About Client
TrueSpot builds connected asset tracking technology for industries that cannot afford to lose what they own. Their platform uses IoT-enabled hardware to monitor the real-time location and movement of high-value physical assets — vehicles, keys, and equipment — across large operational sites.
What They Do
- Real-time location tracking of vehicles, keys, and physical assets using IoT hardware
- Zone departure detection and stationary anomaly monitoring
- Movement pattern analysis across dealership lots and large operational sites
- Location intelligence for missing asset identification and recovery
Their Clients Include
Automotive dealerships, fleet management companies, rental operations, and large-scale logistics providers across North America — from single-site dealerships to multi-location enterprises tracking thousands of assets.
Problem Statement
TrueSpot had the IoT hardware in place and location data was flowing in — but the data was not working hard enough. Finding missing vehicles and keys still required staff to manually search movement logs or walk the lot, and behind the scenes, the same transformation logic was being rewritten across multiple disconnected Power BI files.
The Challenge
The IoT signals were there, but nothing was watching them intelligently. An asset that had not moved in 48 hours looked exactly the same as one parked an hour ago — the system had no way to tell the difference automatically. On top of that, the reporting layer had grown organically over time: different reports contained overlapping transformation logic, timezone handling varied between files, and the same metric could show two different numbers depending on which report a stakeholder opened.
What Was Going Wrong
1. Missing assets hid in plain sight
The hardware tracked every asset, but there was no automated logic to flag abnormal behavior. A vehicle sitting untouched for three days triggered nothing — staff had to manually scan movement logs or physically walk the lot to find it. In the auto dealership world, a single missing vehicle can represent $500+ per day in lost sales, loaner costs, or insurance exposure.
2. Manual inspections ate the clock
Without automated detection, the response to a missing asset report was a physical search. Teams walked lots, checked parking areas, and cross-referenced paper logs against digital records. This was not a backup process — it was the primary process.
3. Reports told different stories
Multiple Power BI reports had been built independently, each with its own version of transformation logic and filtering rules. The same data produced different numbers depending on which file a stakeholder opened. When discrepancies surfaced, the team spent time tracing divergent logic buried in separate files instead of answering the actual business question.
4. Data refresh ground everything to a halt
Full data refreshes took over two hours because the pipeline reprocessed the entire dataset every cycle, regardless of what had changed. A single new file triggered the same processing time as a full historical reload. Stakeholders waited on stale data while the system churned through work it had already done.
Solution Provided
We delivered the engagement in two phases. First, we built a unified Power BI analytics dashboard that automatically surfaces missing assets and exception conditions — no manual review required. Then we migrated TrueSpot’s entire data preparation layer to a centralized Microsoft Fabric pipeline powered by Apache Spark, replacing scattered transformation logic with a single clean source of truth that every report pulls from.
What Is Connected Asset Analytics in Simple Terms?
TrueSpot’s IoT hardware sees where every vehicle and key is at any given moment. Connected asset analytics is the brain that watches those signals and decides what is normal and what is not. A car untouched for two days gets flagged. A key fob that leaves the building after hours gets caught. Instead of a person reviewing hundreds of data points every morning, the system runs that evaluation automatically and puts the answer on screen — what is missing, what moved when it should not have, and what needs action right now.
How the System Works:
01
Ingest and Detect
Raw location signals from IoT-connected assets flow into the system continuously. The Microsoft Fabric pipeline picks up new data files incrementally — detecting only what has changed since the last cycle based on file timestamps, instead of reloading the full dataset every time.
- New asset location data lands in Azure Blob Storage
- The pipeline detects fresh files and skips everything already processed
- Only new records move forward for preparation
02
Clean and Standardize
Every incoming record runs through a series of automated preparation steps built in Apache Spark — normalizing timezones, handling null and missing values, segmenting data by customer, and converting raw CSV files into Parquet format for faster downstream reads.
- Timezone normalization applied dynamically across all regions
- Null and missing values handled systematically at ingestion
- Raw CSV converted to Parquet — faster reads, smaller files, consistent structure
03
Evaluate and Flag
The Power BI analytics layer reads from the standardized dataset and evaluates each asset’s activity against defined thresholds. Vehicles and keys that have not registered movement within the configured window are flagged as missing. Assets that exit designated zones or trigger exception conditions are surfaced automatically.
- Missing Assets module flags every vehicle and key that has gone quiet
- Exceptions Report highlights behavioral anomalies across the fleet
- Exit Locations and Asset Location Exception catch zone and position irregularities
04
Surface and Act
Operations teams open a single dashboard and see exactly what needs attention — filterable by site, asset type, time range, and exception category. Every number on screen comes from the same pipeline, so figures are consistent across every view and every user.
- One dashboard replaces manual lot walks, log reviews, and report cross-checking
- Filters let teams isolate specific sites, asset types, or exception categories instantly
- All metrics trace back to one validated dataset — no conflicting numbers
Results
0
%
reduction in asset search and investigation time
0
%
reduction in manual physical inspections
0
%
faster missing asset recovery
0
%
faster full data refresh cycles — from 2.5 hours to 40 minutes
0
%
faster incremental data loads — from 1.5 hours to 15 minutes
Operations teams stopped spending hours scanning movement logs and walking lots. The dashboard tells them exactly which assets need attention the moment they open it. Missing vehicles and keys are identified earlier, recovered faster, and issues that used to escalate due to delayed detection now get caught before they reach that threshold.
Every Power BI report now pulls from the same centralized, pre-processed dataset. Conflicting numbers between reports are eliminated. Analysts build new reports starting from clean data instead of re-implementing the same transformation logic from scratch.