Product tour
See it work. Ten steps through nexibi, on one sample company.
Follow one question at Tarnwick Supply Co., from the first answer to a follow-up waiting for a person's yes. Click through at your own pace, or watch a short film.
Ten steps through nexibi,
on one sample company.
Step 1Connect a system
Built withData connections
Connect a database, any web system with an API, or a ready-made connector, and add your Microsoft 365 files. nexibi reads your systems where they are, so there is no data warehouse to build first. Where a system allows it, each person connects with their own account.
Data connectionsStep 2Ask a question
Built withAI Assistant
Ask the way you would ask a colleague. It reads the whole list, up to 20,000 rows, and tells you when a list is longer. Counts and totals are calculated from the rows it reads, not estimated by the AI, and when the answer is a list, you get the list itself.
AI Assistant37 open orders are past their promised date, worth $184,920.
| Order | Customer | Value |
|---|---|---|
| SO-10482 | Quellmark Foods | $12,840 |
| SO-10455 | Fenwright Builders | $9,615 |
| SO-10471 | Lumora Hotels | $8,930 |
Step 3Ask your documents
Built withAI Assistant
Keep the files for a piece of work in a project, and ask about them. It reads a PDF page by page and tells you the page each answer came from.
After 10 working days late, Quellmark may cancel the order at no charge.
Quellmark supply contract.pdf, page 4Step 4Keep it as a dashboard
Built withDashboards
Ask for a dashboard and it builds one from your data, checks that each tile shows real numbers, and hands you an ordinary dashboard. Press Edit to change any chart in the designer, or build one yourself from the start.
DashboardsBuilt by the assistant from one request: "open order value by region and month, with a filter by sales person"
Step 5Share it safely
Built withDashboards
A dashboard is private until you share it, with people, teams or roles. Everyone in the company means view only. Sharing it shares the data behind it, only what is yours, and only people signed in to your company can open it.
Step 6See how things moved
Built withHistory engine
Many systems only know how things are today. Keep a daily copy of a list, and a chart can show how it moved, week by week. The chart starts on the day you turn the copy on, for systems with a web API.
History engineCopy kept since 18 Sep
Step 7Find who to act on, and what comes next
Built withAnalysis & prediction
Ready-made analyses list the rows worth acting on, each with its reason, such as customers going quiet. A forecast gives a likely range and says how far off its past forecasts have been.
Analysis & predictionStep 8Let a workflow follow up
Built withWorkflows
Your company's owner builds a workflow in the flow designer: when it starts, what it reads, what it writes. Try it first: nothing is sent and nothing is changed. Add an "Ask someone first" step, and it waits for a person's yes. A workflow reads its lists from systems with an API and from the ready-made connectors; it cannot read a connected database today.
WorkflowsTry it first: nothing is sent
and nothing is changed.
Step 9Give a job to an AI agent
Built withAI Agents
Describe the job in your own words, give it your documents and the workflows it may run, and try it on the same page. Anything it would change in your systems or send to someone waits for a person's yes. It can never approve its own work.
AI AgentsAnswer quote requests from our price list. Check stock before you quote. Never offer more than 5% off.
Step 10Approve what waits for you
Built withWorkflowsAI Agents
One page, called Waiting, shows everything waiting for you: approvals, questions and failed runs. See what each run did, step by step, and run a failed one again from where it stopped.
Prefer to watch?
A 70-second film of nexibi, with drawn screens.
The screens in this film are drawn, not recorded, and some show things a customer cannot use today. Music only, no speech.
Today a scheduled check sends e-mail, and waits for a person's yes where it has an 'Ask someone first' step. WhatsApp is being added.
What the film shows, in words
A 70-second animated film on a light ground, with music and no speech, its drawn screens set in a made-up trading company. A dot travels from "Turn data into insight" to a list of connections, then to a typed question, "Which branch is behind target?", a ranked answer and a dashboard.
At 6:00 am, under "Your day, briefed.", it shows the day's list of what is worth doing. Under "Put the routine on autopilot.", a flow runs every hour, checks whether stock is below its minimum, and messages the branch manager on WhatsApp and purchasing by e-mail; the manager replies. WhatsApp is being added.
It ends with "Wherever you are", the app on a laptop and a phone, and "Business intelligence that acts."
What you saw.
Seven engines, on one set of connections and one set of sharing rules: the AI Assistant; Dashboards, Analysis & prediction and the History engine; Workflows and AI Agents; all on your Data connections.
Platform overviewNow try it on your own system.
Book a 30-minute demo. We connect one of your systems, and you ask the questions.