Explorium MCP

@Explorium what external datasets are available that could enrich our customer demographic records with household income estimates?

Based on your query, Explorium's catalog includes three relevant datasets: US Census Bureau ACS 5-Year estimates (household income by ZIP+4, updated annually), Experian ConsumerView (modeled income brackets at address level, monthly refresh), and a third-party aggregator offering property-value-derived income proxies. The Census data requires no additional licensing. ConsumerView and the aggregator both have per-record pricing. All three can join on US postal addresses. This is an illustrative example of how the MCP surfaces dataset options.

This example shows dataset discovery, a core Explorium capability. In practice, results depend on your Explorium account tier and the vendor's current catalog. API key must have read access to the data marketplace. No data is actually joined or downloaded by this prompt alone.

@Explorium enrich our leads table with company firmographic data — employee count, industry classification, and revenue band — using the domain column as the join key.

Enrichment job initiated. Explorium will match 1,847 rows from your leads table against Crunchbase and Dun & Bradstreet datasets using the domain field. Estimated completion: 4 minutes. Three columns will be appended: employee_count_range, naics_code, and revenue_tier. Rows without a domain match will return null. 12 credits will be consumed (per-record pricing applies). This is an illustrative example; actual job IDs and timing vary by dataset size.

This demonstrates a write action that modifies your data. The MCP submits an async enrichment job; you'll need to poll or wait for a webhook callback. Verify your Explorium credit balance before running large enrichments. Failed matches don't refund credits.

@Explorium compare the coverage and freshness of weather datasets that include hourly precipitation — we need to backfill historical data for our logistics model.

Three datasets match your criteria. NOAA ISD covers 30,000+ global stations, hourly resolution, free, but lags 48 hours behind real-time. Weather Underground offers 15-minute intervals for 250,000 personal weather stations, near-real-time, paid per query. IBM's Weather Company provides modeled hourly precipitation globally with 1-hour latency, subscription-based. NOAA has the longest historical archive (back to 1901 for some stations). This is an illustrative comparison; actual dataset availability depends on Explorium's current partnerships.

This example pairs the MCP's catalog search with AI reasoning to synthesize a comparison. Useful when you need to evaluate trade-offs (cost vs. coverage vs. latency) before committing to a dataset. The MCP surfaces metadata; the AI interprets it for your use case.

When you need third-party signals to qualify leads faster

A 6-person sales team is manually researching company firmographics and technographics before each demo call. They spend 15 minutes per lead on LinkedIn and BuiltWith, then paste notes into Salesforce. Explorium MCP can pull external data signals—company size, tech stack, funding rounds—directly into Switchy threads during pipeline review. The win here is speed: instead of context-switching to five browser tabs, the team asks the AI to enrich a list of 20 accounts in one prompt. The threshold is data freshness. If your ICP changes weekly or you need real-time signals (like a competitor's product launch), you'll still need a human to verify. But for stable enrichment workflows where API key auth is enough, this MCP turns a 15-minute research task into a 90-second query.

When your PM needs industry trends without a data analyst

A 3-person product team at a B2B SaaS startup is planning Q2 features. The PM wants to validate demand for a new integration by checking adoption trends across their target vertical, but they don't have a data team or expensive market research subscriptions. Explorium MCP can surface industry datasets—adoption rates, competitive landscape shifts, regulatory changes—through conversational queries in Switchy. The team runs a shared thread during roadmap week, asks for trend data on specific verticals, and references the output in their PRD. This works when the questions are broad enough for third-party data to answer. If you need proprietary customer usage data or real-time sentiment, this MCP won't help. But for directional market intelligence that informs prioritization, it's faster than hiring an analyst or waiting on a vendor report.

When you want external signals before a QBR, not after

A 4-person customer success team is prepping quarterly business reviews for 30 enterprise accounts. They want to flag churn risk early by checking external signals—layoffs at the customer's company, budget cuts in their industry, leadership changes—but they're pulling this from news alerts and manual Googling. Explorium MCP can automate the lookup: the CS lead drops a list of account domains into a Switchy thread, asks for risk signals, and gets a summary of external events that might affect renewal. The trade-off is scope. If your churn model depends on product usage data or support ticket sentiment, you need your own analytics stack. But for external context that's hard to track manually—macro trends, org changes, competitive moves—this MCP gives CS teams a head start before the customer brings it up.