01
Decision audit
Two weeks inside your company. We map the twenty decisions that move the numbers, find where the data for each one lives, and show which ones are being made blind.
- Decision map
- Data inventory
- Priority list
2 weeks
AI consulting for operations, finance and sales teams
Turn your data into decisions.
Aura builds the models, the dashboards and the weekly rhythm that let your team decide with numbers instead of instinct. Sarajevo based, working across the region.
Aura PulseLast 30 days · live
Revenue
€2.41M
+12.4%
Late orders
3.1%
-31%
Forecast accuracy
94%
+6 pts
Decisions this week
18
4 open
Forecast vs actual
Orders per day
Decision queue
Move 14% of stock from Tuzla to Mostar
Confidence 91%
Delay the March price change by 9 days
Confidence 84%
Hire two drivers for the Zenica route
Confidence 78%
0
Companies advised since 2019
0.0×
Median return on the first year
0 days
From kick-off to a working model
Trusted by teams in Sarajevo, Zagreb and Vienna
- Bosna Logistics
- Klas
- Violeta
- Sparkasse
- BH Telecom
- Mistral
What we do
Three ways we make your data useful.
Most companies do not need more data. They need the twenty decisions that matter to be made from the data they already have, every week, by someone who owns them.
02
Models and dashboards
Forecasts, anomaly alerts, pricing and stock models, built on your own data and shipped as a dashboard the board actually opens on Monday.
- Forecast model
- Live dashboard
- Alert rules
6 to 10 weeks
03
Decision rhythm
The weekly meeting, the owner for every metric and the playbook for what happens when a number turns red. We run it with you for a quarter, then hand it over.
- Weekly review
- Metric owners
- Playbooks
1 quarter
Case study
How Bosna Logistics cut late deliveries by a third in one season.
Their dispatchers were planning 1,800 routes a week from memory and a spreadsheet. We built a delay model on three years of GPS and weather data, put it inside the planning screen, and changed the Monday meeting to start from the model instead of the complaints list.
- 0%Fewer late deliveries
- €0kEuro saved in penalties, first year
- 0 minAverage planning time per route, down from 26
- 0 weeksFrom kick-off to full rollout
“The model does not plan for us. It tells us which routes to look at first, and that alone gave the dispatchers their afternoons back.”
How it works
Listen, model, run. Then we leave.
1
Weeks 1 to 2
Listen
We sit with the people who make the decisions and the people who own the data. No slides. We leave with a decision map and a list of the numbers nobody trusts yet.
2
Weeks 3 to 10
Model
We build the first model on your real data in eleven days, show it to the team, and improve it every week until they stop asking whether it is right.
3
One quarter
Run
We run the weekly review with you, put an owner behind every metric, and step out once the rhythm holds without us in the room.
Clients
What stayed after we left.
We measure ourselves on whether the Monday review still runs a year later. So far it does, in every company we have worked with.
“We had dashboards before. What we did not have was a Monday morning where everyone looked at the same number and agreed what to do about it.”
Lejla KarićCFO, Klas“Aura was the first partner that asked to see our decisions before asking to see our data. The forecast paid for the engagement in the second month.”
Adnan SelimovićCEO, Mistral“The model is good. The habit they left behind is better. Two years later the review still runs every Monday at nine.”
Tomislav MarićCOO, VioletaStart with a call
Bring one decision. We will show you the data behind it.
Forty minutes, one decision you make every week, and a first look at what your own numbers say about it. No deck, no commitment.
09:41●●●
Aura Pulse
Alert · Thursday
Late-order risk on the Mostar route is up 18% for Thursday.
Options
- Reassign 3 loads to Wednesday91%
- Add one driver from Tuzla76%