How can Clarivate Analytics balance the need for comprehensive patent data in Derwent Innovation with the desire for faster database updates and reduced processing times?

Introduction

Clarivate should evaluate a staged model in which a validated source record becomes searchable before later enrichment adds normalization, family relationships, indexing, and expert-authored summaries.

Verified product context and interview assumptions

• Current name: Clarivate now presents the product as Derwent Patent Search (formerly Derwent Innovation). Its official product page emphasizes global patent data, AI search, and descriptive invention summaries. Source: Clarivate Derwent Patent Search
• Why enrichment matters: Clarivate says DWPI subject-matter experts analyze, summarize, and index new patent publications, and that the dataset groups publications into invention families. Source: Clarivate DWPI
• Interview premise: Assume some customers have asked for fresher records or alerts. Do not assume that current update cycles are slow, that competitors update more often, or that one processing step is the bottleneck until data confirms it.
• Unknowns to clarify: Source-publication cadence, latency by pipeline stage and authority, quality contracts, customer workflows, entitlement rules, architecture, staffing, and contractual service levels.

Step 1

Clarifying Questions (3 minutes)

• **Speed and completeness:** Which latency matters, and which fields or relationships must exist before a record is usable for monitoring, prior-art search, freedom-to-operate work, or portfolio analysis?

Why it matters: One average update time and one completeness score can hide stages and workflow-specific risks. Ask for: Source-arrival and stage timestamps, field-level requirements, contracts, support issues, and baselines by patent authority and publication type.

• **Workflow and risk:** Which customers need earlier visibility, what decision do they make, and which errors could materially mislead that decision?

Why it matters: Freshness has value only when it improves a real task, while missing a classification and presenting stale legal status carry different risks. Ask for: Query and alert behavior, exports, interviews, correction history, severity taxonomy, and source-of-record policy by workflow.

• **Bottleneck and delivery consistency:** Where do source delay, queueing, retry, manual review, or rework occur, and must users with the same entitlement receive the same record state across search, alerts, exports, and APIs?

Why it matters: Clarivate cannot fix authority publication delay, and silent delivery differences can make research irreproducible. Ask for: Queue and source-arrival data, entitlement rules, version history, audit logs, and cross-channel consistency requirements.