FactorPrism®
Case Study · Public NYC 311 Open Data

Explaining the Change in the NYC 311 Dataset

How FactorPrism® analyzed 50+ million public 311 records to explain what drove the change in New York City's service requests—pinpointing the specific categories and time periods behind the trend, not just the headline number.

50 Million Records, Countless Patterns

New York City's 311 service handles millions of non-emergency calls annually, from noise complaints to pothole reports. With over 50 million records spanning multiple years, the NYC Open Data initiative published this vast dataset for the community to analyze.

Manually working through a dataset this large — testing one category and time slice at a time — is slow and easy to lose subtle, offsetting movements in. Even then, cross-dimension intersections — where two provided categories combine — are easy to miss when you test one slice at a time. We decided to demonstrate how FactorPrism® could explain what drove this change — automatically locating the specific categories and periods behind it.

The Approach

Using FactorPrism®'s advanced algorithms, we analyzed the period from September 2013 to March 2017—a timeframe showing clear growth trends. Our goal was to understand whether this growth was broad-based across all service types or concentrated in specific categories — and to locate exactly where.

Key Findings

1. Overall Growth Masking Critical Trends

While 311 usage grew 11% overall during the period, this headline number obscured crucial patterns:

+
Housing Authority (HPD) complaints declined 25% A positive trend completely hidden by overall growth
~
Growth wasn't uniform Concentrated in specific service categories
Seasonality effects varied Amplified in some categories, dampened in others

2. The 2014 Pothole Crisis

FactorPrism® isolated a dramatic spike in pothole complaints in 2014 that was distinct from general street condition issues:

Pothole complaints rose sharply between the baseline and 2014—a localized surge at the pothole complaint type, distinct from the broader street-condition category it rolls up into, that aggregate street-complaint totals would have masked. This type of granular insight allows city planners to understand specific infrastructure failures rather than general trends.

3. Seasonal Movement in Water System Complaints

Water system complaints showed clear seasonal movement over time — humps and give-backs the attribution captured rather than averaging away:

~
Sharp spikes every July and December Like clockwork seasonal stress
Summer 2016: Unusually severe spike A distinct movement, larger than the surrounding periods
+
Winter 2016: Unexpectedly mild Suggests infrastructure improvements

The Power of Automatic Change Attribution

What makes these findings remarkable isn't just their value—it's how they were attributed.

Traditional Approach

  • Multiple analysts coordinating manual slicing
  • Weeks of manual hypothesis testing
  • One category and time slice at a time
  • High risk of missing subtle, offsetting movements
  • Difficulty isolating cross-dimension intersections

FactorPrism® Approach

  • Single analyst
  • A single analysis run, start to finish
  • Automatic change attribution
  • Surfaces localized causes and cross-dimension intersections
  • Separates broad-based from localized causes automatically

Technical Excellence

FactorPrism®'s algorithms excel at this type of analysis because they separate overlapping effects (the 25% decline in housing complaints was invisible in aggregate data), isolate localized causes (the 2014 pothole surge was distinct from general street trends), and capture movement over time — humps, waves, and give-backs — so an unusually mild winter 2016 shows up as a distinct movement rather than being averaged away.

What This Means for Your Business

If FactorPrism® can pinpoint exactly what drove the change across NYC's 50-million-record dataset, it can do the same for the metric that matters in your data:

For Retailers

Separate broad-based category and seasonal movements from product-specific causes — across the product, region, and channel dimensions in your data

For SaaS Companies

Identify which customer segments are actually churning while overall growth looks healthy

For Operations Teams

Pinpoint exactly which categories and periods drove a change in your operational metrics — so you know where to fix a problem or replicate an improvement

Ready to see exactly what's driving your numbers?

Don't let critical insights stay buried in your data. FactorPrism® pinpoints the specific categories and periods behind a change in a single analysis run.

Get it on Snowflake Marketplace