How Location Intelligence Helps Field-Service Businesses

For any business that sends technicians to customers, travel is the cost that grows quietly. Location data turns it into something you can measure, schedule around and reduce.

An HVAC contractor, an appliance repair firm, a mobile vehicle or bicycle repair service and a pest control company look different from the outside, but they share an operating model: a skilled person drives to a customer, does a job, and drives to the next one. Every minute behind the wheel is a minute not spent on paid work. Location intelligence, meaning the habit of attaching good coordinates to every job and using them in decisions, is how these businesses keep that travel under control.

The same ideas apply to two broad patterns. Scheduled service (annual maintenance, planned installations, recurring treatments) gives you days or weeks to build efficient routes. On-demand service (a broken boiler, a flat tire, a failed fridge) arrives with a deadline and has to be slotted into a day already in motion. Most businesses do both.

Geocode at the moment of booking

Everything below depends on knowing where each job is, and the best time to establish that is while the customer is still on the phone or the booking form. Geocode the address immediately, check what level the match reached (building, street or just the town), and if it is not building-level, ask. A short "is this pin on your front door?" with a draggable map marker solves most problems before they reach a technician.

Store the confirmed coordinate with the job, along with any access notes (gate code, which entrance, where to park). Repeat customers then cost nothing to locate the second time. The geocoding in logistics article explains precision levels and delivery points in more depth; the same principles apply when the "parcel" is a technician.

Service-area checks before you accept the job

With a coordinate in hand, you can tell a customer immediately whether you cover their location instead of discovering it at dispatch. The simplest check is a radius around a base: "within 15 km of the workshop". It is easy to compute with the haversine formula, which you can try in the distance calculator, but straight-line distance ignores rivers, highways and one-way systems, so two customers the same distance away can have very different drive times.

Polygon service areas, drawn to follow roads and natural barriers, are usually more honest. They also let you offer tiered terms, such as a standard zone and an extended zone with a travel surcharge or longer lead time. Designing them is covered in service coverage areas. For a quick rectangular approximation of a district, the bounding box explorer lets you draw a box and export it as GeoJSON or WKT.

Travel-time-aware scheduling

A calendar that only knows job durations will happily book a technician into a 9:00 job in the far north and a 10:00 job in the far south. Travel-aware scheduling treats the drive between jobs as part of the schedule:

  • Estimate travel time between consecutive jobs from a routing engine, or at least from straight-line distance with a realistic speed and a margin.
  • Offer customers time windows that are cheap for you to serve: if a technician is already in their area on Thursday, show Thursday first.
  • Group scheduled work geographically, for example by assigning each weekday to a zone, so on-demand jobs can be fitted into nearby gaps.

The detailed mechanics of sequencing stops are in route planning for mobile technicians.

Dispatch: nearest, zone or skills

When an on-demand job arrives, someone or something has to choose who goes. Three strategies are common, and most real systems blend them:

Nearest-available
Send whoever can get there soonest. It minimizes response time for this job, but it can drag technicians out of their area and leave gaps elsewhere.
Zone-based
Each technician owns a territory, and jobs go to the zone owner. It is predictable, builds local knowledge and keeps travel short, but it struggles when one zone is overloaded while its neighbor is idle.
Skills-based
Filter first by who is qualified (a certified gas engineer, a technician trained on a particular brand, someone carrying the right parts), then choose among them by time or zone.
Dispatch decision flow: check service area, filter by skills, then choose the earliest arrival among available technicians or book the next slot in the zone New job, geocoded Inside service area? no Decline / refer yes Filter by skill and parts Anyone free in time? yes no Pick earliest arrival free time + travel time Next slot in that zone offer window to customer
A blended dispatch rule: location decides eligibility first, skills narrow the pool, and arrival time (not raw distance) picks the technician.

Worked example (hypothetical)

The figures below are invented for illustration. A customer reports a gas boiler fault at 10:00. Three technicians are on shift:

TechnicianGas certified?Free fromDrive time to jobEarliest arrival
AYes10:00 (now)22 min10:22
BYes10:308 min10:38
CNo10:00 (now)5 minnot eligible

A pure nearest-available rule picks C, who is closest and free, and who then cannot do the job: a guaranteed second visit. A pure distance rule among qualified staff picks B, who is closest but busy, giving a 10:38 arrival. Ranking qualified technicians by free time plus travel time picks A, arriving at 10:22. If the customer's window allowed 10:30 to 11:00, B might still be the better choice because it keeps A available for the next urgent call. The point is that the decision needs three inputs (skills, availability and location) and that "nearest" means arrival time, not distance on a map.

Honest ETAs

An ETA is a promise, so build it from the same data you dispatch with: the technician's current job end time, a travel-time estimate, and a margin that reflects how often jobs overrun. Update it when the previous job finishes rather than leaving the morning's estimate in place. When a technician is running late, notifying the customer early is better received than a precise ETA that turns out wrong. If you share live technician location with customers, do it only for the active trip and stop sharing on arrival.

Heatmaps of demand

Once every job has a coordinate, aggregate a few months of bookings into a grid and color the cells by job count. Patterns appear quickly: neighborhoods that generate far more work than expected, areas you cover but rarely serve, and seasonal shifts (cooling calls in summer, heating calls in winter, pest calls after particular weather). These maps inform where to base technicians, how to draw zones, and where local marketing might pay off.

A convenient grid is the geohash: a six-character geohash cell is roughly 1.2 km by 0.6 km at the equator (narrower east to west at higher latitudes), coarse enough to hide individual homes while still showing neighborhood patterns. You can see how a coordinate maps to a geohash in the coordinate converter.

Measure first-time fix and travel share

Two metrics connect location data to the bottom line.

First-time-fix rate is the share of jobs resolved on the first visit. Every repeat visit doubles the travel for that job. Location data helps indirectly: skills-based dispatch sends the right person, and job history at a geocoded address tells the technician what equipment was found there last time.

Travel share is travel time divided by travel time plus on-site time. Another hypothetical day for one technician with an 8-hour (480-minute) shift:

Before zoningAfter zoning
Jobs6 × 50 min = 300 min7 × 50 min = 350 min
Drives (incl. start and return)7 × 25 min = 175 min8 × 15 min = 120 min
Total475 min470 min
Travel share175 ÷ 475 = 36.8%120 ÷ 470 = 25.5%

In this made-up example, shorter average drives free enough time for a seventh job within the same shift. Your numbers will differ, but measuring travel share per technician and per zone shows exactly where scheduling or zone design needs work.

Customer location privacy

A list of customer addresses with coordinates and visit times reveals when people are home and what equipment they own. Under data-protection laws such as the EU GDPR, location data is personal data, and it deserves deliberate handling:

  • Minimization: collect the address and coordinate needed to do the job, not continuous customer location. Track technicians during working hours only.
  • Retention: decide how long precise job locations are kept (for example, for the warranty period) and delete or coarsen them afterwards. Aggregated heatmap cells can usually be kept longer than individual points.
  • Access: technicians need today's addresses, not the full customer map. Restrict exports.
  • Third parties: check your geocoder's terms. The public Nominatim service, for instance, asks users not to submit personal data, so customer address geocoding at scale belongs on a contracted or self-hosted service. The tools on this site, such as address to coordinates, are suited to spot checks rather than processing your customer database.

Key takeaways

  • Geocode and confirm the location at booking; it is the cheapest point to fix an address.
  • Check service areas with polygons that follow roads and barriers, not only radius rules.
  • Dispatch on skills and earliest arrival (free time plus travel time), not raw distance.
  • Track first-time-fix rate and travel share per technician and zone to see where location decisions pay off.
  • Treat customer locations as personal data: minimize, set retention periods and restrict access.