Skip to content
Case study · Logistics and last mile

US last-mile delivery operator: from WhatsApp voice notes to a working browser extension

KINTO co-developed with a US last-mile delivery operator a dispatch cockpit that reconciles platforms that don’t talk to each other and proposes assignments with a human in the loop, delivered as a working browser extension.

By KINTO · Published

Figures, each with its source

  • about ten

    competitors analyzed

    Measured

Ranges or orders of magnitude where a figure could identify the client, each with its source.

How the transformation unfolded From the problem to what changed, in the five steps every KINTO case follows.
  1. Problem

    Every day, a last-mile operator builds the dispatch sheet by hand (driver, vehicle, route, departure times and load) from several platforms that don’t talk to each other: the marketplace portal, the shift scheduler and the fleet system.

  2. Signals

    Four signals: market, business, user experience and support.

  3. How we understood it

    We synthesized weeks of voice notes and messages between the partners (transcribed locally), mapped each platform section by section with facts verified live, ran competitive and compliance analysis, and turned long meetings into a task board.

  4. What we built

    A browser extension that reads the portal with the user already signed in and proposes assignments with a human in the loop, a sync engine with alerts, and a minimal server that holds no personal data.

  5. Transformation

    From an idea spoken in an afternoon to a working browser extension, with a simulated-station demo to sell without live access to the platforms.

Before and after, by dimension Editorial levels from 0 to 4 read from the text of this case, never a figure drawn to scale.
  • Dispatch sheet

    Before

    Built by hand daily (level 1 of 4)

    After

    Proposed in the pilot (level 2 of 4)

  • Platform sync

    Before

    Platforms don’t talk (level 0 of 4)

    After

    Engine built, access lost (level 2 of 4)

  • From idea to product

    Before

    Idea in voice notes (level 0 of 4)

    After

    Working extension, pilot (level 3 of 4)

The four signals we read What the market, the business, the user experience and support were saying before KINTO built anything.
  • Market

    About ten competitors analyzed on pricing; none integrates the shift scheduler or the fleet.

  • Business

    Monetization is still open (local vs. cloud); the billable value is synchronization without personal data.

  • User experience

    The dispatcher lives in a daily sheet with alerts for absences and replacements.

  • Support

    Drivers get a bilingual English/Spanish panel with history and incidents.

How the idea became a product
  1. Voice notes Transcribed locally
  2. Platform mapping Section by section, verified live
  3. Market and compliance Competitors and rules analyzed
  4. Human-in-the-loop extension Proposes, a person decides

Connections

  • Voice notes → Platform mapping
  • Platform mapping → Market and compliance
  • Market and compliance → Human-in-the-loop extension

KINTO co-developed with a US last-mile delivery operator a dispatch cockpit that reconciles platforms that don’t talk to each other and proposes assignments with a human in the loop, delivered as a working browser extension.

Problem

Every day, a last-mile operator builds the dispatch sheet by hand (driver, vehicle, route, departure times and load) from several platforms that don’t talk to each other: the marketplace portal, the shift scheduler and the fleet system. Market tools store and display, but almost none act, and per-driver pricing punishes an industry with very high turnover.

Signals

  • Market: About ten competitors analyzed on pricing; none integrates the shift scheduler or the fleet.
  • Business: Monetization is still open (local vs. cloud); the billable value is synchronization without personal data.
  • User experience: The dispatcher lives in a daily sheet with alerts for absences and replacements.
  • Support: Drivers get a bilingual English/Spanish panel with history and incidents.

How we understood it

  • Synthesis of weeks of voice notes and messages between the partners, transcribed locally.
  • Section-by-section mapping of the platforms, with facts verified live.
  • Competitive and compliance analysis.
  • Long meetings with the operator turned into a task board.

What we built and why

  • Browser extension that reads the portal with the user already signed in, reconciles against the shifts, and proposes or writes assignments, always with a human in the loop.
  • Sync engine with alerts and tiered assignment, also operable by a remote agent through an agent contract (dry run by default).
  • Minimal server with no personal data: per-company licenses, pseudonymized backup and a protected license panel.
  • Compliance by design: no evasion techniques, no acting on the marketplace’s behalf, and minimized driver data.

Why this way: Compliance was treated as design: no evasion techniques, no acting on the marketplace’s behalf, and minimal driver data, so the product is sellable and does not depend on a shortcut.

Transformation

  • From an idea spoken in an afternoon to a working browser extension.

  • A simulated-station demo to sell without live access to the platforms.

  • competitors analyzed: about ten (measured)

What we don’t know yet

  • Business results (paying customers, dispatch hours saved): no evidence.
  • The pilot lost access to the operator’s platforms and monetization is undecided. We publish it as a pilot, not as a results case.
Case studies

Related case studies

Have a problem like this?

Tell KINTO what is going on. The first conversation is to understand it, not to sell.