POI Location Specific Insights Overview

POI Location Specific Insights (POI LSI) is an Altitude module that lets operators benchmark any specific location against its competitive environment using first-party commercial vehicle telematics.

What is POI Location Specific Insights?

Rather than relying on consumer-device foot-traffic estimates or manually maintained competitive zone overlays, operators query a location and radius and receive named competitor brands alongside the commercial vehicle metrics that matter for site decisions: stop counts, dwell-time distribution, vehicle class, vocation, fleet size, and fuel type.

The module is designed for site-level competitive analysis. Every query is anchored to a specific point of interest and bounded by a radius, covering the competitive landscape within a defined distance, not brand-level or national data.

You can read the API documentation POI Location Specific Insights here: POI LSI API Module Overview.

Who is it for?

POI LSI is designed for petroleum and gas operators, network planners, and commercial site developers who make site investment, expansion, and competitive-monitoring decisions. Typical use cases include:

Site evaluation: quantifying commercial vehicle traffic and competitor density at candidate locations before a multi-million-dollar investment.
Competitive benchmarking: comparing own-site performance against named competitors within a chosen radius, with methodology transparent enough to present to an investment committee.
Network monitoring: running regular competitor scans to detect new entrants, track trade area stop share over time, and replace manual zone maintenance workflows.
Remote site validation: replacing field visits with on-demand commercial vehicle data at owned and competitor locations.

How it fits in Altitude

POI LSI is built on the same stop-event pipeline and commercial vehicle taxonomy, including vehicle class, vocation, fleet size, and fuel type, as Altitude's Stop Analytics module. Where Stop Analytics answers "what is happening at my site?", POI LSI answers "how does my site compare to competitors within a defined radius?" The two modules are complementary: POI LSI provides the competitive context that internal transaction data cannot.

For operators running multi-module workflows, POI LSI pairs naturally with Open-Ended Origin/Destination (for understanding through-traffic capture at a site versus passing commercial vehicle volume) and AADT (for road-segment commercial vehicle volume context on routes adjacent to a candidate site).

Phase 1 scope

POI LSI Phase 1 is delivered as an API. A visual layer for lower-fluency users is planned for a future release. The Phase 1 API does not include:

Brand-level national queries (all queries are site-anchored and radius-bounded)
Corridor or route-based analysis (queries are radial)
Alerting or change-detection notifications
Unique vehicle count at individual POI level

Attribution and coverage

POI LSI attributes commercial vehicle stops to POIs using an anchor-tenant model. Roughly 40-50% of commercial vehicle stops are attributed to a specific POI. This is not a data gap but the reality of commercial vehicle operations: private facilities, construction sites, staging areas, and other non-POI destinations make up the remainder. A stop is attributed only when GPS trace, dwell time, and proximity align (confidence over coverage), and the same rules apply across every location in a query, so competitive comparisons are valid. See the Methodology page for the full explanation.

Reviewing POI LSI Results

This page explains how to interpret the key output fields from getPOIAnalytics.

POI LSI results are commercial vehicle metrics at a specific point of interest, not consumer foot-traffic estimates and not aggregate trade-area totals. Understanding what each metric represents, and how attribution coverage affects the numbers, is essential for presenting results accurately to internal stakeholders.

Trade area stop share is a market share proxy: the fraction of all attributed commercial vehicle stops in the radius captured by a specific location. Compute it by dividing a location's stopCount by the tradeAreaStopCountTotal field, which represents total stop events across all locations in the radius.

tradeAreaStopCountTotal is always computed across the full radius regardless of any vehicle class, vocation, or fleet size filters applied in your request. This ensures the denominator reflects the full competitive pool, not just the subset you filtered for, so share figures are comparable across queries with different filter configurations. Read trade area stop share as a share of attributed commercial vehicle stops, not as a share of total commercial activity.

Interpreting dwell-time bins

Stop duration is reported across six commercially meaningful bins:


Bin	Range	Typical interpretation (fuel retail context)
1	0-5 minutes	Pass-through; unlikely a fuel or convenience customer. Vehicles that slowed or paused without stopping for a service.
2	5-15 minutes	Refuel or quick-service transaction. Primary indicator of a fuel stop.
3	15-45 minutes	Extended transaction; food, combined services, or short break.
4	45 minutes-2 hours	Meal break or longer driver rest.
5	2-8 hours	Extended rest or regulatory break.
6	8+ hours	Overnight stay or extended parking.

The 0-5 minute bin is the most important for separating pass-throughs from genuine service customers. A high proportion of stops in this bin at a competitor location indicates the site may be capturing incidental slowing rather than real service activity. These interpretations are directional; the actual behavior at any site will vary by location and corridor type.

Interpreting vehicle class, vocation, and fleet size

Commercial vehicle stops are broken out by three dimensions of commercial vehicle identity. These breakdowns are not additive across dimensions; a stop is counted once in stopCount and separately in each breakdown.

Vehicle class (stopCountByVehicleClass): gross weight categories segmenting heavy-duty, medium-duty, and other vehicle types. For fuel retail, high proportions of heavy-duty stops indicate a site that is capturing long-haul commercial traffic.
Vocation (stopCountByVocation): the operational type of the fleet vehicle (for example, long-haul, local delivery, service). Vocation mix reflects which industries are using the site.
Fleet size (fleetSizeBins): the size of the fleet the stopping vehicle belongs to. Applies to Geotab fleet devices only; not a full-universe metric.

Interpreting fuel type

The stopCountByFuelType field breaks stop counts into seven fuel types: diesel, gasoline, electric, hybrid, flexible, other, and unknown. For petroleum and gas operators, the diesel proportion is the most commercially significant; it directly indicates the share of heavy commercial diesel customers in the trade area. Note that fuel type is derived from vehicle telematics, not from fuel transaction records; it reflects the vehicle's primary fuel type, not necessarily the fuel dispensed at the site.

Note:

Each breakdown (vehicle class, vocation, and fuel type) includes a null or unknown bucket for stops where that attribute could not be classified. Breakdowns sum to stopCount only when the null bucket is included.

Understanding attribution coverage

POI LSI attributes commercial vehicle stops to specific POI locations using an anchor-tenant attribution model. Roughly 40-50% of all commercial vehicle stops in a geography are attributed to a named POI; the remainder occur at private facilities, construction sites, staging areas, and other non-POI destinations. This is not a data gap; it reflects the nature of commercial vehicle activity, and a stop is attributed only when GPS trace, dwell time, and proximity align.

Attribution coverage is consistent across locations within a query. When comparing your own site against competitors in the same radius query, the same attribution methodology applies to all locations, so comparisons are consistent across locations. For a detailed explanation, see the Methodology page.

The isSourceLocation and isOwned flags

isSourceLocation = true identifies the anchor POI (the location you queried) in the response. isOwned = true identifies all locations in the response that belong to the same brand as the anchor, both the anchor itself and any owned locations that fall within the radius.

Radius and discovery geography

The radius used in a getPOIAnalytics query is a straight-line distance and is not bounded by the discovery geography used during the getPointsOfInterest step. As a result, competitorDensity from the analytics response will not match the count of locations returned by the discovery step.

POI LSI Methodology

This page describes how POI Location Specific Insights constructs site catchments and attributes commercial vehicle stops to POI locations.

Why catchments

A latitude/longitude marks where a business is registered, not where commercial vehicles stop. A fixed buffer over-captures (adjacent businesses, shared parking); a building footprint under-captures (fueling lanes, truck parking, access roads). POI LSI constructs a purpose-built boundary per location, a catchment, through a five-stage process, then attributes stops within it.

Site and catchment creation

Raw Site: group co-located POIs (shared polygon, or coordinates within approximately 10 m) into a shared building; cluster related buildings into a site. The site, the full physical property, is the unit of analysis.
Buffer: circular buffer around the footprint: 65 m (under 10,000 m2) or 100 m (larger).
Raw Catchment: intersect the buffer with the local road network into a road-bounded land block; prevents bleed across highways or into unrelated parcels.
Service Road Clustering: where adjacent sites overlap, group service roads into entry/access clusters and assign each portion to the site whose traversal zone it falls in; unresolved blocks stay shared. Stable month over month as sites open and close.
H3 Hexagonal Grid Assignment: overlay H3 cells (L13; L12/L11 for larger catchments), assign each to the nearest site, union and clip. Result: non-overlapping catchments, every stop counted once.

Site attribution

Anchor-tenant model. Anchors (fuel, truck stops, warehouses, distribution centres) independently draw commercial vehicles; non-anchors (restaurants, retail) do not. Four rules apply:

Anchor present: 100% to anchors only.
No anchor: split evenly across non-anchors.
Multiple anchors: proportional by POI count per anchor category.
Excluded (weight = -1): Never attributed. This includes ATMs, EV charging stations, and other point-geometry locations.

Percentages are calculated to two decimals; largest-remainder ensures each site sums to 100% and stop metrics sum exactly.

Tradeoffs

Misattribution at the margin: a stop-to-eat at a truck stop is attributed to fuel; per-stop intent is not reliably inferable from GPS.
Private and unregistered locations are not captured (own yards, non-listed distribution centres).
Attributable coverage: not every stop falls in a named catchment (corridors, unmarked, sub-threshold). The methodology favors accuracy over coverage.

Coverage

Roughly 40-50% of commercial vehicle stops are attributed to a specific POI. This is not a data gap; it reflects the reality of commercial vehicle operations: private facilities, construction sites, staging areas, and other non-POI destinations make up the remainder.

Confidence over coverage: a stop is attributed only when the vehicle's Global Positioning System (GPS) trace, dwell time, and proximity all align with a known POI. Coverage is not inflated with uncertain matches.

Truck-stop coverage is higher. Highway-adjacent businesses, including truck stops, travel centers, and rest areas, are well-represented, so attribution for these sites sits at the upper end of the range.

Three principles hold: Defensible (GPS-based spatial matching with dwell-time validation; methodology documented for review), Consistent (the same rules apply across all geographies and time periods), and Comparable (no bias toward any POI type, so location-to-location and brand-to-brand comparisons are valid). Petroleum and gas operators have confirmed this level of attribution supports multi-million-dollar site decisions.