Cleansing the complex

Cleansing CRM data doesn’t have to be an overwhelming task - it can be an easy, manageable and efficient process, as Oleg Rogynskyy, CEO of People.ai explains.

The origin of Customer Relationship Management (CRM) can be traced back to the 1990s, when companies such as Siebel helped gradually drive the evolution of contact management software towards CRM systems. Previously, CRMs were built on hierarchical databases, but these have since been wiped out by SQL (Structured Query Language) CRMs. Since then, the likes of SalesForce have moved SQL CRM into the cloud, but the problems that inhibited the platforms 20 years ago, such as inaccurate, incomplete and untrustworthy data, still exist today. 

This is a problem that limits the true potential of CRM software. The technology was built for static data while today’s business data is, in fact, very dynamic. Information is constantly developing and so can quickly become outdated. The current use of CRM is like using flipbooks to try to watch a movie: the method has become obsolete and overtaken by newer, more efficient forms of technology. 

The main issue is that modern CRM platforms, despite their sophistication, focus primarily on processing and consuming data instead of collecting and keeping it accurate. According to Ben Horowitz, we have witnessed the demise of systems of record from the rise of AI. CRMs were built in the point-in-time sales world, meaning that they were built in the days of one-time sales, where activity data and the dynamic nature of contacts didn’t matter. Since then the world has transitioned into a continuous sales world, leading companies like Zuora and Gainsight to try to fix the point-in-time nature of CRM and successfully address data inaccuracy and duplication. 

Specialised tools

A ‘CRM Scan’ can quickly identify data quality metrics and incorporate them into an overall metric called the ‘CRM Health Score’, revealing where efforts need to be focused. This assessment sheds light on CRM fitness and, when combined with a strong understanding of how sales and marketing teams are using the activity data, elevates confidence in prioritising efforts to improve the CRM system. 

Within this process, it is paramount to focus on three primary dimensions of CRM data quality to establish the baseline: 

1. Is the activity data complete?

2. Is there a single representation of the activity data?

3. Does the activity data correctly represent the real world?

Although it is possible to create the metrics internally, this would take several weeks. Not only does this discourage teams who are investing significant time in this work, but it also paralyses them as they often don’t know where to start or whether their efforts are making a difference. 

Important first steps

Identifying data duplication is another hurdle that can undermine productivity. Duplication is typically due to a lack of standard and unique identifiers for companies and the people that work for them. Despite the use of common proxies, including web domain and email addresses, these are often not unique, as the names of companies and people can change or have variations. To tackle duplicates, businesses need to: 

1. Define duplicates 

The first step is to define what is considered a duplicate. For instance, in contacts and leads this can be email address matches, identical name matches and account associations. 

2. Set up preventative dedupe rules in the CRM

Businesses should then use features established by Salesforce to block and prevent the creation of duplicate records. 

3. Identify and clean existing data duplicates

The ‘CRM Scan’ can be used to identify duplicates and clean them up. This requires some planning based on the CRM system in use. There are specialised tools that make this process easier, but in some cases it can be a good step to reinforce the process by taking it offline to use spreadsheet analysis. 

4. Implement ongoing monitoring for new duplicates

Once data duplicates have been identified and cleaned, it is important to set up preventative de-duplication rules in the CRM platform to monitor and repair duplicates. 

Quick, visible results 

Specialised scan tools, custom reports and dashboards are used to identify, clean and enrich data. This focuses on finding invalid data, such as digits or special characters in contact names, email addresses, web domains and incomplete mailing addresses. This can be done by combining spreadsheets and simple scripts to build update files for a CRM loader, as well as using a database built for this purpose. 

The timescale of this process varies depending on data quantities, the number of duplicates and the amount of data that needs cleansing. With the right tools, reliable measurement and on-going commitment, results can be visible almost immediately. 

In order to achieve this, organisations need to set targets that are tied to business priorities. This will enable businesses to communicate results, rebuild trust in the data and celebrate milestones to keep the momentum going. Benefiting from CRM data doesn’t have to be overwhelming, impossible or disheartening. It can be relatively easy, straightforward and more than satisfying.