This article was posted on LinkedIn on April 29, 2026.
Most people dislike entering project data into spreadsheets and project planning applications. The data entry process is divorced from the daily project work and few people ever look closely at the data. It’s often a tedious and bureaucratic task, but it is vital for the success of innovation programs.
In many organizations, reporting systems are designed for leadership visibility—not for the teams doing the work. As a result, the people entering the data see little direct benefit. When reporting doesn’t help teams manage timelines, resources, or risks, accuracy becomes a low priority. Yet the same data increasingly feeds AI dashboards and decision systems, making its quality more important than ever.
Innovation management systems rely on quality data
As organizations grow beyond just a few innovation projects to a larger project portfolio, an innovation management system become indispensable. Such a system works as a database of project information including project plans, budgets, and financials. The best allow dashboards to be set up that show at a glance what projects are on track for timely completion. Dashboards may also display other important information, such as the sustainability impact of a project, its probability of success, whether a project is within budget, and net present value of the project and the whole portfolio.
These systems rely on the accuracy of the data presented. If project data is incorrect, out of date, or incomplete, any business decisions made using that data will be questionable at best. The level of risk of a company’s decisions depends on the data quality. However, most data is entered into the system by humans who are prone to making mistakes and errors. It’s critical to improve data quality as much as possible. To do that we need to find out why data quality is poor, how it can be improved, and how that will help the business succeed.(1)
In most of the organizations that I have worked with, data entry was left up to the project teams and project leaders. There was little checking by the data customers – management and finance departments – that the data was accurate. Some project teams did an excellent job keeping their project plans and financial projections up to date. Others did not really update the data after the launch of the project, or for small projects, did not set up a project plan at all.
Why data quality is poor
This is a general industry problem, and not only in the consumer goods and chemicals spaces. There are several main reasons that data quality is poor.
Data entry is time consuming and boring
Project team members have many other tasks, so entering data in the project plan is often an afterthought. Connectivity of the project planning software with other business applications such as Salesforce and Microsoft Power BI may be problematic, so data has to be entered twice, making data entry even more tedious and error-prone. Project leaders and team members see data entry as bureaucracy because it takes time away from project execution.
Data fields are poorly defined
It may not be clear what data to enter. The methods set up to process data may be inaccurate or have questionable validity, such as estimated sales forecasts, risk analysis, or probability of success calculations.
Business leaders don’t look at data in detail
Business leaders don’t look at data in detail, because they expect it to be correct. In project review meetings, data is often not questioned or discussed at all. This removes the incentive to enter accurate data. Often there are no processes in place to audit data and correct wrong, obsolete, or missing data. Erroneous data remains in place or worse, propagates to other projects or programs.
Performance incentives can orient away from accurate data
Performance incentives can orient away from accurate data, leading to overly optimistic interpretations. Sales forecasts are often taken at face value without a careful evaluation of the assumptions made to develop the forecasts. Risk management techniques such as scenario analysis are rarely done for small projects.
Important data for decision making may not exist
Important data for decision making may not be captured or may not exist. For example, methods or processes may exist that would make teams more effective and efficient. Other data that is not routinely considered includes competitive and patent activity. In some organizations, data is siloed: housed in separate departments with database systems that don’t communicate. The data owners may feel ownership and may be reluctant to share access.
Business impact of poor data quality
Poor data quality impacts businesses in multiple ways. In all these examples, business decisions are compromised because they are based on poor quality data. Sometimes there is a vicious circle when business leaders distrust data quality and use it to a lesser extent to make decisions, leading to project teams spending less effort to improve the quality of their data and to keep reports and dashboards up to date.
Also, in the absence of data audits and other internal checks, project teams are motivated to present data in the most optimistic way. For example, sales forecasts are often not challenged or tested. When forecasts are compared to actual project outcomes, they are very rarely on point.(2) In my experience, sales predictions can be wildly off the actual outcome, both in a positive direction and (more often) a negative one.
None of the outcomes of these deficiencies are good for the organization. Project teams may miss deadlines or launch late. Project budgets may be exceeded, impacting funding for other projects. Business decisions may be based on faulty data, so that resources are not allocated efficiently and innovation opportunities are lost. Artificial intelligence and machine learning models may be trained on inaccurate data.
A culture change is needed
The business impact of efforts to improve data quality is clear, but it requires a culture change in the organization. Such culture change initiatives only take root if they have strong and visible involvement of senior management. Senior leaders need to make it a core objective to address all the deficiencies outlined above. This cannot be left to IT, because data excellence is an organization-wide objective.
Instead, leaders should empower everyone in every part of the organization to address data quality issues. They should challenge data presented in meetings and incentivize accuracy, not rosy presentations. Moreover, generating accurate data is a collaborative effort.(3) Existing data silos need to be dismantled and databases should be integrated so they can communicate.
Importantly, data collection and entry should be fast, intuitive, and integrated with tools that employees already use. No one enjoys entering the same data in multiple tools. Where possible, data collection should be automated, to eliminate data entry errors and reduce mistakes in collection of data. Dashboards may be used to make the impact of the just-entered data immediately visible to project team members.
In conclusion, poor data quality plagues many organizations. It has the potential to seriously impact business performance. Innovation programs including AI implementation may fail if business data is not reliable. Improving data quality is possible, but it requires the focus of the whole organization, including senior leaders.
1. Tom Redman “People and Data, Uniting to transform your business,” Kogan Page, New York, 2023.
2. Marc de Mul, PhD MBA , “How well can innovators predict the future?”, published on LinkedIn, August 4, 2020.
3. Tom Redman , Donna L. Burbank, “How to make everyone great at data”, Harvard Business Review January 6, 2025.