Data & analytics consulting · Vancouver, BC

Turn messy data
into clear decisions.

I'm Ashita Diwan, a Vancouver-based data scientist and analytics consultant. I help teams cut through the noise and answer the questions that actually move the business.

Scattered data points converging into clear, parallel lines of insight
From complexity to clarity
Dashboards
people actually use
Models
you can interpret
Processes
that stop leaking time
Answers
in plain English
What I do

Three ways I help.

Most teams don't have a data problem. They have an answers problem. There's plenty of information; what's missing is a clear read on what it means and what to do next. That's the gap I work in.

01

Data Science & Predictive Analytics

Forecasting, risk scoring, and AI tools you can actually interpret, with no black boxes. I build models your team can explain to stakeholders and act on with confidence.

ForecastingRisk & churn modelsRAG / LLM toolsAnomaly detection
02

Data Analytics & Business Intelligence

Reporting and dashboards that answer real questions instead of adding more charts. KPI frameworks, performance analysis, and the one view your team checks every morning.

DashboardsKPI frameworksReportingData cleanup
03

Business Systems & Process Optimization

Find where time and margin leak, then fix it. I map how work actually flows, spot the bottlenecks, and design practical automation that fits the team you already have.

Workflow mappingAutomationData pipelinesOps analysis
Selected work

Real problems, real answers.

A couple of recent engagements, anonymized. Same pattern each time: start with the question that's actually keeping someone up at night, then let the data settle it.

Multi-location restaurant group

Is delivery quietly killing the business?

The problem
The owner was convinced third-party delivery apps (DoorDash, Uber Eats, Skip) were cannibalizing higher-margin in-house orders and eroding the business, but no one could prove it.
What I did
I pulled two years of point-of-sale data across brands and channels, then analyzed order mix, per-order value, and discount behavior to separate real cannibalization from normal shifts in how customers order.
The result
Delivery share had actually plateaued, not grown. The real driver of declining takeout was the drive-thru, not the apps. Along the way I surfaced a more urgent issue no one had caught: a 4.7% margin leak from one platform's discounting, worth more than the question they'd originally hired me for.
Build · Document Q&A assistant

Ask a folder of spreadsheets a question

The problem
Even a small knowledge base gets hard to use when it is split across formats. Answers you need are sitting in a few Excel sheets and a couple of Word and PDF files, but getting them out still means opening every file and reading.
What I did
I built a retrieval assistant over a Google Drive of mixed documents, using n8n to run the workflow and Claude to write the answers. Ask a question and it pulls the relevant rows and passages from the spreadsheets and documents, then returns a plain-language answer that cites the exact file it came from, so nothing is invented.
The result
A working assistant that turns a scattered set of spreadsheets and documents into something you can simply ask. It shows the pattern I can extend to a team's real knowledge base: grounded answers, sources attached, running on tools they already use.
Ashita Diwan, data scientist and analytics consultant based in Vancouver
About

Hi, I'm Ashita.

I'm a data scientist and analytics consultant based in Vancouver, with a Master of Data Science from UBC. I started VysionCo to do one thing well: help organizations make sense of their data and turn it into decisions they can stand behind.

My work runs the full range, from predictive models and LLM/RAG tools to the less glamorous but critical work: cleaning messy data, untangling reporting, and figuring out why two numbers that should match don't.

Whether I'm working with a leadership team, an operations department, or the technical folks, the goal stays the same: give you an answer you can act on, explained in language you don't need a data team to translate.

UBC Master of Data ScienceData ScientistAnalytics ConsultantVancouver, BC
Why work with me

A different kind of consulting partner.

01

Answers, not jargon

I translate technical work into decisions you and your stakeholders can actually understand, then defend in the room.

02

Interpretable over impressive

I'd rather ship a simple model you can explain than a complex one nobody trusts. No black boxes, no hand-waving.

03

Built for real constraints

Limited data, tight budget, small team? That's most projects. I find pragmatic solutions that fit what you actually have.

How it works

Four steps from question to clarity.

Step 01

Discover

Understand your goals, the real question, and the systems behind it.

Step 02

Analyze

Dig into the data to find the patterns, bottlenecks, and opportunities.

Step 03

Design

Turn findings into practical recommendations, models, or dashboards.

Step 04

Deliver

Implement it, explain it in plain English, and support the decisions.

Book a consultation

Let's make sense of it together.

Trying to understand your data, tighten a process, or make a call you can defend? Grab a free 30-minute intro call and we'll talk through what you're working on and whether I can help.

Contact

Start a conversation.

Tell me a little about what you're working on. I read every message myself and reply personally.

VysionCo logoVysionCo

Data science & analytics consulting · Vancouver, BC

© 2026 VysionCo. All rights reserved.