Free Diagnostic
90 minutes. Three opportunities. Zero cost.
Structured conversation with your leadership. We leave with a short report showing where AI and automation can free margin, time or risk in your operation. In 5 business days.
Free test · instant result
How mature is your company's data & AI?
Ten questions, two minutes. No email, no sign-up — the result appears right here, with recommendations for your level.
Inside the diagnostic
What we do, with whom, and what each area takes away.
The diagnostic is a structured session of around 90 minutes with management and the heads of the areas involved. It gathers, in one guided conversation, three kinds of information: the company's strategic priorities, each area's concrete difficulties, and the state of the systems and data that support the operation.
Praxis has its own diagnostic support tool, DataVault, which can connect to one or two systems chosen by the company — read-only — and objectively measure the state of the information: missing fields, duplicate records, inconsistencies between sources. This step is entirely optional. It is natural that, before a contract exists, a company prefers not to open its systems; the session works fully without it, and the analysis can wait for a later stage. When it happens, it replaces perceptions with facts and makes the report considerably more precise.
What we discuss with each stakeholder
The preparation required is minimal: no report or prior survey — just come with the area's processes and difficulties in mind and, if available, some data that helps quantify.
Finance / Controlling
What they bring
Knowledge of closing, reconciliation and reporting processes, and where rework originates.
What we discuss
The time and cost of manual work in closing and reconciliations; the reliability of the numbers behind decisions; how easy it is to get margin by product, customer or asset.
What they take
A first reading of what today's friction costs and where automation and better data quality free up time and reduce error.
Operations
What they bring
How operational processes really work and where information arrives late, incomplete or not at all.
What we discuss
The efficiency of critical processes, planning and maintenance; the availability and reliability of operational data; the dependence on specific people to produce or interpret it.
What they take
The points where the operation loses efficiency for lack of timely, reliable data.
Commercial
What they bring
Knowledge of the sales cycle, invoicing and the customer relationship, and where customer data is scattered.
What we discuss
The quality and consolidation of customer data; demand forecasting and pricing; the points in the commercial process where decisions lack a solid basis.
What they take
Where better data improves forecasting, reduces invoicing errors and supports the commercial relationship.
IT / Data
What they bring
The map of systems, existing integrations, the data architecture and the history of initiatives.
What we discuss
How information flows between systems and where it duplicates; the state of the data, from raw to decision-ready; access rules and governance; what sits in the backlog and why.
What they take
An external, objective reading of the state of the data and a prioritisation that helps justify investment to management.
The themes that run through the conversation
Process efficiency and cost
Much of the hidden cost sits in manual work and rework — and much of that rework exists because the data cannot be trusted: missing, duplicated or inconsistent information that forces constant checks. By quantifying this effort, the diagnostic links data quality to cost reduction, which is the language management cares about.
Characterising the data
We distinguish three levels: raw data, as it arrives from source systems; data already cleaned and consolidated, with business rules applied; and data ready to serve reports and decisions. Knowing where the company sits on this scale shows where to invest first — the right sequence starts with the foundation.
Proper use of data
Who accesses what, whether sensitive information is protected, whether decisions rest on the correct, up-to-date source, and whether the company meets its obligations. Misuse of reliable data is as real a risk as poor data quality — and usually far less visible.
What the company gains
- 1
A reading leadership can use
The report describes problems in terms of money, time and risk, not technical language — management decides with its own criteria.
- 2
A view across areas
By crossing the conversations, we often find difficulties that different areas describe in their own way but that share a common origin. Seeing it exposed by an outsider helps align teams.
- 3
A sequenced plan
Initiatives are organised in phases, so early results sustain the next steps. Each phase has owners, prerequisites and indicators.
- 4
Clarity on what not to do
As important as the recommendations is what to avoid. Sparing the company an investment that would not pay off is often worth more than any project.
- 5
A measured starting point
The figures established become the reference against which progress is measured. Without an agreed baseline, improvement is hard to demonstrate.
What you get
A short, dense, actionable report.
This is how the diagnostic output arrives in your inbox within 5 business days. Content below is fictitious but represents the actual format.
What you get.
- 90-minute session, online or in person in Lisbon/Porto.
- 3 concrete AI & automation opportunities identified in your operation.
- Impact estimate (hours/month or margin) and expected payback.
- One recommended lighthouse project to start — if it makes sense.
Industrial Diagnostic · Praxis
AI & Automation opportunity report
Indústria Norte SA · Automotive components · 320 employees
Executive summary
Three concrete AI and automation opportunities identified in current operations, with combined potential to free ~280 hours/month and average 6-month ROI. Recommend starting with opportunity #1.
Opportunity 01
Intelligent invoice automation
Pain point
2400 invoices/month processed manually. 4–6h/day of finance team typing. Posting errors and lost early-payment discounts.
Solution
Centralized intake + OCR/LLM extraction + 3-way match + ERP integration. Exception workflow for humans.
Impact
~120
h/month freed
Payback
4–6 months
Effort
▌▌ Médio
Opportunity 02
Maintenance copilot
Pain point
MTTR of 38min. Knowledge concentrated in 2 senior technicians. High risk if one leaves.
Solution
AI agent with RAG over manuals + CMMS history. Tablet/voice access on shop floor.
Impact
-45%
MTTR (estimated)
Payback
6–9 months
Effort
▌▌▌ Alto
Opportunity 03
Predictive stock analysis
Pain point
Stockouts on critical SKUs. 15% excess on slow movers. Working capital poorly allocated.
Solution
Demand prediction model + alerts + WMS integration. Weekly automatic reorder point adjustment.
Impact
-20%
stock capital
Payback
3–5 months
Effort
▌ Baixo
Recommendation
Start with invoice automation — fastest payback, manageable effort, validates architecture for following flows. Next step: 2h workshop with Finance + IT to scope and define KPIs.
Book your diagnostic.
Reply within 24 business hours. Confidential.