Customer data platform at Sprinklr

Sprinklr is a complex enterprise platform spanning more than twenty products. Every customer generates signals across all of them: purchases logged in Salesforce, product consumption coming from the platform itself, support tickets in a separate module, and health scores filled in manually by success managers in Excel and processed later. No single person had a coherent view of any of it.

The result was that field teams, success managers, account executives, renewal managers, only looked at customer data when they absolutely had to. Before an annual business review. Before a renewal. When a customer had already flagged a problem. By then, pulling everything together took hours, and they were always reacting, never ahead of it. The signals were there. Nobody was reading them in time.

Three distinct groups needed this, and they needed different things from it.

I led the initiative end to end, from discovery through to internal adoption. This meant working across data engineering, field teams, and executive stakeholders to define what the platform needed to do and how the data would flow into it. I partnered with one other PM specifically for the GTM and field rollout phase, where the volume of training and feedback calls across success teams required two people on it full time.

This was a CTO-led initiative, which shaped how it moved. Decisions that might have taken weeks got made in days. That also meant the bar for what we shipped and how we presented it was consistently high.

The first question was not what to build. It was what data actually meant anything. Sprinklr has more than twenty products, each with its own consumption patterns. Raw usage numbers without context are noise. A customer could have high consumption driven entirely by automated rules, which looks healthy on a dashboard but signals nothing real about engagement.

I ran discovery across field teams and executive leaders to understand two things: what signals they already relied on when assessing a customer, and where their current process broke down. From that I worked with the respective product managers for each Sprinklr product to define adoption metrics that reflected genuine, intentional usage, not just volume.

Once the right metrics were identified, the pipeline work began. Consolidating data from Salesforce, the platform, the support module, and manual inputs into a single clean feed was the foundational piece. Everything downstream depended on getting this right.

A unified customer data platform that gave field teams, leadership, and end customers a single, coherent view of customer health for the first time. Success managers stopped building spreadsheets before every customer call. Regional leaders could assess portfolio health at a glance without waiting for reports. Customers could see their own consumption data directly, without routing requests through their success manager.

The platform became the primary tool field teams used to identify at-risk accounts before those accounts raised a flag themselves. Within the first quarter of going live, it contributed directly to improvements in ARR and net revenue retention, and a measurable reduction in churn, driven by the ability to spot and act on signals proactively rather than reactively.

The hardest part of this project was not building the pipeline or designing the dashboard. It was earning trust with the field teams who had been burned by inaccurate data before and were skeptical that this would be any different. Every data point they questioned had to be investigated and corrected quickly, or the whole thing lost credibility.

That taught me something that I now think applies to most internal data products: adoption is not a GTM problem. It is a data quality problem. Once the numbers felt right to the people using them, adoption followed without much pushing. The training sessions and feedback calls were valuable, but the real unlock was fixing the data until teams trusted what they were seeing.