TALK TO most forwarding and broking offices and you'll find they aren't short on software. They've got a freight system, a customs system, an accounting package, a shared inbox and a spreadsheet or two holding it all together. What they're short on is time. And when you dig into where that time goes, a lot of it is the same job done twice. Sometimes three times.
A consignment comes in. Someone keys the details into the booking. Someone keys them again to build the entry.
Someone keys them again when it's time to invoice. Each of those is a person reading one screen and typing into another. Each one is a chance for a digit to slip, a weight to be rounded or a reference to be missed. A small typo on a booking is a nuisance.
The same typo on an entry can mean a hold-up at the border and a customer on the phone wanting to know why.
So, before you sit through another demo of the next big platform, I'd suggest something cheaper. Count the rekeys.
Pick a normal job. Not the nightmare one, just an ordinary import. Then walk it through the office from the first email to the final invoice and write down every time someone types something that already exists somewhere else.
Do it with the people who do the work, not from the org chart. The process on paper and the process at 4pm on a Friday are rarely the same thing. You'll find the workarounds: the spreadsheet that tracks containers because the main system doesn't show it the way the team needs, the email that gets forwarded to three people so one of them can copy the reference into a different screen.
The count usually surprises people. Not because anyone is doing anything wrong, but because each step made sense when it was added. Nobody ever stood back and looked at them all in a row.
It tends to cluster in a few places.
The handover from sales or customer service to operations is a big one. A quote gets agreed, then the booking gets built from scratch from the email instead of from the quote. The jump from operations to the entry is another. Commercial invoice details, tariff classifications and supplier information get read off a PDF and typed in by hand. Then there's invoicing. Charges that were known at booking get re-entered at the end, plus whatever changed along the way.
That's usually where the disputes start, because the change was agreed on a phone call and never made it back into the file.
None of these are dramatic on their own. Added up across a few hundred jobs a month, they're hours of skilled people doing data entry instead of the work you actually pay them for.
Here's the part people don't expect from someone who builds automation for a living. A lot of the fix needs no new software at all. Agree on one place where each piece of information lives, and make everyone pull from there. Kill the spreadsheet that duplicates the main system or make it the main system for that job. Decide who owns a change when a customer rings to move a date or add a charge, and make sure it gets written down once, in the file, not on a sticky note.
Cut steps before you speed them up. If you automate a messy process, you just get a faster mess. Then automate the repeat steps, and keep a person on the risky ones.
Once the process is clean, the repetitive bits are good candidates to hand off. Pulling details from a commercial invoice into a draft. Building the booking from the agreed quote. Carrying charges through to the invoice so nobody is required to retype them. Flagging a job where the numbers don't match between documents.
What shouldn't be handed off is judgement. Classification calls, valuation questions and anything lodged under a licensed broker's name still need a licensed person to check and sign off. Good automation does the typing and prepares the draft.
The broker still makes the call. That's not a limitation. It's the right way to split the work, and it means your most experienced people spend their day on the decisions only they can make.
You don't need a transformation program. Pick the one handover that annoys your team the most, count the rekeys in it, fix the process, then see what's left to automate. Do one, learn from it and move to the next.
The next platform might be the right move for your business. But you'll choose it far better once you know exactly how many times your team is typing the same consignment, and why.