Solutions · Transport admin

Stop being the routing layer.

You are the reason the programme works, and that is exactly the problem: the queue, the policy and the history of every decision live in your head and your inbox. This page is about moving them somewhere a colleague could work from on a Tuesday you are not there.

The job today

Four things that are true of almost every transport desk.

If none of these describe yours, you are running a better programme than most and the rest of this page is probably not for you.

  • The inbox is the system

    Requests arrive as email, WhatsApp and phone calls, in no format, at any hour. The queue exists only in your head, and the only person who knows the status of anything is you — which is why you cannot take leave.

  • Every request is a negotiation

    Is this person eligible? Is a sedan enough? Who approves a 2am airport run? Each answer is reasoned out again from scratch because the policy lives in a PDF rather than in the request form.

  • Allocation is phone work

    Finding a vehicle means calling operators until one says yes. When the first choice declines at 11pm, the fallback is whoever picks up, at whatever rate they quote.

  • Nobody knows why

    Three months later someone asks why a particular trip cost what it did, or why an employee was refused. The reasoning was real and it was never written down.

What changes

Six things that move work off your desk.

Not dashboards. Each of these either removes a decision you were making manually or records one you were making from memory.

One queue, every trip type

Rostered transport, ad-hoc cabs, airport transfers and rentals land in the same board with the same shape — requester, cost centre, policy status, vehicle class, price basis. You work a list, not an inbox.

Policy checked before it reaches you

Eligibility, vehicle class, approval threshold and advance-notice rules are evaluated when the request is made. What arrives at your desk is already compliant or already flagged, with the specific rule named.

Auto-allocation with fallback

Trips allocate against your contracted supply by rate, proximity and acceptance history. A decline cascades to the next operator automatically instead of waiting for you to notice and start calling.

Deviation alerts while it matters

A vehicle that has not moved, a route that has diverged, a pickup running late against its window — raised while the trip is live and there is still something to do about it, not in tomorrow’s report.

Exceptions, with the reason attached

When you override policy — and you will — the override records who approved it and why. The next person to ask about that trip gets the answer without asking you.

A desk someone else can cover

Because the queue, the policy and the decision history are in the system rather than in your memory, a colleague can work it for a week. This is the part that is hardest to put in a feature list and the one most admins care about.

How a request runs

Four steps, and you appear in one of them.

The exceptions are where you belong. The volume is not.

  1. 1

    The request arrives structured.

    Employee app, web, or raised by you on someone’s behalf. Cost centre, purpose, pickup, drop and timing are captured as fields rather than as prose you have to interpret.

  2. 2

    Policy runs, not you.

    Eligible and within threshold, it proceeds. Outside policy, it routes to the named approver with the breach spelled out. Either way the decision is logged against the request.

  3. 3

    Supply is found automatically.

    Allocation goes to contracted supply at the contracted rate, with fallback on decline. You see the ones that could not be filled, which is a far shorter list than the ones that need filling.

  4. 4

    The trip reports on itself.

    GPS trail, actual times, deviations, the driver and vehicle that ran it. Billing is computed from what happened rather than from what the operator submits.

What you can report on

The numbers that make your case for you.

Most transport desks cannot evidence their own workload, which is why the headcount conversation never goes well. These are produced from trip data rather than estimated.

Auto-allocated share

The proportion of trips placed without you touching them. The single number that describes how much of your week the desk is taking.

Time to allocate

Request to confirmed vehicle. Worth watching by hour of day, because the 11pm number is the one that hurts.

Vendor acceptance rate

Who actually takes the trips they are offered, by city and vehicle class. The evidence behind a supply conversation.

Exceptions per week

Overrides, by rule. A rule that is overridden constantly is usually a wrong rule rather than a discipline problem.

SLA breaches

Against the window promised to the employee, measured from the GPS trail rather than from a driver’s report.

Unfilled requests

Where supply ran out, by city, time and class. The gap list that tells you which contract to renegotiate first.

What stays with you

Four things automation should not touch.

A system that claimed these would be a system that had not run a transport desk.

  • The genuine exceptions

    A medical emergency at 3am, a VIP movement, a site evacuation. Automation handles the volume so that your attention is available for the handful that need judgement — it does not pretend to have judgement.

  • Vendor relationships

    Scorecards make performance arguable with evidence. The call about whether a long-standing operator gets another quarter is yours, and it usually involves things no scorecard holds.

  • What the policy should be

    If the approval threshold is wrong, the system will enforce the wrong threshold faithfully and show you how often it is overridden. Changing it is a conversation with HR and finance.

  • Local knowledge

    Which gate to use at a particular park after 9pm, which pickup point floods in monsoon. The system records what you tell it; it did not know in the first place.

Questions

From transport and admin teams

Still unanswered? Our solutions team replies within one business day.

Talk to the team →

We run four vendors and none of them use software. Does this still work?

Yes, and that is the common case. Operators get the Partner Suite — a dispatch board, driver and document compliance, and auto-generated invoices — which most adopt because it removes their own paperwork rather than adding to it. Where an operator genuinely will not use an app, allocations can still reach them by SMS and WhatsApp with confirmation back, so the trip is tracked even if the vendor is not on the platform.

How long before the desk actually gets quieter?

The honest answer is that it depends on where your volume comes from. Rostered transport improves immediately, because routing and allocation were always the mechanical part. Ad-hoc requests improve as employees move to the app, which takes a few weeks of nudging and is the part that needs you. The slowest piece is usually vendor adoption, and that tracks how much of their volume you represent.

What happens when auto-allocation cannot fill a trip?

It surfaces as an unfilled request with the reason — no vehicle of that class in that city at that hour, or every contracted operator declined. You then place it manually, and that manual placement is recorded like any other. The system does not silently substitute a different vehicle class or a different rate to make a trip look filled.

Can I override anything?

Yes. Overrides are a feature, not a loophole: policy that cannot be overridden gets worked around outside the system, which is worse. Every override names the approver and the reason and appears in the exception report, which is what makes the pattern visible.

Does the employee app reduce calls or just add a channel?

It reduces them where it answers the question people were calling to ask, which is overwhelmingly "where is my cab". Live ETA and driver identity remove most of that traffic. Requests still arrive by phone from people who will always phone, and the app lets you log those as structured requests in a few seconds so the record stays complete.

Who else can see the queue?

Access is role-scoped. A site admin sees their site, a regional lead sees theirs, finance sees cost and policy status without personal detail they do not need. The audit log records who viewed and changed what, which is also how a shared desk stays accountable.

See your own queue

Bring a week of requests, however they arrived.

Email threads, a WhatsApp export, a spreadsheet. We will show you what that week looks like as a queue with policy applied and how much of it would have allocated itself.

  • A real week of your requests, restructured
  • How much would have auto-allocated
  • What would have reached you, and why

Request a demo

Three details, and we'll call you to arrange a time.

We'll only use these details to contact you about taSki. No newsletter, no list sharing. See our privacy policy.