Learn to anticipate risk and recognize the signs of looming problems to keep your digital transformation projects from going off the rails.
Digital transformation projects create the opportunity to deliver value to engineers and their organizations every day. However, like other projects, digital transformation projects routinely face the risk of disasters. With awareness of the source of digital transformation disasters, projects can mitigate their impact by:
- Listing the associated risks in detail in the project charter to educate stakeholders.
- Including specific tasks in the project plan to avoid or minimize the impact of disasters.
- Conducting parallel projects to address these situations before they turn into disasters.
Here is a list of the most common issues that cause digital transformation disasters. Anticipating these issues will position your project for success, allowing the organization to squeeze more value from its data.
Too many data problems
Data problems often overwhelm digital transformation projects. Some typical data problems include:
- Missing values such as incomplete component specifications or missing product descriptions.
- Incompatible key values for data columns, such as customer and vendor codes across systems.
- Incorrect data such as material and supplier codes, dates and discount percentages.
- Missing transaction history where data such as engineering change history or warranty claims is essential to analyzing trends.
- The absence of data quality standards.
Correcting data problems will add to the project cost and extend the schedule, undermining the project team’s efforts. The significant effort required to make corrections will surprise management and potentially reduce their commitment to digital transformation.
To manage data problems in ways that are helpful to advancing digital transformation, engineers can undertake the following actions:
- Recognize the risk of data issues in the project charter to set stakeholder expectations.
2. During the feasibility phase of the project, profile all the potential data sources to determine the extent of data issues.
3. Share the data issues identified with the data stewards and encourage them to take action to make corrections.
4. Start the project by focusing on data sources that exhibit fewer data issues.
Lack of data literacy
Employees’ lack of data literacy is impeding the realization of benefits from digital transformation because they are not using the available digital data.
This lack of data literacy means the planned benefits of digital transformation are not a reality in the organization. The absence of visible benefits will reduce management’s commitment to digital transformation.
To overcome employees’ lack of data literacy, project teams can take the following actions:
- Develop a library of data analytic routines that employees can run and modify to suit their needs.
- Offer in-house, instructor-led training for the available data and data analytic tools.
- Offer one-on-one coaching for employees.
- Point employees to specific YouTube videos that will improve their conversancy with the available data analytic tools.
- Develop a library of frequently requested reports that employees can run and export the data to Excel.
Viewing generative AI as a silver bullet
The explosion of generative AI during the past two years has caused some to view this incredibly capable technology as a silver bullet that can be easily applied to many problems, including digital transformation.
Delivering generative AI features as part of a digital transformation project is not trivial and can lead to undesirable consequences, including:
- Poorly constructed prompts that produce erroneous results and then misleading recommendations.
- Leakage of commercially sensitive intellectual property into the hands of others.
- Risk of inadvertently infringing on the copyrights of others.
- Investment in multiple generative AI software packages that increase cost more than value.
- A more reasonable approach to applying generative AI for engineering applications includes implementing these elements:
- Orient your organization on generative AI capabilities, risks, and limitations.
- Architect a data and analytics environment that will include a data lakehouse to manage both structured and unstructured data.
- Design an AI computing infrastructure, including cloud components that is efficient, scalable, well-governed and at least somewhat future-proof.
- Strike a balance between leveraging vendor capabilities that provide little competitive advantage and developing in-house models and related software that will be costly.
- Choose where to deploy open-source and proprietary technologies.
- Identify which of the many AI use cases are suitable for your company and can deliver tangible business value.
- Build trust in AI-driven solutions through detailed verification of results.
Chasing the latest technology
Some digital transformation project teams become excited by or even fixated on the latest vendor announcements about information technology advances. Examples include:
- Incorporating a sophisticated data visualization software package when a simpler and cheaper one is sufficient.
- Including generative AI capability when it’s of limited value to the digital transformation.
- Using a graph DBMS when a relational DBMS is sufficient.
- Building a data warehouse when integrating data from multiple data sources is straightforward.
- Introducing a new integrated software development environment that is unfamiliar to the organization.
Often, teams see the potential benefits of new information technology without considering how the immature technology will add cost, create delays and introduce quality problems.
Changing technologies or adding more and more technologies mid-project will distract and overwhelm digital transformation projects. Impacts will include reworking software, training staff, and building familiarity with the new technology.
Engineers can take a superior approach by carefully selecting a set of information technologies near the beginning and sticking with the choices for the project’s duration. Information technologies do not advance so quickly that older technologies become obsolete within a system’s planned existence. Engineers successfully use software packages and application development tools that aren’t the latest and greatest every day.
Fantasy business case
Some companies approve digital transformation projects based on an unrealistic business case. Engineers can recognize a fantasy business case because it will include one or more of the following elements:
- Estimated future revenue increases that exceed the historical trend.
- Estimated future operating costs will decrease more than the historical trend.
- The project cost estimate is unrealistically low, does not include a contingency amount and does not recognize the cost of likely change orders.
- There is no discounting of the value of future benefits.
- Quantification of intangible benefits such as brand value or customer satisfaction. Intangible benefits can be essential aspects of digital transformation projects. However, quantifying these benefits is not realistic.
Engineers can promote a credible business case based on tangible benefits and a more reasonable project cost. While digital transformation offers companies many benefits, those benefits often indirectly support other goals, such as reduced operating costs, compressed product development work or increased market share.
