How to Tell Ahead of Time If Your IT Transformation Project is Going to Fail (1 of 7)
We’ve started using a phrase in my office that encompasses all of the various migrations, reorganizations and modernization trends going on in IT departments today: “21st century IT transformation.”
Yes, I know. It’s a mouthful. Someone smarter than me will probably coin something more concise. And catchier. The phrase refers to the adoption and combination of technologies such as “Continuous Integration/Continuous Delivery” (CI/CD), “Infrastructure as Code” and “Artificial Intelligence” (AI); and methodologies such as “Agile” and “DevOps;” and service models such as “Cloud Hosting.” Combined, these areas help IT organizations meet business requirements and deliver business value.
Learning from Failures
I’ve been working in the IT Ops space for more than 30 years. The last 10 of these years have been spent assisting clients of all sizes understand and ultimately make the transformation to 21st century IT models and approaches. While I have a great winning percentage overall, I am experienced enough that my scorecard also reveals a few failed projects over the decade; that is, efforts that produced neither the desired business outcomes nor technical outcomes.
Series: 21st Century IT
- The Right Partner is Better than a Crystal Ball (7 of 7)
- When Best Efforts Aren’t Good Enough (6 of 7)
- Cloud: The Mirage of Massive Cost Savings (Ketchup on the side) (5 of 7)
- A Tale of Two Models: Provisioning vs. Capacity (4 of 7)
- Cloud, All In or All Out? (3 of 7)
- Revolution, not Evolution (2 of 7)
- You are currently reading: How to Tell Ahead of Time If Your IT Transformation Project is Going to Fail (1 of 7)
During my journey, I’ve been able to learn from my own mistakes as well as from those of others – I’ve observed many projects without being involved. And, while every project is unique in its own way, I’ve recognized that there are a few hallmarks that almost always foretell failure. In this week’s blog, I’ll highlight several of the most prevalent warning signs from a high level. Then, in each subsequent post over the coming months, I’ll go increasingly into more and more depth. I’ll write primarily from a business perspective in this series because most transformation projects that will ultimately fail can be identified before the first engineer is assigned. This comes from a fundamental misunderstanding – at the executive level – of the potential impacts of new concepts such as Cloud Adoption, DevOps, and CI/CD, and principles such as “Fail fast, fix fast.” A quick note before I go into this week’s list. Your particular cloud initiative is not necessarily doomed to failure because one or two of the factors described below apply. That being said, the odds can quickly tip in favor of failure as more and more applicable issues appear.
Your particular cloud initiative is not necessarily doomed to failure because one or two of the factors described below apply.
The List of Dreaded Pitfalls
1. You did not start your transformation with an agnostic application or business feature-based assessment of your current IT estate.
This, more so than any other item in this list, will – if not properly performed – lead to budget/timeline overruns, failed deployments and ultimately unhappy internal/external clients. A properly performed assessment should answer the following questions, at a minimum:
- What are the business requirements of my IT estate?
- Why should I transform my IT/What do I want out of it?
- What is my current application/business feature inventory?
- For each business feature, what is (or are) the:
- Actual infrastructure requirements
- Currently deployed infrastructure
- Licensing requirements
- Cost of operation per month
- Business Continuity/Disaster Recovery posture
- Actual cost (in lost productivity/revenue) of availability per hour and workday
- Governance model
- Security/compliance requirements
- Scalability requirements
- Associated development, Quality Assurance, Configuration and sandbox environments
- Integrated applications, and
- Ideal post-transformation destination (i.e., SAS, Cloud, AWS/Azure/GCP, Physical/Virtual Infrastructure, or other).
- Based on the inventory above, what functionality or performance-related issues need to be proven out through Proof of Concept efforts before ultimate decision are made?
- What is the appropriate high-level budget/timeline required to complete this work? It’s easy to get lulled into the complacent thought that you’ve been operating your infrastructure and applications for a long time, so you know it very well. In practice however, the knowledge required to operate is not the same as the knowledge required to transform.
This leads to the next red flag:
2. You consider the transformation effort to be an evolution of your previous IT models, practices and tools.
21st century IT is a revolution – not an evolution of what we have been doing for decades. The skills required to transform through this revolution are different than those required to operate the existing state, which are different still from those required to operate the new state. Many efforts have failed because the executives responsible did not understand that a cloud migration does not and should not resemble a data center migration – regardless of what the various migration tool and cloud hosting partners will tell you.
3. Without first performing an assessment to evaluate the skills or effort required, you and your senior IT staff dictated the timeline, budget and technology decisions.
This seems so absolutely ridiculous that it can’t possibly be true. Could you imagine going to a heart surgeon and demanding a transplant, without understanding the impacts or even if the transplant was needed? Of course not. But somehow enterprises do this every day with the beating heart of their businesses – a.k.a. IT – without giving it a second thought.
4. Prior to starting your transformation, you didn’t have a complete understanding of the financial and operational models behind 21st century IT in the enterprise.
Yes, you understand OpEx vs CapEx, and are maybe even able to make a solid “Net Present Value of Cash” argument regarding your future IT directions. But have you considered Provisioning vs. Capacity planning models? Do you understand the cost and value related to Infrastructure as Code? Can you articulate the risks inherent in Best Efforts Availability as opposed to Five Nines? How will you manage costs in a world where a single button click can result in thousands of dollars in Monthly Recurring Costs?
5. You view the transformation as a strictly technical effort.
This is a big one. To make IT more efficient through transformation, you have to address two areas: The How and the What. Assigning the work to your IT organization addresses the How. Involving business resources drives the definition of the What. Without both groups working together, operational efficiencies won’t be realized, budgets will be blown, and opportunities lost. If some of the issues listed above are factors in your current transformation effort, it’s never too late to try to resolve them. In the coming months, I’ll expand on the thoughts above and share some anonymous war stories while providing pointers on how to avoid pitfalls in the business of 21st century IT transformation. Andrew Pope is a Co-Founder of effectual, Inc.
21st century IT is a revolution – not an evolution
The skills required to transform through this revolution are different than those required to operate the existing state, which are different still from those required to operate the new state.