Why projects fail: The dangers of inaccurate project data

We’ve all been bombarded with messages telling us that the right project management (PM) solution is the key to productivity. It’s true–project management systems can do wonders when it comes to aligning team priorities, enabling stakeholder visibility, and, most importantly, maintaining a team’s sanity. 

But what if the same tool creating success could also contribute to process breakdowns? This can be the case regarding inaccurate data in project management systems. 

I sat down with George, VP of product at Zenhub, who has spent countless years thinking about optimizing project management. He believes that one of the biggest reasons projects fail is that the information contained in PM systems–the info that is supposed to keep the team aligned–just isn’t accurate enough. 

Why a lack of data integrity is a bigger problem than we thought

“Data integrity, in the simplest sense, is making sure you have the most up-to-date information about the work your teams are doing,” states George. 

Unfortunately, for most teams, most work is siloed and disconnected from other data sources, making getting an accurate view of what’s happening a scavenger hunt. “The combination of multiple systems used by multiple personas that don’t connect is the biggest source of inaccurate, out-of-date data. For example, developers work in GitHub, product managers work in Notion, and stakeholders view reports in PowerPoint slides. All of those systems are silos, resulting in stale information.”

While, of course, all of these tools have unique use cases, what exacerbates this problem is that most of them are competing to be the “one tool to rule them all.” This causes competing tools to have less than impressive integrations. 

“Imagine a team working in Jira, Slack, GitHub, and Notion. GitHub is a competitor to Bitbucket, so the GitHub-Jira integration isn’t great. Slack is a competitor to Microsoft Teams, so the GitHub-Slack integration isn’t great. Notion is a competitor to Confluence, so the Jira-Notion integration isn’t great, and so on,” George explains. 

To make matters worse, tech companies are incentivized to require users to spend more time in their tools than those they integrate, causing more context-switching.

“That’s what makes Zenhub special. Our goal is to get out of your way so you can focus on the jobs you need to get done. We want you to have to use Zenhub less so you can be the most productive version of yourself,” says George.

How projects fail when they don’t have accurate data

It’s clear why data inaccuracy is such a common problem in the industry, but so what? Why should my team care? After all, most of us have been operating in this less-than-optimal way for a while now. 

Here are four big reasons that might get you to rethink the importance of data integrity: 

1. Misrepresenting information to clients and stakeholders

The first and most obvious danger when it comes to having PM systems that include misinformation is the potential to misrepresent information like project end-dates.

“Years ago, I heard a manager ask an engineer for the latest update on a product we were building. ‘It’s probably going to be another 3-4 days,’ I said. And that’s when they realized that they had incorrectly told stakeholders it would be done sooner because they were relying on an out-of-date system.” 

The stakes get significantly higher when communicating with clients. This is why accurate project management data is even more important for agency and consultancy owners. 

“[Without data accuracy], you start making promises to customers that are impossible to keep because you didn’t realize how far behind a project was, how much scope was left, or how much effort was required. These are potentially business-ending consequences.”

2. Publishing uncompleted work 

Project management systems carry critical information about the status of work, making it easy to pass the baton and move work to the next stage of a process. But what happens when teams miscommunicate task completion?

“Imagine thinking you are supposed to publish a site when you were only supposed to publish a draft, and now you’ve unveiled a product strategy before it was ready. Or seeing an early version of a pricing proposal that didn’t have a full cost analysis and, as a result, putting an app for sale that starts losing money.”

Of course, these are often worst-case scenarios. You might get lucky and be able to restore something from a backup and recover quickly, but you may not. This is why it’s better to play it safe and consider what details must always be accurate in your system. 

3. Discovering blockers too late

Dependencies have been a critical concept in Zenhub for a while–and it’s for very good reasons. “Recently, we worked with a team that had not yet started using Zenhub, so they were not tracking dependencies. When it came time to deploy their service into a new data center, their service was unavailable due to a dependency that had not been worked on.”

This small mistake greatly impacted the team’s ability to ship according to plan. “As a result, they were blocked for a long time, and it was just because one person did not connect that their work depended on someone else’s work,” says George. This could have been avoided if the PM system had been up-to-date and contained the information on dependencies. 

4. Not building the right things

Data in PM systems goes beyond labels, dependencies, and status updates–it can also include the requirements of the work being tracked. An example of how this can go awry is when the Acceptance Criteria (AC) isn’t properly filled out. 

“One time, we were working on upgrading one of our features, and someone read the issue description wrong, so the first iteration of the feature was extremely off. Accurate acceptance criteria is incredibly important because that’s how you know you’ve built the right thing.”

When teammates rely on any type of task instruction contained within a PM tool, there is a risk that the task won’t be completed properly if the data is incorrect. “We figured out the feature was wrong by comparing it against the Acceptance Criteria, which is another reason why it must be accurate.” 

Forward-thinking: improving data accuracy with AI and automation

After discussing the dangers of bad data in a PM system, hopefully, you’ve been inspired to take more action to solve this problem. Luckily, you can implement quite a few simple solutions today to help keep your data as clean as possible. 

From the Zenhub side, many AI tools we’ve built have been made with data integrity in mind, like AI-suggested labels for ensuring label consistency, AI-generated Acceptance Criteria for writing thorough AC, and AI sprint reviews for accurately documenting past work. Additionally, using automations like Smart Pipelines, Workflow Automations, and our Native GitHub integration can help reduce human error that compromises data integrity. 

If you’d like to learn more about how Zenhub can help you improve your project’s data quality, please book a demo here.