I have been leading IT programs for 10 years. A big part of any implementation includes implementing reports generated out of the system. I have had my fair share of rework in this space. It's tough to design reports when you don't have data to work with, so it becomes an after thought. I wanted to share some of my best practices to help you prevent any mistakes or rework when creating reports.
Here are 8 best practices for report creation:
1) Include key dates - consider created date, closed date (if applicable) as a minimum. I recall making a customer configuration report. The report was generated at the end of the production run, so dates weren't important. Wrong! When you want to look back at a configuration in the future, the dates help with filters and context of time. It's best to add them in the beginning.
2) If a filter exists to generate the report, include the column in the report. Nothing is more frustrating than having to pull multiple reports to slice the data.
3) Order the columns from left to right in the order of the hierarchy. For example, if the report contains product family, part number, serial number and customer, order the columns as follows: customer, product family, part number, serial number. Of course if the product family spans multiple customers, switching the order makes sense.
4) Include quantities if they exist. If possible, keep quantities in numeric format so that calculations in Excel are easier.
5) Include names of people and roles if possible. This helps identify who to call or how to track down an action if the person has moved on.
6) Before making changes to an existing report, double check with the users. I remember one time changing a report layout and it completely messed up someone's Excel macros. It created hours of rework on their end. Oops.
7) If the reports are emailed on a regular basis, check with the users to make sure that the report is adding value and that the right people are on the distribution list. If users don't need the information, remove them from the distribution list. If they are sending the email to a different set of users, correct the distribution list.
8) Check with the users to identify if they are doing manual manipulation such as creating categories or cleaning the data after the report is sent. It is amazing what people will manually do on a regular basis. Focusing on automation and data standards will eliminate this non-value add time, freeing up capacity for more strategic things.
The thing to keep in mind with reports is that most people use them to get access to data and then manipulate the data in Excel. I have seen this with systems like Salesforce, Quality Management Systems, and reporting platforms such as Oracle's BI product. Even visualization tools such as Qlikview, Sisense and PowerBI are used as mechanisms to extract data for analysis. With this behavior in mind, giving people more data is a priority over making the data look pretty on the screen. Understand your end users, what they need the reports for and design accordingly. Happy reporting!