Datrix.

The problem

Logistics and operations teams produce massive amounts of weekly performance data but the analytics work that turns it into decisions is manual, repetitive, and analyst-bottlenecked. Data is fragmented across Excel, ERP, and WMS exports; column formats are inconsistent; pivot tables and chart-building eat hours per week; and writing the summary email "what changed and why" takes longer than the analysis itself.

The deeper pain is uncertainty about what to look for. Most operational analysts know the data has trends but don't always know which are worth surfacing to leadership — and the trends they do find often lack explanation ("orders dropped 12% — why?") that managers actually want.

The solution

Datrix collapses the entire weekly reporting loop into a single drag-and-drop. Upload Excel → the system detects column types and structures messy data → auto-generates charts and KPI cards → an AI insight panel explains what's happening in plain language → suggested prompts offer obvious next questions ("why did Region B drop?") → a chat input handles anything else.

Crucially, the AI explains why changes happened — not just that they happened — giving managers the talking points they'd otherwise ask analysts for. Anti-hallucination guardrails enforce factuality: the model must reflect direction (increase/decrease), magnitude approximation, and never invent trends or numbers that aren't in the data.

How it works

Flow: drag-and-drop area → "Upload your data" (or click sample dataset) → automatic structuring → KPI summary cards → auto-generated charts → insight cards (AI text output) → suggested prompts → chat input at bottom. V1 must-haves: file upload, chart visualization, AI insight generation, basic chat. V2 backlog: ERP/Google Sheets integrations, advanced analytics, anomaly detection alerts, root cause analysis.

System prompt: the AI is positioned as a senior data analyst and strategy consultant specialized in operational and logistics data — clear, analytical, concise, business-focused tone. Primary model: GPT-4o-mini, selected for cost-efficiency, speed, and sufficient reasoning capability for structured data insight generation without the latency of a frontier model.

Who it's for

Primary buyers: Logistics Managers, Operations Leads, Supply Chain Directors at companies with operational complexity but no full data science team. Revenue-impacting end-users: the logistics and operations analysts generating weekly performance reports — whose pain points are direct and whose adoption decisions the buyer will follow.

Secondary consumers: managers reading the AI-generated insights — they get trend explanations and reasoning they previously had to ask an analyst for.

Why it matters

The global BI and Analytics market is growing at 8–12% CAGR, while the AI-powered analytics subsegment is growing at ~20% — and the wedge between the two is exactly Datrix's positioning. Mainstream BI tools require analysts to know what to look for; AI-native analytics tools generate that "what to look for" automatically.

The strategic moat is domain awareness. Generic AI analytics tools work across any dataset but understand none of them deeply. Datrix's logistics-specific schema awareness and KPI defaults create a faster path to value for the target buyer than a horizontal tool can match — a deliberate niche-first strategy that becomes a moat as the model learns the segment.

At a glance

Project

Datrix

Built by

Burcu Kilinc

One-liner

AI no-code analytics for logistics & operations — upload Excel, get auto-generated insights, trend explanations, and weekly summaries.