GROUNDED
Answers with receipts
Every number arrives with the SQL that produced it, the source it came from, and what the scan cost. Nothing is asserted that wasn't fetched.
Agentic analytics · open beta
The AI analyst that shows its work.
DataLens plugs into your data warehouse, Google Ads, Meta Ads, and your spreadsheets, answers plain-English questions with live queries — and hands you the receipt behind every number.
DataLens AI is an agentic analytics product for marketers and analysts that turns plain-English questions into live, verifiable SQL across BigQuery, ad platforms, and spreadsheets. Every answer ships with its query, sources, and scan cost — how to verify any AI answer and what a warehouse query costs before you run it.
Bring your own AI modelsYour keys stay with youYour data stays in your stack
Preview of a DataLens analysis session: the analyst asks what drove blended ROAS up last week across BigQuery, Google Ads, and Meta Ads; DataLens discovers the schema, writes and runs the SQL, cross-checks the Meta API, and answers that blended ROAS rose to 3.86x with a verified query.
You asked
What drove blended ROAS up last week?
BigQueryGoogle AdsMeta Ads
Discovered schema ads_performance
Wrote & ran SQL 12.4 MB scanned
Cross-checked Meta Insights API
Built visualization 7-day trend
Blended ROAS rose to 3.86× — Meta prospecting scaled efficiently while branded search held steady.
- Meta prospecting CPMs fell 12% while conversions held — cheaper reach, not just more of it.
- Branded search volume stayed flat, so the lift isn't cannibalized demand.
- Spend tracked to plan — the gain came from efficiency, not extra budget.
Blended ROAS3.86×+13.9% w/w
Spend₹3.8Lon plan
FreshnessLiveverified now
Query verifiedyour model, your keys
One analytical layer across
- BigQuery
- Google Ads
- Meta Ads
- Google Sheets
- CSV & Excel
- Gemini
- Claude
- GPT
- Grok
- Local models
Product
Everything between a question and a decision.
DataLens does the analyst’s operational work — discovery, querying, checking, charting — and keeps every step inspectable along the way.
CONNECTED
One conversation, six sources
Your data warehouse, Google Ads, Meta Ads, live Google Sheets, CSV and Excel — joined in a single thread instead of six browser tabs.
SHIPPABLE
From answer to artifact
Interactive Plotly charts become read-only share links you can send to anyone — without leaving the chat.
BYO-MODEL
Your models, your keys
Run Gemini, Claude, GPT, Grok — or fully local models via Ollama and LM Studio. API keys live in your browser, not in someone’s database.
ON-BRAND
Reports in your brand
Drop in a logo and DataLens derives a contrast-safe palette for every chart and share link — client-ready by default.
HANDS-FREE
Ask out loud
Real-time dictation turns a spoken question into a running analysis — useful mid-meeting, when the number is needed now.
How it works
From “what happened?” to “here’s why” in four moves.
Built around the actual rhythm of analysts and performance marketers — not a staged demo path.
01
Connect
Point DataLens at a warehouse, an ad account, a live Sheet — or just drop a file on it.
02
Ask
Type — or say — the question the way you'd ask a colleague. No SQL, no report builder.
03
Inspect
Watch the agent work: schema discovery, the exact query, data freshness, scan cost.
04
Share
Send a read-only share link to anyone — no login required.
Built for trust
Numbers you can defend in the Monday meeting.
An AI that invents figures is worse than no AI at all. DataLens runs on strict grounding rules: it must fetch real data before it speaks, cite exactly what it fetched, and let you re-run the same query against live data — any time someone asks “where did this come from?” Seven checks before trusting an AI answer · Estimate a query's cost first
6source types
10+AI providers
15agentic steps / run
AES-256credential encryption
METRICBlended ROAS
DEFINITIONrevenue ÷ ad spend
SOURCESwarehouse · google_ads · meta
FRESHNESSlive · re-run 14s ago
QUERY SCAN12.4 MB · within budget
STATUSVERIFIED
CR
built where marketing meets data
About the builder
Built by an analyst, for analysts.
I’m Chinmay Raibagkar — a marketing analyst focused on data and automation at Maino AI. My week runs through data-warehouse pipelines, ad platforms, and dashboards, which is exactly where DataLens was born: as the tool I kept wishing existed.
Every feature traces back to a real Monday-morning problem — numbers that need defending, reports that keep getting rebuilt, and questions that deserve answers faster than a ticket queue.
See my work and backgroundFAQ
Asked before you ask.
The five questions every first conversation starts with — data, sources, models, and why this is not another dashboard.
Does DataLens store or copy my database rows?
DataLens does not replicate your database. Every query runs live against your connected warehouse or ad accounts — there is no background sync and no copy of your tables. What is persisted is the conversation itself: your questions, the answers, the generated SQL, your visualization settings, and up to 50 rows per message of the result behind an answer, so a past analysis still renders when you reopen it. Deleting a conversation removes it. The Privacy page has the full retention disclosure.
Which databases, warehouses, and platforms are supported?
Today: Google BigQuery, Google Ads, Meta Ads (Facebook/Instagram), Google Sheets, and direct CSV/Excel uploads. PostgreSQL, MySQL and Snowflake are on the roadmap but are not connectable yet — if one of those is your primary warehouse, tell us and it moves up the queue.
Can I bring my own OpenAI, Claude, Gemini, or Grok API key?
Yes. DataLens is model-agnostic. Use the managed models, or bring your own key for Google Gemini, Anthropic Claude, OpenAI, xAI Grok, Moonshot Kimi, NVIDIA NIM or OpenRouter. You pick the model per conversation, so you retain control over which provider ever sees your questions.
How do I know the AI-generated SQL query is accurate?
Every answer ships with the exact SQL that produced it, expandable in place — you can read the joins, the WHERE filters and the aggregation grain without leaving the conversation. Alongside it, one-click Verify re-runs that same query live against your warehouse and flags any drift from the numbers the answer reported, along with how many bytes it scanned and what that cost.
How is DataLens different from traditional BI tools like Looker or Tableau?
Traditional dashboards only answer questions that were anticipated months ago when the dashboard was built. When you need an ad-hoc breakdown (e.g., "Show me refund rates for first-time buyers who came from TikTok Ads last week"), a dashboard requires submitting a ticket to the data team. DataLens answers ad-hoc questions in seconds with live, verifiable SQL.
All questions, answered in depth →
Content last reviewed September 2026.
Plans
Simple to start. Shaped to fit.
DataLens is in active beta, so pricing is scoped per engagement — the first step is always a conversation about your stack.
Individual
Explore the full product on your own data.
- Full product with your own model keys
- Files, Sheets, ad accounts & your data warehouse
- Read-only share links for your analysis
Team pilot
A guided rollout around your reporting stack.
- Hands-on setup with your data sources
- Governed metric definitions the team agrees on
- A direct feedback line that shapes the roadmap
Agency & custom
Cross-client reporting, deployed your way.
- Cross-account & cross-client analysis
- Client-branded charts & share links
- Custom deployment & integrations
No public price card while DataLens is in beta — tell me about your stack and I’ll come back with a scoped proposal.
Contact
Bring a messy data question.
Want to explore the product, run a team pilot, or scope a custom setup? Send the context — specific problems make the best starting point.
Email directlycontact.datalensai@gmail.com