If you have a list of “bad things to avoid with my Salesforce investment” then dirty data should be in the top 3. It’s right up there with user adoption and executive commitment, in reality the 3 are interrelated. Maintaining high standards of data quality is tough and a continual process or battle in some cases. Once dirty data creeps in it becomes a huge trust issue as management and users alike don’t believe the data and usage spirals down. If you can’t trust the data, you can’t trust the system so why bother using it. It enters into your system with breathtaking ease and before you know it, using Salesforce is considered a waste of time. Because of dirty data.
Of course the right time to deal with dirty data is during the implementation then continually throughout usage. That’s really easy to say but few of us are blessed with the gift of hindsight so it probably needs to be dealt with post-implementation. I’ll break this post into 2 sections, dealing with the mess after the event and staying on top.
Dealing with the mess after the event
At some point you need to say “enough is enough” and deal with the mess. Make the assumptions that it’s not going to get better and data entry behaviours won’t change because Jedi mind tricks don’t actually work (sorry for that bombshell). Step 1 of dealing with the mess is understanding the problem.
- Dirty data generally creeps into your system from the following areas;
- It’s easier to create a new record instead of searching to see if it already exists. So we end up with duplicates.
- Its often too much effort to enter the correct information into a given field because it’s not immediately to hand. Phone numbers are a common example – who’ll know it’s wrong?
- Contact and account data imports from users often generate duplicates
- Data from other systems may not be clean on entry. We’ve seen a SAP integration that had multiple codes for identical products. Therefore consolidated sales reporting needed an Excel spreadsheet as different salespeople used different codes for identical products
- Missing data entries can also count as dirty if they’re required by a colleague or other department
- Records that have not been updated and become invalid over time. Old opportunities are a common example – run a report of opportunities with a close date in the past.
It’s often the case that dirty data is a combination of some/all of the above.
To help you through Operation Clean-Up there are some great tools on the App Store, notably Duplicate Check for Salesforce (free), Cloudingo and RingLead. The available tools have a range of different functions for de-duplication, merge, convert and cleanse. There isn’t a universal choice of tool, it depends on your data cleansing needs.
Staying on top
All or some of the points below should help you stay on top.
- Speak to users and understand why data got into a mess. The reasons may surprise you and give you a good indication of how users interact with Salesforce. In many cases you’ll find confusion rather than apathy so a training refresher may be required.
- Understand where users gain benefit. Salesforce, implemented correctly, will see users gain benefit from their own data. Help your users understand how they can benefit from Salesforce and their own data so they’re motivated to ensure data is correctly entered.
- Ensure data entry screens are logical and quick to navigate. Cut out anything that doesn’t add value to the user using page layouts or record types. Focus on fewer fields and clicks to complete forms, laid out on the most logical manner.
- Use validation on critical fields for data quality. Be careful not to validate everything because you can, it’ll just make users resent it.
- Use tools such as Data.com to maintain data quality. Data.com is on of many tools that will help maintain your data quality.
- Consider disabling quick create which ignores data validation rules. It’s one of the quickest and easiest ways to introduce dirty data into Salesforce.
- Use the carrot and stick principle of rewarding your star users and calling out the poor performers. Chatter is a great tool to make sure the company knows who to recognise and who not.
- Create reports and dashboards focused on data quality. Use RAG graphics on forms to visually alert users to issues. Review reports frequently and correct behaviours if data quality is starting to decline.
- Check field completeness with free apps such as Field Trip.
Above all, the battle for data quality is won on 3 fronts:
- A culture of clean data. Teams see a virtue in clean data, individuals link clean data will job performance.
- If it’s not visible in SFDC it doesn’t exist.
- Train users continually and make sure Salesforce benefits them