Datapad
Datapad is an autonomous data analyst agent that connects fifty plus sources, answers questions in plain language, monitors key metrics proactively, and produces ready to deploy campaigns and reports.
Datapad is an autonomous data analyst built to replace static dashboards with an agent that answers business questions in plain language and watches key metrics on its own. Founded by Cem Ruso, who earlier exited BluTV to Time Warner, and Orkun Soylu, formerly of Insider, the company is based in Istanbul and backed by investors including 500 Startups and Pitchdrive. Its pitch is simple. Take a company database and marketing stack, then tell the owner what is working and what is not, without the cost and delay of building analytics capacity by hand.
Users talk to a single chat, but behind it a team of specialist agents for marketing, search optimization, conversion, and business strategy work together to build each analysis. The agent converts plain language into SQL and executable Python, blends data across sources, and returns charts, tables, and dashboards. It goes beyond describing what happened and generates deliverables, producing campaign plans, optimization checklists, and written content that a team can act on immediately rather than interpreting raw numbers themselves.
Datapad connects to more than fifty data sources, spanning SQL databases, BigQuery, Snowflake, advertising platforms such as Google Ads and Facebook Ads, social channels, and customer relationship management tools. Rather than waiting for a question, it monitors metrics constantly, detects anomalies, and pushes alerts and scheduled analyses to a team through a Slack bot, email, or even voice interaction. A memory layer learns a company over time, holding its business context, preferred metric definitions, and internal terminology so answers grow more tailored the longer it is used.
A library of more than fifty ready to use workflows gives teams a starting point for common analytical jobs, and agencies can spin up white label client dashboards that consolidate many marketing platforms into one branded view. The product is delivered as a hosted service rather than something a customer runs on its own hardware, and it centers on its own agent stack rather than published named security certifications or a model routing gateway. Datapad remains an independent venture backed startup serving more than one hundred customers, and it fits growth and marketing teams that want proactive answers and finished deliverables rather than another dashboard to maintain.
Vendor details
Canonical URL
https://datapad.io
Category
Data analyst agent
Company status
independent
Use cases & customers
In practice
A growth team gives Datapad a weekly brief, and the agent monitors campaign spend across advertising platforms, flags anomalies, and delivers a written performance analysis to their inbox before the Monday meeting.
A marketing agency connects several client advertising accounts, and Datapad consolidates them into white label dashboards while generating optimization checklists and content the agency can hand to each client the same day.
A founder asks in plain language for customer lifetime value by acquisition channel, and Datapad writes the SQL, runs it across blended sources, and returns an explained chart she can drill into.
Sources & related URLs
Research notes
Score 6.0 (4F/4P/6N). Fulls: Int (50 plus sources across DBs, ad platforms, social, CRM, plus SQL and Python tool calling), Orch (100 percent autonomous, multi agent team, takes a brief and acts on a schedule, produces deliverables), Mem (persistent cross session memory layer: business context, metric definitions, terminology, historicals), Trig (proactive constant KPI monitoring, anomaly alerts, scheduled tasks, Slack and voice). Partials: Know (business context and metric memory but no formal governed semantic layer; schema grounding level), Obs (drill through, anomaly explanations, query transparency), Pack (50 plus ready to use workflow library, not a cloneable agent marketplace), Ext (RESTful API present but no clear SDK or MCP). N: HITL (branded 100 percent autonomous, no documented guardrail or approval gating), Sec (Istanbul startup, no named certs verified), Dep (SaaS only), Model (own agent stack, no documented multi provider routing), Eval (no documented test harness), Comp. Marketing analytics leaning autonomous agent.
Capability coverage
6.0 / 14 capabilities · 43%
| Integrations & Tool CallingDatapad connects to more than fifty data sources across databases, advertising platforms, social channels, and customer relationship management tools while writing SQL and executing Python, so full. | Full |
|---|---|
| Workflow OrchestrationDatapad runs a team of specialist agents for marketing, optimization, and strategy that work together autonomously, taking a brief and producing full analyses on a schedule, so full. | Full |
| Knowledge Grounding & RAGDatapad grounds answers in a memory of business context, preferred metrics, and terminology, a learned schema level grounding rather than a formal governed semantic layer, so partial. | Partial |
| Human Oversight & GuardrailsDatapad markets itself as a fully autonomous analyst and no runtime guardrail, approval gating, or clarification workflow could be verified as a first class feature, so not documented. | Unable to verify |
| Security, Identity & GovernanceDatapad publishes no named third party certifications such as SOC 2, and none could be verified for this early stage company, so not documented. | Unable to verify |
| Observability & AuditabilityDatapad lets users drill through any chart or insight and explains detected anomalies, offering query and result transparency rather than full trace auditability, so partial. | Partial |
| Memory & State PersistenceDatapad runs a memory layer that persists business context, preferred metric definitions, and company historicals across sessions so answers grow more tailored over time, so full. | Full |
| Deployment & Data ResidencyDatapad is delivered only as a hosted service with no self host, on premises, or open source deployment option documented, so not documented. | Unable to verify |
| Prebuilt Agents, Templates & PacksDatapad ships a library of more than fifty ready to use workflows that create deliverables, a template library rather than a marketplace of cloneable agents, so partial. | Partial |
| Triggers & Channel CoverageDatapad proactively and constantly monitors key metrics, detects anomalies, and pushes scheduled analyses and alerts through a Slack bot, email, and voice, so full. | Full |
| Model Flexibility & RoutingDatapad runs its own agent stack and no multi provider model selection or routing gateway could be verified, so not documented. | Unable to verify |
| APIs, SDKs & MCP ExtensibilityDatapad exposes a representational state transfer application programming interface for integrations, but no software development kit or Model Context Protocol surface could be verified, so partial. | Partial |
| Testing, Debugging & OptimizationDatapad offers anomaly detection but no agent testing, benchmarking, or evaluation harness could be verified as a first class feature, so not documented. | Unable to verify |
| Browser & Computer UseDatapad is a data analytics tool with no browser or computer use capability, as expected for this category, so not documented. | Unable to verify |
Pricing
Free tier; Pro paid (monthly/annual)
Related vendors
- AskYourDatabase — AskYourDatabase is a conversational data analyst that turns plain…
- Athenic AI — Athenic AI is an agentic data analyst that connects business apps…
- BlazeSQL — BlazeSQL is an AI data analyst that learns your SQL database, turns…
- Brewit — Brewit is a conversational business intelligence agent that turns…
- Buster — Buster is an open source, artificial intelligence native data…
- Definite — Definite is a full stack, artificial intelligence native data…