---
title: "Real-Time Tracking for CPG in Southeast Asia in 2026: Why the Outlet has no Coordinate to Track Against"
id: "26570"
type: "post"
slug: "real-time-tracking-cpg-southeast-asia-2026"
published_at: "2026-09-15T13:30:00+00:00"
modified_at: "2026-09-15T13:24:55+00:00"
url: "https://locus.sh/blogs/real-time-tracking-cpg-southeast-asia-2026/"
markdown_url: "https://locus.sh/blogs/real-time-tracking-cpg-southeast-asia-2026.md"
excerpt: "Geofenced arrival detection assumes the destination has a reliable coordinate. In dense general trade it does not, and no fence radius produces a verified visit more than 7% of the time."
taxonomy_category:
  - "General"
---

#### [General](https://locus.sh/blogs/category/general/)

# Real-Time Tracking for CPG in Southeast Asia in 2026: Why the Outlet has no Coordinate to Track Against

[Aseem Sinha](/author/aseem_locus/)

Sep 15, 2026

15 mins read

Real-time tracking for CPG distribution in Southeast Asia is the practice of confirming that a delivery vehicle reached a specific retail outlet at a specific time, across route structures that commonly run 40 to 80 stops per vehicle per day. It breaks in general trade for a structural reason rather than a technical one: the small independent outlets that make up most of the channel have no reliable address and therefore no stable coordinate, so a geofence has nothing accurate to be drawn around. The consequence is that arrival events degrade into driver self-report, and a tracking system that looks complete on a dashboard cannot actually verify which outlet was served. Locus, the world’s first agentic Transportation Management System, treats the outlet coordinate as something the delivery network earns through verified transactions rather than something the address database supplies.

## Key Takeaways

- In a dense general trade cluster with a mean geocode error of 80 metres, no geofence radius produces a verified visit more than about 7% of the time, because the radius that catches the arrival also swallows neighbouring outlets.
- The same geofencing feature returns roughly 98% verified visits in modern trade and 7% in general trade, in the same city on the same day.
- Anchoring coordinates on the first verified transaction lifts general trade from about 7% to about 47%, which is a large gain that still does not reach modern trade performance.
- The residual gap is outlet density, which no data quality programme can remove, so the delivery transaction and not the geofence must be the system of record.
- Locus resolves addresses natively and anchors outlet coordinates from confirmed deliveries, with a global FMCG deployment across ten Asian markets reaching 1.8 million retail outlets.

## Why Address Quality Decides Tracking Performance in Southeast Asia

Southeast Asia is a large and fast-growing distribution market, which is why the tracking problem matters commercially rather than academically. The [e-Conomy SEA 2025 report](https://www.temasek.com.sg/en/news-and-resources/news-room/news/2025/e-conomy-sea-2025-report-aseans-digital-economy-poised-to-surpass-300-billion)
 from Google, Temasek and Bain puts the region’s digital economy on track to pass $300 billion in gross merchandise value in 2025, growing at 15% year on year across a population of more than 680 million. CPG volumes ride on the same road networks and the same outlet base.

The structural fact that governs tracking is channel mix. A large share of consumer goods volume in Indonesia, the Philippines, Vietnam and Thailand still moves through general trade, the dense network of small independent outlets known locally as warungs, sari-sari stores and their equivalents. These outlets frequently have no formal street address, sit inside lanes that do not appear in commercial map data, and are identified operationally by landmark rather than by number.

Satellite positioning is not the limiting factor. The [US government publishes GPS accuracy of around 4.9 metres](https://www.gps.gov/systems/gps/performance/accuracy/)
 for a smartphone-class receiver under open sky, which is far tighter than the spacing between adjacent outlets. The error that breaks tracking is not in the vehicle’s reported position but in the recorded position of the destination, and that error is commonly one to two orders of magnitude larger.

Independent assessment points at the same weakness. In the World Bank’s 2023 [Logistics Performance Index](https://data.worldbank.org/indicator/LP.LPI.OVRL.XQ)
, Indonesia fell to 63rd of 139 countries with its score easing from 3.15 to 3.0, and the sharpest declines came in the timeliness and tracking and tracing components rather than in customs or international shipments. Those two components are precisely the ones that depend on knowing where a shipment is and when it arrived, which is the capability that address quality governs.

Density is what converts that error into an unrecoverable problem. Because general trade routes concentrate many stops into a small area, the economics of the channel depend on drop density in exactly the way [McKinsey has quantified for out-of-home delivery](https://www.mckinsey.com/de/publikationen/2024-10-28-ooh-delivery)
, where raising drops per stop from one to five cuts labour and vehicle cost by more than half. The same density that makes the route viable makes the outlets indistinguishable to a geofence.

| Also Read: How Locus Reduces Cost Per Delivery for CPG Distributors in SEA |
| --- |

## How Outlet-Level Tracking Actually Works

### 1. Separate arrival detection from visit verification

Detecting that a vehicle entered an area and verifying that it served a specific outlet are two different claims, and most platforms report only the first while implying the second. Keeping them apart is the precondition for measuring either one honestly.

### 2. Measure your geocode error before you tune your geofence

Sample a few hundred outlets, capture the true coordinate at the door, and compare it with the stored coordinate to get a distribution rather than an anecdote. Every downstream decision about radius depends on this number, and almost no distributor has measured it.

### 3. Test the radius against outlet density, not against detection rate

A larger radius always improves detection, which is why tuning on detection alone drives the radius upward until the fence covers half the street. The correct test asks how many other outlets fall inside the same circle, because that count is what destroys attribution.

### 4. Make the delivery transaction the system of record

Proof of delivery, payment capture, a photograph and a signature are events that cannot be produced from the wrong location without deliberate effort, which makes them stronger evidence than a position reading. Treat the geofence as corroboration for the transaction rather than as a substitute for it.

### 5. Anchor the outlet coordinate on the first verified visit

When a transaction confirms that this vehicle served this outlet, capture the position at that moment and write it back as the outlet’s coordinate. From the second visit onward the outlet is addressable, and the whole problem shrinks.

### 6. Re-verify anchors on a schedule, because outlets move

General trade has real churn: outlets relocate, close and reopen under new ownership. For outlets served weekly this is absorbed easily, but for outlets served monthly or quarterly the anchors decay faster than the route refreshes them, so low-frequency outlets need a deliberate re-verification cycle.

| Also Read: Route Optimisation for Southeast Asia: Why Address Quality Caps Your Routing Gains |
| --- |

## Why No Geofence Radius Solves General Trade

The trade-off can be modelled directly. Take a conservative general trade cluster of 12 outlets within a 150 metre radius, a mean radial geocode error of 80 metres, and a 15 metre positioning error for a vehicle in an urban canyon. A visit counts as verified only when the arrival is detected and no other outlet sits inside the same fence.

| Fence radius | Arrival detected | Other outlets inside the fence | Attribution unambiguous | Verified visits |
| --- | --- | --- | --- | --- |
| 25 metres | 7% | 0.3 | 72% | 5% |
| 50 metres | 25% | 1.3 | 26% | 7% |
| 75 metres | 48% | 3.0 | 5% | 2% |
| 100 metres | 69% | 5.3 | 0% | 0% |
| 150 metres | 93% | 12.0 | 0% | 0% |

The two columns move in opposite directions and their product peaks at about 7%, at a radius of 50 metres. Widening the fence to 150 metres lifts detection to 93% and takes verification to zero, which is the configuration most operations actually run, because detection rate is the number on the dashboard. The density assumption used here is deliberately generous: many real general trade streets carry outlets every 20 to 40 metres, which pushes the peak below 1%.

## What Fills the Gap When Verification Fails

When a geofence cannot confirm which outlet was served, the completion record does not become blank. It quietly falls back to the driver marking the stop complete in the app, which is the weakest available evidence and the one least visible in reporting, because it looks identical to every other completed stop on the dashboard.

That substitution matters because the incentives around it are not neutral. Route completion is frequently tied to driver pay, shift end or performance scoring, which means the one party with an interest in the record is also the only party producing it. This is not an argument that drivers are dishonest, and most are not. It is an argument that a control system should not rest on self-reported data from the party being measured, which is a principle no finance function would waive and most logistics functions accept without noticing.

The test is simple and uncomfortable. Take a week of completed stops and ask how many carry evidence that could not have been produced from somewhere other than the outlet. In most general trade operations the answer is far smaller than the completion rate implies.

## General Trade Against Modern Trade on the Same Platform

| Dimension | Modern trade outlets | General trade outlets |
| --- | --- | --- |
| Typical geocode error | Around 20 metres, formal addresses | 80 metres or more, landmark identification |
| Outlet spacing | Hundreds of metres to kilometres | Tens of metres within a cluster |
| Best achievable verified visits | About 98% at a 75 metre radius | About 7% at a 50 metre radius |
| Limiting factor | Positioning error | Outlet density relative to geocode error |
| Reliable arrival evidence | Geofence crossing | Delivery transaction with payment or photograph |
| Effect of widening the radius | Detection and attribution both stay high | Detection rises, attribution collapses |
| Effect of a data quality programme | Marginal, already sufficient | Large, roughly 7% to 47% |

| Also Read: How to Reduce Failed Deliveries in Southeast Asia |
| --- |

## What Anchoring Recovers, and What It Does Not

Rerunning the same model with the geocode error reduced to 10 metres, which is what first-visit anchoring achieves once a coordinate comes from a confirmed delivery rather than from an address database, changes the picture substantially. At a 25 metre radius, detection rises to 66% and attribution holds at 72%, producing about 47% verified visits. The optimal radius also shrinks, and that is the actual mechanism: accurate coordinates are valuable mainly because they permit a small fence, and a small fence is what keeps neighbouring outlets out.

Coverage accumulates quickly because CPG routes are repetitive, and the rate depends entirely on how much of the outlet base a normal week already touches.

| Share of outlets served each week | Week 1 | Week 2 | Week 4 | Week 6 | Week 8 |
| --- | --- | --- | --- | --- | --- |
| 25% | 25% | 44% | 68% | 82% | 90% |
| 40% | 40% | 64% | 87% | 95% | 98% |
| 60% | 60% | 84% | 97% | 100% | 100% |

The practical reading is that a distributor serving 40% of outlets weekly reaches usable coverage inside a month and near-complete coverage in six weeks, without a data cleanup project, a field survey or a master data vendor. The work is already being done, and the only change is that the delivery app writes the coordinate back.

The honest limit is that 47% is not 98%. Anchoring delivers roughly a sevenfold improvement and still leaves general trade well short of what the same platform achieves in modern trade, because the remaining constraint is outlet spacing rather than data quality. No address programme can move outlets further apart. That is the argument for making the transaction the system of record permanently rather than treating it as a stopgap until the master data improves.

## What to Look for in Southeast Asian Tracking Software

**Native address resolution rather than a third-party lookup.** Commercial geocoding services are trained on formal address systems and degrade precisely where general trade lives. A platform that resolves incomplete and non-standard addresses as a first-class capability behaves differently from one that passes the string to an external API.

**Write-back of verified coordinates into the master outlet record.** Capturing a position at the point of delivery is only useful if the platform updates the outlet permanently. Ask specifically whether anchoring is automatic, whether it is governed by a confidence threshold, and who can override it.

**Transaction-anchored proof of delivery.** Payment capture, photographs and signatures should be bound to the outlet record and timestamped, so that the delivery evidence stands on its own without reference to a geofence crossing.

**Configurable fence radius per outlet, not per route.** A single radius applied across a route that mixes a supermarket and thirty warungs will be wrong for one of them by construction. Per-outlet radii, derived from the measured confidence in each coordinate, are the only configuration that fits the channel.

**Reporting that distinguishes detected from verified.** Insist that dashboards separate arrivals detected by geofence from visits verified by transaction, because a single blended completion number conceals exactly the failure this article describes.

## Outlet-Level Tracking in Action

A global FMCG manufacturer operating across ten Asian markets with 1,000+ distributors and 5,000+ riders reached [1.8 million retail outlets while saving more than 12,000 trips per month](https://locus.sh/case-studies/global-fmcg-logistics-automation/)
 and returning 3X ROI on more than $4 billion of optimised orders. The scale is the point: an outlet base of that size cannot be cleaned by a central data team, so coordinate quality has to be produced by the delivery network itself as a by-product of normal execution.

A leading ASEAN apparel retailer running [multi-carrier parcel management](https://locus.sh/case-studies/apparel-multi-carrier-parcel-management/)
 faced the same underlying addressability problem in a different channel, where the destination is a consumer address rather than an outlet but the address quality distribution is similar. The pattern that transfers is separating the carrier’s scan events from the platform’s own verification of what was delivered where.

A Fortune 50 enterprise operating 51 sites and a 4,500-strong driver pool lifted [weekly execution rate from 75% to 92%](https://locus.sh/case-studies/fortune-50-parcel-centralized-dispatch/)
 after centralising dispatch decisions, alongside more than $14 million of unused capacity identified. Execution rate improves when the system can tell which planned stops were genuinely completed, which is the capability that outlet-level verification provides.

| Also Read: Reimagining TMS in Southeast Asia |
| --- |

## Common Southeast Asian Tracking Mistakes to Avoid

**Tuning the geofence on detection rate.** Optimising the number the dashboard reports drives the radius upward until arrivals are detected almost always and attributed almost never.

**Treating geocode error as a data cleanup project.** Cleaning a general trade master file centrally is slow, expensive and obsolete within months because of outlet churn, whereas anchoring produces coordinates continuously as a by-product of delivery.

**Applying one fence radius across a mixed route.** Modern trade and general trade stops on the same vehicle need different radii, and a single setting guarantees the wrong answer for one of them.

**Reporting completion without distinguishing evidence type.** A completion percentage that blends geofence arrivals with transaction-verified deliveries reads as certainty while resting largely on driver self-report.

## How Locus Approaches Tracking Where Addresses Are Hard

Locus, the world’s first Decision-Intelligent, Agentic TMS, treats address resolution as core infrastructure rather than as an integration detail, because the markets it was built in made that unavoidable. Outlet coordinates are resolved natively from incomplete and non-standard inputs, then improved from confirmed deliveries so that the outlet record gets better every time the network serves it. The [control tower](https://locus.sh/control-tower-software/)
 presents detected arrivals and transaction-verified visits as distinct states rather than merging them, and the Hub, Dispatch and Customer agents act on the verified state. The six governance mechanisms, Explainability, Traceability, Evaluation, Autonomy Levels, Execution Sandbox and Human Review, determine when a coordinate update is applied automatically and when it is held for review. The platform evaluates against 250+ real-world constraints across 1,000+ carriers at 99.99% uptime.

Locus is [recognized by Gartner for seven consecutive years](https://locus.sh/analyst-recognition/)
, featured in the 2026 Hype Cycle for Supply Chain Execution and Logistics Technologies, named a Leader in TMS by QKS Group in the SPARK Matrix, and ranked #1 in Route Planning on G2’s 2026 Best Software Awards. In October 2025, Ingka Investments, the investment arm of Ingka Group, the world’s largest IKEA retailer, acquired Locus. Locus continues to operate independently.

Real-time tracking for CPG in Southeast Asia fails in general trade not because positioning is inaccurate but because the destination has no reliable coordinate, and the fence radius that detects an arrival in a dense outlet cluster cannot also identify which outlet was served. Verified visits peak near 7% before anchoring and near 47% after it, against roughly 98% for modern trade on the same platform, which is why the delivery transaction rather than the geofence has to be the system of record. Locus, the world’s first agentic TMS, resolves outlet addresses natively, anchors coordinates from confirmed deliveries, and reports detected and verified visits separately so the numbers mean what they appear to mean. [Request a Locus outlet-level tracking assessment](https://locus.sh/schedule-demo/)
 to see how your current geocode error maps to verifiable visits.

## Frequently Asked Questions

**What is real-time tracking for CPG distribution in Southeast Asia?** It is the confirmation that a delivery vehicle reached a specific retail outlet at a specific time across routes that commonly run 40 to 80 stops a day. In modern trade this is straightforward, but in general trade the outlet often has no formal address, so the tracking system has no accurate coordinate to compare the vehicle’s position against. The practical output is therefore a verified transaction rather than a map position.

**Why does geofencing not work for general trade outlets?** Because the fence has to be wide enough to absorb the error in the outlet’s recorded position, and at that width it also contains neighbouring outlets. Detection and attribution move in opposite directions as the radius grows, and their product peaks at roughly 7% in a dense cluster with 80 metres of geocode error. There is no radius that resolves both.

**How accurate is GPS for last mile delivery tracking?** Smartphone-class GPS is accurate to around 4.9 metres under open sky according to the US government, and somewhat worse among tall buildings. That is far tighter than the spacing between outlets, which is why positioning accuracy is rarely the constraint. The limiting error is in the stored location of the destination, not in the vehicle.

**What is first-visit anchoring for outlet coordinates?** It is the practice of capturing the vehicle’s position at the moment a delivery is confirmed by transaction, then writing that position back as the outlet’s coordinate. The outlet becomes addressable from the second visit onward, and because CPG routes repeat weekly the outlet base converges quickly, typically passing 90% coverage within four to six weeks where 40% of outlets are served each week.

**How should a CPG distributor measure delivery verification in general trade?** Report detected arrivals and transaction-verified visits as separate numbers rather than as one completion rate. Measure geocode error directly by sampling outlets and comparing the captured coordinate with the stored one, and set fence radii per outlet from that measured confidence rather than applying one value across a route.

**Does improving address data fix tracking in general trade?** It helps substantially, moving verified visits from roughly 7% to roughly 47% in the model above, but it does not close the gap to modern trade. The remaining constraint is how close the outlets are to each other, which no data programme can change, so the delivery transaction should remain the system of record permanently.

MEET THE AUTHOR

Aseem Sinha

Vice President - Marketing

Aseem, leads Marketing at Locus. He has more than two decades of experience in executing global brand, product, and growth marketing strategies across the US, Europe, SEA, MEA, and India.

### Related Tags:

[https://locus.sh/blogs/real-time-visibility-uk-fmcg-booking-slot-2026/](https://locus.sh/blogs/real-time-visibility-uk-fmcg-booking-slot-2026/)
#### [General](https://locus.sh/blogs/category/general/)

## [Real-Time Visibility for UK FMCG in 2026: Why the Booking Slot is the Only Deadline That Counts](https://locus.sh/blogs/real-time-visibility-uk-fmcg-booking-slot-2026/)

[Ishan Bhattacharya](https://locus.sh/blogs/author/ishan_locus/)

Sep 15, 2026

UK grocery supplier manuals grant tolerance to the supplier who warns and withhold it from the one who simply arrives late. That makes notification lead time, not position accuracy, the measure of a visibility system.

[Read more](https://locus.sh/blogs/real-time-visibility-uk-fmcg-booking-slot-2026/)

[https://locus.sh/blogs/cross-border-visibility-eta-error-north-america-2026/](https://locus.sh/blogs/cross-border-visibility-eta-error-north-america-2026/)
#### [General](https://locus.sh/blogs/category/general/)

## [Cross-Border Visibility in North America in 2026: Why 91% of Your ETA Error Sits Where GPS Cannot See it](https://locus.sh/blogs/cross-border-visibility-eta-error-north-america-2026/)

[Anas T](https://locus.sh/blogs/author/anas_locus/)

Sep 15, 2026

A truck in a customs queue is still moving on your map. Most of the ETA error on a cross-border move lives in the release decision, which telemetry cannot observe and your broker already can.

[Read more](https://locus.sh/blogs/cross-border-visibility-eta-error-north-america-2026/)

## Real-Time Tracking for CPG in Southeast Asia in 2026: Why the Outlet has no Coordinate to Track Against

- Share
- [Print](javascript:window.print())
- [Download](#)
- [Schedule a Demo](https://locus.sh/schedule-demo/)

### Is your team spending more time on fixing logistics plan than running the operation?

- Agentic transportation management from order intake to freight settlement
- Route optimization built on 250+ real-world constraints
- AI-driven dispatch with automatic execution handling

20%Cost Reduction

66%Faster Planning Cycles

[Schedule a demo](/schedule-demo/)

Insights Worth Your Time

#### [General](https://locus.sh/blogs/category/general/)

## [Locus 2026 UK Consumer Survey: Why Returns Visibility is Now the Conversion Engine for AI-Driven Shopping in UK Retail](https://locus.sh/blogs/returns-visibility-conversion-engine-ai-shopping-uk-retail-locus-q2-2026-consumer-survey/)

[Aseem Sinha](https://locus.sh/blogs/author/aseem_locus/)

May 29, 2026

#### [General](https://locus.sh/blogs/category/general/)

## [Locus 2026 US Consumer Survey: Generative AI isn’t Just Changing How Consumers Shop, it’s Breaking the Demand Patterns US Retail Was Built On](https://locus.sh/blogs/generative-ai-shopping-effect-retail-fulfillment-operations-locus-q2-2026-consumer-survey/)

[Ishan Bhattacharya](https://locus.sh/blogs/author/ishan_locus/)

May 29, 2026

#### [General](https://locus.sh/blogs/category/general/)

## [Embedded vs Bolted-On AI: The Architecture Question European Logistics Buyers Are Asking](https://locus.sh/blogs/embedded-vs-bolted-on-ai-european-logistics-platform-architecture-business-benefits/)

[Aseem Sinha](https://locus.sh/blogs/author/aseem_locus/)

May 21, 2026

#### [General](https://locus.sh/blogs/category/general/)

## [Hybrid Fleet Management: How Owned, 3PL, Gig, ICE, and EV Capacity Actually Operate at Most Enterprises](https://locus.sh/blogs/three-workforce-fleet-reality-owned-3pl-gig-drivers/)

[Aseem Sinha](https://locus.sh/blogs/author/aseem_locus/)

May 7, 2026

#### [General](https://locus.sh/blogs/category/general/)

## [US Returns Hit $850 Billion in 2025: Why US Retailers Are Restructuring Reverse Logistics in 2026](https://locus.sh/blogs/850-billion-us-returns-ai-routing-reverse-logistics-2026/)

[Ishan Bhattacharya](https://locus.sh/blogs/author/ishan_locus/)

May 7, 2026