The scenario
A commercial director wants revenue by segment for last quarter, excluding intercompany, in the same definition the board uses. Today that is a ticket, a spreadsheet and a two-day wait.
DataBot demo
Ask the enterprise a question, get the answer, the source and the query
A working conversational analytics session on your own schema: Arabic or English in, a governed number out, with the SQL and the source system attached to every answer.
In Saudi Arabia
What turns the DataBot demo into a production system
A working demo and a system a Saudi reviewer will sign off are two different things. These are the questions asked before go-live, and what we hand over against each one.
AuthorityWhat it requiresWhat we buildWhat you keep
SDAIA, PDPL
Lawful basis for every personal-data field the demo touches, before it runs on real records.
Field-level data map produced during scoping, sanitised extract used until that map is signed off.
Data map and lawful-basis note the client's DPO can file before production access is granted.
NDMO
Data classification and quality controls on the sources the demo would depend on in production.
Source-by-source classification, with quality thresholds agreed before the pilot widens.
Classification map and a quality gate report your data office keeps.
NCA ECC
Access control, logging and read/write boundaries clear before wider rollout.
Entitlements scoped per role, every action logged, write paths gated behind approval.
Control matrix and an action log your risk function can replay.
Where it runs
The demo itself can run on a sanitised or representative extract, but where the data cannot leave a system, the build happens inside your environment inside Saudi Arabia, not ours.
Arabic, as it is spoken here
Arabic questions and documents are handled as first-class input, not translated to English before processing, including organisational and Gulf-dialect terms.
Who it is for
Finance, commercial and operations leaders who wait days for a number that a analyst has to rebuild each time.
Sample workflow
The chain of work behind the answer
1
Question and identity
- Arabic or English question
- User identity and entitlements resolved
- Row and column scope applied
2
Semantic resolution
- Business terms mapped to governed metrics
- Ambiguity returned as a clarifying question
- No metric invented outside the catalogue
3
Query generation
- SQL generated against the approved model
- Validated and cost-checked before execution
- Read-only execution path
4
Answer with evidence
- Number, trend and short explanation
- Source system and SQL shown
- One-click export or chart
5
Log and learn
- Every question and query recorded
- Unanswered questions become backlog
- Accuracy measured on a test set
Answers are graded against a question set your team wrote, so accuracy is a number rather than an impression.
What you will see
Two examples, start to finish
“Revenue by segment last quarter, excluding intercompany.”
- Behind it
- Entitlement check → segment and revenue resolved to catalogue metrics → intercompany filter applied from the governed definition → SQL run read-only on the warehouse.
- What comes back
- A segment table with quarter-on-quarter movement, the definition used, the source tables and the exact SQL, exportable to Excel.
«كم عدد الطلبات المتأخرة عن الوعد التعاقدي هذا الشهر؟»
- Behind it
- Arabic parsed without translation to English first → “late against contractual promise” resolved to the SLA metric → order and SLA tables joined.
- What comes back
- Count, breakdown by cause and the ten worst orders, each linked back to the record in the source system.
Deliverables
What you keep after the session
Every item below is an artefact your team can reuse, whether or not the engagement continues.
Working DataBot on your schema
Connected to one governed subject area in your environment, not a sample database.
Semantic layer extract
The metric definitions, joins and filters we resolved, as a reusable artefact.
Accuracy report
Scored results against your own question set, with every failure explained.
Governance notes
How entitlements, logging and read-only access were enforced during the demo.
Rollout sequence
The subject areas to add next, in the order that reduces analyst load fastest.
Session agenda
- Your questions, live, on your data, not a scripted script
- The SQL behind three answers, read line by line
- What happens when the question is ambiguous or out of scope
- Entitlements: the same question asked by two different roles
What we need from you
- Read-only access to one warehouse schema or a representative extract
- Ten to twenty real business questions your team asks today
- Existing metric definitions, however incomplete
Where data cannot leave a system, the demo is built inside your environment instead of ours.
FAQ
Frequently asked questions
Does it write to our systems?
No. DataBot runs on a read-only path. Anything that changes a record belongs to AgentOS, where approvals and action logs are designed in.
What if a metric is defined differently in two systems?
The demo surfaces that conflict instead of hiding it, that disagreement is usually the most valuable output of the session.
Is Arabic handled natively?
Yes, Arabic questions are resolved against the same semantic layer, including organisational terminology, without a translation round-trip.
Topic cluster
Explore the topic cluster
Curated links across our Saudi hubs, sector pages and long-form blog research.