AI automation for logistics and warehouse companies
One late delivery starts a chain. The customer asks, the dispatcher calls the driver, the driver gives a status from memory, the dispatcher calls back. We pull data from the transport system, the warehouse and telematics into one place, so the answer is ready before anyone picks up the phone.
- 14 daysto the first working automation, counted from the end of the workshop
- EUR 3,500net per week of workshop, a published price we do not negotiate
- 10 to 25%how much a routing algorithm cuts fleet mileage at the same delivery volume
The problem
Why does a dispatcher spend half the day on the phone?
A European road haulier runs three kinds of traffic at the same time: domestic, transit and cross-border. Each one carries its own paperwork, its own filing deadlines and its own way of proving what happened, and all three land on the same few people in the back office. The administrative load doubles while the headcount stays flat.
Planning routes for a few dozen vehicles a day is a computation, not a matter of experience. A person solves it below the optimum every time, because nobody runs several hundred variants in their head before the shift starts.
On top of that sits a stream of data nobody asked for: the Mobility Package, where the driver return and the vehicle return run on separate clocks, second generation smart tachographs, electronic filings for sensitive goods and electronic tolling. Each of these demands compliance and none of them talks to the others.
- Shipment status comes from a call to the driver, so it is as accurate as his memory
- The consignment note and the proof of delivery are filled in by hand after the drop, and invoicing waits for the paper to come back
- Service, insurance and tachograph dates live in a spreadsheet, so sometimes the roadside check is the reminder
- Nobody knows what a single run actually earned until the month is closed
Why now
What changed that makes this pay off today
Transport systems, telematics and warehouse software now expose interfaces that three years ago either did not exist or cost as much as a separate rollout. Connecting them is counted in working days, not quarters. The connector catalogues of the automation platforms say the same thing: what used to be a custom project is now configuration.
The second change matters more. A rule will copy a consignment note number from one system into another, but it will not answer the question "which runs last week came in below the profitability threshold, and why". A language model reads the transport system, the fuel data and the settlements at once and answers in a sentence.
The third one is commercial. At current freight rates and fuel costs, ten percent on fleet mileage stopped being an optimisation and became the difference between a run worth taking and a run worth passing on.
What changes
What can you see after the first month?
Status without calling the driver
Position and delivery stage come together from telematics and the transport system. The customer gets an answer and the dispatcher does not stop planning to give it.
The phone call to the driver stops being the source of status
Documents ready when the drop is confirmed
The consignment note, the proof of delivery and the basis for the invoice are built from data that is already in the system, instead of being copied off a sheet of paper.
Invoicing no longer waits for paper coming back from the route
Routes computed, not arranged
The plan comes out of the constraints: time windows, payload, driver hours. The dispatcher corrects it rather than building it from scratch.
10 to 25% less mileage at the same delivery volume
Compliance watched by the system
Driver and vehicle returns, tachograph validity, servicing, insurance and filings for sensitive goods all have dates, and the dates remind you about themselves.
The deadline speaks up before the check, not during it
Run profitability as it happens
Revenue, fuel, tolls and driver hours lined up per run, so you can see which jobs are worth taking before the month is closed.
Rate decisions made on data instead of instinct
Proof
Do you use this yourselves?
Yes, and it is the only proof we can show without asking a client for permission to publish their figures. Two AI agents running in production inside our founders' own companies, both launched in March 2026 and built by the method described above. Neither company is a haulier, so read this as evidence about the method, not about your fleet. Below is six months of measurement, together with what these numbers do NOT prove.
Built for ourselves. These are not client deployments but implementations inside the founders' own companies. We say so plainly, because proof you cannot verify is not proof.
JustBottheir agentJustIdea Agency, a performance marketing agency of around 30 people
- 23
- people from the company work with the agent
- 38
- systems wired in permanently
- 112
- processes run without a human
- 475
- thousand lines of the agent’s code
An agent wired into Google Ads, Meta, the CRM, the task system, the SEO tool, Analytics, Search Console and the data warehouse. First commit 1 March 2026, today 23 people use it.
8 of 38 systems it works in
Google Ads
Meta Ads- HubSpot
- ActiveCollab
- Ahrefs
GA4
Search Console
BigQuery
- Before
- In the twelve months before launch the agency ran at a margin of minus 4.42 percent. Across a twenty-month window there was not a single run of three profitable months.
- After
- May, June and July 2026 are three profitable months in a row, the first such run in the whole window. Revenue barely moved, so the change sits on the cost of delivery, not on sales.
Margin up 5.55 percentage points year on year, with revenue up 1.5 percent
SurfBottheir agentSurf Inc., a premium clothing brand with an online shop and two physical stores
- 16
- people from the company work with the agent
- 21
- systems wired in permanently
- 89
- processes run without a human
- 323
- thousand lines of the agent’s code
An agent wired into the order system, the ERP, the shop, the data warehouse and the ad platforms. Launched in March 2026, fully running by May. Sixteen people use it, warehouse and store staff included.
8 of 21 systems it works in
PrestaShop- BaseLinker
- Comarch Optima
- BigQuery
- Google Ads
- Meta Ads
- GA4
Zalando
- Before
- From February to August 2025 every unit of ad spend came back as 3.53 in online revenue, and advertising absorbed 28.3 percent of that revenue.
- After
- In the same months of 2026 the ratio is 4.19 and advertising is down to 23.9 percent of revenue. The ad budget is 6.9 percent lower with sales 10.3 percent higher.
Return on ad spend from 3.53 to 4.19, 18.5 percent higher, on a budget 6.9 percent lower
2026, with the agent 2025, before the agent revenue per unit of spend
Return on every unit of ad spend at the clothing brand, February to August, 2026 against 2025. The 2026 curve starts LOWER than the year before, breaks away in March and holds the lead for the rest of the season. The peak is 4.91 in May. We point out the weaker start because it shows we did not pick a convenient baseline.
Measurement: Online revenue from the order system against spend across every ad platform, the same months in both years. Cost data in the warehouse starts in February 2025, so we cannot rule out that the ratio was already rising on its own.
Operations run by the agency agent in successive months of 2026, from the first commit in March. From 83 operations in March to almost 63 thousand in August, with users growing from 4 to 23. This curve is here because a deployment nobody uses looks identical in a report to a successful one.
Measurement: Agent call log in the data warehouse, March to August 2026, counted at the level of a single tool invocation.
Breakdown of 29,854 operations run without a human over thirty days, by kind of task. Most of it goes to tracking team availability and monitoring client sites: things that have to run without a break and that nobody wants to do by hand.
Measurement: Agent audit log, the 30 days before 2 September 2026, counted by the kind of tool invoked.
Of 64,903 operations run in the last thirty days, 29,854 went through without a human: client site monitoring, warehouse loading, ad account reviews, catching form submissions. The rest is work someone requested and reviewed. We give both numbers, because autonomy without oversight is not a feature but a risk.
Measurement: Agent audit log, the 30 days before 2 September 2026, split by whether a call had an associated human session.
What about us
What if it works differently here?
Four sentences come up in transport companies more often than any others. We answer them here the way we would answer them in the room.
We have a transport system and we are not replacing it.
And there is no need to. We read data out of it through its interface and leave it where it is. Replacing a TMS in a haulage business is a year of work, and here we are talking about 14 days to the first effect.
Our drivers will not key data into an app.
That is why we start with the data that appears on its own: telematics, fuel card transactions, tolling records, scanned documents. We add manual entry only where the process cannot close without it, and we keep it as short as it can be.
Our fleet is too small for this to pay off.
With a small fleet the money sits somewhere other than with a large one: not in route optimisation, but in documents and in answering customers. At the workshop we say which of those effects is real for you, and if none of them is, we say that too.
Our rate and customer data is sensitive.
We work on your infrastructure or on a cloud you name, and the scope of access is agreed at the workshop and written down. Telematics positions and tachograph records are personal data about your drivers, so <a href="https://eur-lex.europa.eu/eli/reg/2016/679/oj" rel="noopener">the GDPR</a> applies to them and we handle them on that basis. Nothing goes beyond what was signed.
- Transport systems
- TMS, forwarding systems, freight exchanges, customer portals
- Telematics and fleet
- GPS, digital tachographs, fuel cards, electronic tolling
- Warehouse
- WMS, BaseLinker, scanning systems, dock booking
- Documents and settlements
- ERP and accounting systems, invoicing tools, document workflow, customs filings
- How the team works
- Microsoft 365, Google Workspace, Slack, task systems
- Automation layer
- Make, n8n, custom integrations over an API
What exactly we automate in a transport company
A typical split of the work in a haulage or forwarding business before the rollout and after it.
| Area | Today | After the rollout |
|---|---|---|
| Shipment status | a call to the driver, status from memory | position and stage from telematics, available to the customer |
| Transport documents | filled in by hand after the drop | generated from data as soon as delivery is confirmed |
| Route planning | arranged by the dispatcher | computed from constraints, corrected by the dispatcher |
| Fleet deadlines | a spreadsheet of services and insurance | a reminder before the date, not after it |
| Tolls and filings | handled separately on every load | filing and settlement inside the same flow |
| Run profitability | visible once the month is closed | revenue against fuel and tolls as it happens |
The scope of the first rollout is agreed at the workshop. We take the three items with the highest return, not all of them at once.
How we work
Four steps from the phone call to the system
Every step ends with something you can judge before you decide on the next one.
Workshop on your site
One week. We walk the process from the order, through planning and the run, to the documents and the settlement. We come out with a map and a list of the places where data goes missing or gets keyed in twice.
Picking three automations
From the map we take the three with the highest return and we count them before we start, so that after 30 days it is clear whether they worked.
Rolling out the first one
14 days to a working solution. In transport we usually start with shipment status, because that is what takes the most calls off the dispatcher.
Measurement, then the next one
After 30 and 90 days we check the metrics agreed at the workshop. Only then do we decide on the next area.
Before and after
How a dispatcher's day changes
Before
- Half the day on calls: the customer asks for a status, the driver gives it from memory
- Routes arranged in the morning, with no calculation of time windows and driver hours
- Documents coming back from the route days later, and settlement waiting for them
- Servicing or a tachograph reminds you about itself during a roadside check
After
- The customer sees the status without asking, and the dispatcher takes only the exceptions
- The plan is ready before the shift, to be corrected rather than built
- The basis for the invoice appears the moment delivery is confirmed
- The deadline reports itself early, so the vehicle does not stop unexpectedly
What changes is who does the watching: the system instead of a person. The number of things to watch stays the same.
For the owner and the transport manager
What the person signing this gets out of it
The first thing shows straight away. The dispatcher stops being a switchboard and goes back to planning, which is the work you are paying for.
The second one is quieter and matters more when you sell the company or sit down with a bank. The profitability of a single run stops being one person's knowledge and becomes a number in a system that you can put on the table.
What it costs
The workshop price is public, the rollout price comes after it
The workshop costs EUR 3,500 net per week and we publish that number before the first call, because a company has the right to know what it is agreeing to. It runs from 1 to 7 weeks, and most transport companies need one to three. We quote the rollout after the workshop, once we know how many systems we are connecting to and what state they are in.
Process workshop
EUR 3,500 net
per week, usually one to three
A closed piece of work. It ends with a process map and a 30 day plan you can run without us.
- A week of work inside your company and inside your systems
- A process map from the order to the settlement
- The three automations with the highest return, with the hours counted
- Metrics to measure after 30 and 90 days
- Most popular
Workshop and first rollout
quoted after the workshop
For a company that wants shipment status working within two weeks of the workshop ending.
- Everything in the workshop
- The first automation rolled out in 14 days
- A connection to the transport system and to telematics
- Training for the person who runs it on your side
- Measurement after 30 and 90 days
AI agent for logistics
quoted after the workshop
For a company where questions about shipments, the fleet and profitability come up a dozen times a day.
- Everything in the rollout
- BOND, the company AI agent, answering from your own data in 5 to 10 seconds
- Alerts on fleet deadlines and on runs below the threshold
- Access for dispatchers, customer service and the board
If after the workshop we conclude there is nothing here worth automating, we say so plainly and we do not propose a rollout. It has happened before and it will happen again.
Frequently asked questions
What is logistics automation?
It is a system taking over the work a person does today: establishing where a shipment is, filling in transport documents, watching fleet deadlines and putting the cost of a run together. The data needed for all of that is usually already in the company, it just sits in several systems at once. Automation is about pulling it into one place, not about buying new software.
Do we have to replace our transport system?
No. We read data out of it through its interface and leave it where it is. Replacing a transport system in a haulage business is a year of work, and the first working automation is meant to exist within 14 days of the workshop ending.
How does automation help with the Mobility Package, tolls and filings for sensitive goods?
All three produce deadlines and submissions that nobody wants to track by hand: driver and vehicle returns to base, a filing for every load of sensitive goods, toll settlement. We put them into one flow with reminders, so the deadline reports itself before a check rather than during one.
How much does fleet mileage actually drop?
Route planning algorithms cut mileage by 10 to 25% at the same delivery volume. The upper end applies to companies that plan by hand and have many stops on a route. On simple line runs the gain is smaller, and we say so at the workshop instead of promising everyone the maximum.
Will this work with a fleet of a few vehicles?
It will, but the money sits somewhere else. With a small fleet the saving on routes is minor, and the real effect comes from automatic status for the customer and documents that build themselves. At the workshop we say which of those two is worth paying for in your case.
How long does it take from signing to the first effect?
One week of workshop and 14 days for the first rollout, counted from the end of it. We measure the effect after 30 and 90 days, on the metrics agreed before we start. We run at most two rollouts at a time, so the dates hold.
Start with one process
On a free 30-minute consultation we walk through one of your processes, usually handling a status enquiry, and tell you whether we see anything in it worth automating. If we do not, we say so on the call.
We run two workshops a month, because a week inside a company is a week of one person's time on our side.
Book a free consultationPS. If you would rather see what a workshop looks like before anything else, we described it step by step on the product page.