Roche Diagnostics automates data cleanup with Plauti Deduplicate rule‑based auto‑merge
We sat down with Ana Marquez Salazar from Roche Diagnostics. She shared how Roche’s CRM team streamlined their workflow by replacing manual data merges with reliable automation. They also fostered trust through colleagues’ collaboration and clear, transparent rules.
Martina: Hi Ana. Thank you for joining us. Could you tell us a bit about the company you work at and your current role?
Ana: Of course. I am Ana Marquez Salazar. I work at Roche Diagnostics as part of the global CRM and Digital Sales team, where I am one of the global CRM managers.
Martina: Could you describe the data quality challenges your organization faced and how you were managing before Plauti?
Ana: The main challenge was data structure, using our CRM constantly means every customer interaction can create its own record. Over time, this leads to multiple records for the same person, fragmented record histories, and a poor customer experience with things like duplicate communications. Previously, merging duplicates was an entirely manual task handled by our Global IT but now with Plauti this is an easy and fast activity to do.
Martina: So before Plauti, all deduplication was manual?
Ana: Yes, exactly. There was no way for teams to do massive or automated merges; it was one record at a time or done in different chunks due to limitations in the amount of data volume that could be processed. That was the case across business teams in different countries, especially when cleaning up after massive data imports, it meant a lot of manual checks.
Martina: How big of a burden was this for teams in terms of time and effort?
Ana: I don’t have specific numbers for time spent yet, since this is an activity managed by other colleagues/teams, but local business colleagues have told us that it was a significant and frustrating workload.
Martina: You mentioned using Plauti because of its flexibility. Could you explain what made a difference compared to other tools?
“ When we compared Plauti to other tools, the key advantage was granularity. With Plauti you can set very specific merge criteria: choose which fields to compare, set thresholds, and control what happens with empty fields. Also, Plauti lets us run automatic, large-scale merges, where other tools require everything to be done manually, one by one. ”![Ana Marquez Salazar]()
Ana Marquez Salazar Roche Diagnostics
Martina: Can you give a practical example of how you made use of this flexibility?
Ana: Sure. In our setup, records can be created manually or digitally, or come from external systems. During a merge, rules need to decide what information to keep as “the winner.” For example, if both are manually created, we keep the most recent info for things like email, phone and consent. If it's digital versus manual, we prefer the digital, unless a relevant field is blank. If it’s two digital records, we avoid merging altogether to prevent disruptions for customers logging into our portals or placing orders. This level of customization was not possible elsewhere.
Martina: And what benefits did these new capabilities bring the teams?
Ana: Data management became much simpler. Instead of spending time on manual checks, teams can now run bulk merges with a button and be confident in the results. People appreciate finally having a tool that just works, you hear a lot of “it was about time.” Our colleagues can trust the database, and customer care sees fewer issues caused by duplicates.
Martina: Are there business outcomes you can share, in terms of satisfaction or efficiency?
Ana: The feedback from local colleagues has been very positive. Customer experience is better, no more sending duplicate communications or updating the wrong record. Colleagues spend less time figuring out what’s what, and the data they use is more accurate, which also makes decision-making easier for everyone.
Martina: Is Plauti Deduplicate also helping admins or mainly business users?
Ana: Both of them, 1) For admins in the Deduplicate instance, since now they can handle auto-merge operations efficiently. The process is much less labor-intensive, and they can focus on more strategic work and also 2) For our business colleagues, since leaving duplicates mean person info cannot unequivocally be linked to an individual, marketing activities and campaigns are allocated to wrong people which make them more difficult, digital services cannot be leveraged easily and most importantly, data compliance requirements are not met.
Martina: When launching a new solution like this, getting buy-in is often a challenge. How did you approach selecting the first users or testers, and did you involve influential colleagues on purpose to help drive adoption?
Ana: We’re running pilots in selected countries. This helps us refine the merge rules and build success stories before rolling out globally. We’ve also made sure local business colleagues who are recognized for being CRM experts, are involved in testing and refining the tool: when they say it works, others trust the process.
Martina: When rolling out automation at this scale, what concerns or objections did you face?
Ana: There was a lot of concern about data loss: what happens to related data like assets or orders, and what if someone picks the wrong main record. We make sure everything is merged to the master record. Some also fear losing control, so we take time to clarify that with rules in place, the process is safe and reliable.
Martina: How did you help teams trust the automation instead of manual work?
Ana: It’s about clarity and transparency. We explain there’s no magic: auto-merge uses the same logic a person would, just faster and at scale. Involving detail-oriented, trusted users as testers and advocates made a huge difference. When our most experienced colleague said it worked as intended, everyone followed her lead.
Martina: Can you clarify for others the practical difference between auto and manual merge?
Ana: Manual merge means a person reviews records, compares data, and chooses what to keep. Auto-merge runs the same rules automatically, eliminating repetitive work and mistakes. Once people see that automation actually makes the process safer, because of consistent rules and fewer human errors - they’re much more confident.
Martina: Which customization and automation features have had the most impact?
Ana: The most important is being able to compare extra fields and insert logic to always keep the most up-to-date consent or contact data, not just blindly favor the winner record. That’s made a real difference in accuracy and compliance.
Martina: What advice do you have for organizations looking at data management automation?
“ Really understand your data’s structure before setting up automation - every field, every scenario. Build your merge rules based on what you actually need. Involve your experts early. And always focus on validation and intelligent comparison rules, not just default settings. ”![Ana Marquez Salazar]()
Ana Marquez Salazar Roche Diagnostics
Martina: Are there any outstanding support or rollout points you’d like to mention?
Ana: We’re in production in three countries and hope to add three to five more in coming months. Support from Plauti has been excellent - very collaborative. We’ll be able to share more concrete statistics as rollout expands.
Curious to learn more? Watch the webinar with Ana Marquez Salazar, CRM Manager at Roche, to see how they set granular merge criteria, protected consent data, and rolled out pilots safely.
