and assess the following:
a) Whether you currently have these functions on staff
b) If you do, whether these staff members can be allocated to this project
c) If not, whether you can get them
d) The associated costs.
These functions include, among others:
• Business/functional analysts
• Architects—product and deployment
• Project/product management
• Coders/developers
• Test/QA/validation
• Deployment
• Documentation
• Help desk support
• Change Management
Once you have staffed-up appropriately, you will need to decide for how long you’ll need each function. If you decide to use implementation partners, you must ensure they are thoroughly familiar with this new application to deploy it effectively. Additionally, you’ll need a DevOps infrastructure suitable for this level of system complexity, along with licenses for the components and environment your team plans to use in the final system build.
Category #2: Opportunity Costs
This may sound like an MBA buzzword, but it’s a crucial consideration. The cash, time and resources applied to building an MES will likely turn out to be a significant investment. That investment could alternatively be used for other initiatives that might yield a return, making it essential for the project to be worthwhile for both the company and its investors.
To assess this, you will have to know how long the project will take to deliver value and at what level. When analyzing competing investment options, your Finance Department will need to know the overall project timeline: the duration over which money will be spent, and how quickly and substantially value will be returned—and this includes some longer-term costs, that we’ll cover in a minute.
Another important aspect to consider under this category is the notion of “not knowing what you don’t know”. This applies not only to the project cost and timeline, but also to potential gaps in your build, due to a limited perspective based solely on your own processes – a phenomenon known as “process myopia”. Let me explain that a little further.
While no one knows your manufacturing processes and environments better than you do, your competitors may have different, possibly more efficient approaches. You should expect this to be true. Do you have 100% market share? Is competitive pricing significantly better/worse than yours, and at what margins do your competitors operate? MES suppliers who work with multiple manufacturers across various industry segments, incorporate diverse insights and best practices into their products. It could be that you look at their pricing and say that you can build it for less than buying it, but are you getting the accumulated industry knowledge that’s baked into a product used by multiple customers?
Category #3: Longer-term Costs
When planning for an MES, it’s essential to account for future costs that go beyond the initial build and deployment. Considering these longer-term expenses in your initial project planning, will help prevent unpleasant surprises that come from improper planning. Here are some key cost categories to consider:
- Maintenance: while software doesn’t need lubrication or replacement gaskets like hardware, it will require ongoing maintenance. You’ll discover new needs or ways to improve processes, add new equipment or products to the production floor, or integrate new software to your company’s portfolio. One survey article suggests that “software maintenance is the most expensive part of the life cycle” and can account for ‘“more than 90% of the total cost of software’”.1
- Risk: It may be difficult to assign a value to this category, but there are very serious considerations around the long-term stability of what will become a critical application. Risk categories include:
- Longevity/stability/migration ease of third-party components: While these components may save time initially, they can turn problematic when they reach end-of-life or lose-support. You’ll need a plan to manage this inevitable situation.
- Personnel turnover: If the key person or team responsible for building, knowing and deploying the system leaves the company, can you keep the system up and running without interruption? Is your system documentation up to the task of supporting a new maintenance team
- Operational risk: Since this application will run your plant(s), any downtime could halt production or shipping. Will your customer-facing teams be able to communicate order statuses or lead times? Taking this risk into account will drive investment in support capabilities.
- Scale and flexibility: Will the system be able to keep up with changes to the manufacturing value chain? If more capacity is added, will the system be able to support it? Can it handle multiple or new plants, or different types of plants? As you generate more data, is the underlying infrastructure capable of handling the increased load without compromising on latency and reliability?
This post series may seem to advocate for a particular point of view, but please take it as experience-based advice. We know that software systems are challenging to build; it’s all we do. Today, many products companies are also very accomplished software developers, because software is an increasingly important part of finished products, leading them to create highly capable software teams. If you’ve thoroughly addressed the strategy, product and cost questions and decided to proceed as an extension of your manufacturing data strategy, you’ve made a business decision that you feel is best for you, your company and your investors.
However, if you ever decide to reconsider, we will be here to support you.
- Galanopoulos, Ilias, SAPM: Course Blog, “Let’s talk about Software Maintenance”, University of Edinburgh. ↩︎
2. Delve deeper into the topic on our LinkedIn: Build vs Buy Software: What should manufacturers consider?
Read the complete series on Build vs Buy Software:
You must be logged in to post a comment.