Data typed twice
The same order, item or customer entered two or three times, in different programs.
Automated checks
Connections between programs
Custom management systems
The method
Technology
Systems that talk
Warehouse, fleet, ecommerce, logistics: each often has its own system, and data moves between them by hand. We write the custom piece that connects them, so data entered once reaches everyone.
“You stood out from over 60 candidates because you had a clear plan. You explained exactly how you would tackle the problem.” Verified client, supplier selection for a software project · translated from Italian
Programs linked 4 of 5
data entered once
40 minutes a day to retype 80 lines from one program to another, 30 seconds each
160 hours a year, over 240 working days
Once the data is entered, then it reaches the other programs on its own
Reference values for a business retyping 80 lines a day between two programs.
Free check
Tell us which programs you use and where data moves by hand. We find which connection saves you the most time and where to start.
What it costs today
Lines retyped every day from one program to another, times the time each takes. With the connection data is entered once and only the exceptions stay with a person.
A line is an order, an item, a customer or a delivery: anything someone types twice today.
Try it with your numbers
a year of work back to the team
Let us talk about these numbersThe sum: lines a day times seconds per line, over 240 days. After, about 5% stays manual, the exceptions. Retyping errors avoided are not included.
The problem
Each program does its job well. Time is lost when someone has to copy, compare or retype data from one to the other.
The same order, item or customer entered two or three times, in different programs.
Each program has its own figure, and nobody is sure which one is right.
Exports, spreadsheets and imports done by hand to move data from one system to another.
What happens in the warehouse or on the road only shows up when someone updates it by hand.
The part nobody sees
The connection itself is the easy part. The real work is agreeing on who leads, on which codes and on what to do when a figure does not match.
The ERP calls it 10233, the website ST-40, the supplier AB-4471. You need a table saying they are the same thing.
For exampleWith 1,500 items and 3 programs you need 1,500 mapping rows: built once, then updated only for new items.
The ERP decides the price, the warehouse the stock, the customer the address on the website. Every piece of data has an owner, the other programs follow.
For exampleIf two programs can both change price and stock, sooner or later one overwrites the other without anyone noticing.
It is not decided at random: a rule says which one wins, and when it is unclear the line goes to a person, with the reason.
For exampleA customer with two different addresses in two programs lands in an exception queue, instead of receiving the parcel at the wrong place.
API, export, files in a shared folder, database, email: every program allows something. We start from what is there, without changing program.
For exampleAn ERP with no API that saves a CSV every evening is enough to update the other programs every night.
Not everything has to be instant. Stock yes, the price list maybe once a day, customer records when they change.
For exampleUpdating every 5 minutes 20 fields that change once a month is wasted work for every program involved.
Every handoff leaves a trace: what left, what arrived, what stopped and why.
For exampleIf an order never reaches the logistics program, you find it in the log with the time and the reason, without asking three people.
A connection at work
The data is entered once. The connection takes it to the other programs with the right codes, and flags only what cannot get through.
| Time | From | To | Data | Result |
|---|---|---|---|---|
| 09:12 | Ecommerce | ERP | Order 5531 · 2 × ST-40 | Arrived as item 10233 |
| 09:12 | ERP | Logistics | Order 5531 to ship | Ready for pickup |
| 09:40 | Logistics | Fleet | Delivery 41 · 18 parcels | Assigned to van 2 |
| 10:05 | Supplier | ERP | Delivery note with code AB-4471 | Code to link, sent to a person |
| 10:30 | Ecommerce | ERP | Customer with a new address | Differs from the records, to confirm |
| 11:00 | ERP | Ecommerce | Price of item 10233 | Updated on the website |
A close case: an online store received files from several suppliers, each with different codes. The connection made them the same for the catalogue and today stock updates on its own. See the real case
One single system or a connection
If the programs you have work, connecting them often pays off. If one no longer copes, only that one gets replaced.
Everything new, at once
The piece that connects them
Data stays in the company
Data can move from one of your programs to another without a detour through outside services.
Before you write
It makes sense if
You don’t need it if
What clients say
You stood out from over 60 candidates because you had a clear plan. You explained exactly how you would approach the problem. That gave me the confidence to choose you.
Verified review, anonymised for privacy.
Frequently asked
No. We start from the ones you have. If one no longer copes with your work, we look at replacing only that one.
An export, a file, the database or a mailbox is often enough. We start from how data leaves today.
We decide it together during analysis, one piece of data at a time: price, stock, customer records, order status. From there the connection knows who leads and who follows.
The connection does not hide it: it flags it to a person, with the reason.
The data waits in a queue and restarts when the program is back. If something stays stuck too long, a person gets an alert.
Yes, as long as data can get in or out somehow, with APIs, files or the database. We check it program by program at the start.
Yes. We start from the handoff that takes the most time, and the others are added later.
Start with the problem
We will tell you whether connecting them pays off and which handoff to start from.
Prefer to write directly? [email protected]