A KPI dashboard that pulls HubSpot, GA4, ad platforms and your own database into one refreshing view, then sends the client report out on a schedule.
What is
Databox
The numbers a GTM team argues about on a Monday live in five places at once, with pipeline in HubSpot, sessions in GA4, spend in Google and LinkedIn Ads, and revenue in Stripe. Databox exists to end that copy-and-paste, pulling all of it into one dashboard that refreshes without anyone touching it. It connects through roughly 130 native integrations, plus Google Sheets, Excel, SQL and NoSQL databases, and a REST API for anything with no connector of its own. You pick metrics from a prebuilt library or build your own with filters and calculations, then drop them onto dashboards and reports that send themselves on a schedule. The newer layer tries to answer questions rather than only draw charts. Genie is an in-app AI analyst that runs real queries against your data, and a Databox MCP server hands the same governed metric definitions to Claude or ChatGPT. Agencies are the heaviest users, with sub-accounts and white-labelling built around client reporting. Databox reports and it does not attribute. It will show that demos rose 18 percent and that LinkedIn spend rose alongside them, and joining those two facts stays a human job. We have not run it at Nebor, and a team that wants credit assigned across a buying journey should look at Dreamdata or HockeyStack.
Integrations and the modern GTM stack
Databox is a read-layer, so its integrations all point one way: data comes in, dashboards go out. Roughly 130 native cloud connectors sit alongside spreadsheets, warehouses and a push API, and the number you connect is also the number that sets your bill.
HubSpot | The deepest connection on the list, split into HubSpot Marketing, CRM, Service and Custom Objects, each counted as its own data source. It is the reason most B2B teams land here, since it lets pipeline sit on the same board as ad spend instead of behind a HubSpot report nobody outside marketing opens. |
Google Analytics 4 and the ad platforms | GA4, Google Ads, Facebook Ads and Pages, LinkedIn Ads, TikTok, Pinterest and Reddit. Every property and every ad account counts separately, so a business running two GA4 properties and three ad accounts is at five sources before the CRM is wired in. |
Salesforce, Pipedrive and Copper | Standard CRM pulls for pipeline, deals and activity. Reviewers single out Salesforce as the weaker of the CRM connectors for anything that needs custom queries, so check your specific objects during the trial rather than after the contract. |
Warehouses and databases | BigQuery, Snowflake, Postgres, MySQL, Microsoft SQL Server, Oracle, MongoDB and Azure, through a Query Builder that avoids hand-written SQL. This is the escape hatch when a metric has no native connector and lives in your own tables. Databases need Analyst or above. |
Google Sheets, Excel, Zapier and Make | The route for numbers that live nowhere official, such as a manual pipeline sheet or a finance export. Zapier and Make push rows in from tools with no direct connector, which is how most odd internal metrics get onto a board. |
Databox REST API | Push raw structured data in from anywhere and build metrics on top of it. For a Clay-run GTM stack this is the practical path, since you can send counts of enriched accounts, replies or booked meetings straight in from an n8n or Make step. |
Databox MCP server | Released 1 June 2026 and published openly on GitHub at databox/databox-mcp. It exposes governed metric definitions to Claude, ChatGPT and custom agents so answers come from real queries against your data. It burns AI credits from the shared monthly pool, and it needs Analyst or above. |
What is missing | There is no attribution model anywhere in the product, so journeys, touch weighting and channel credit are not on offer at any price. Cold email infrastructure is also absent, with no native Instantly, Smartlead, Lemlist or Clay connector, meaning outbound numbers reach Databox through the REST API or a Sheet. |
Pro’s
The connector list is the real product, with roughly 130 native apps plus BigQuery, Snowflake, Postgres, MySQL, MongoDB, Sheets and a push API, so almost any number your team argues about can land on one board without a data engineer.
Every team plan carries unlimited users, so you can hand a dashboard to sales, finance and the client without a seat negotiation, and that is unusual in a category that usually meters people.
Scheduled snapshots, Slack digests, scorecards and anomaly alerts mean the weekly update sends itself, and agency reviewers credit exactly that with taking hours of manual reporting out of the month.
The prebuilt metric library and dashboard templates get a HubSpot or GA4 board live in an afternoon, so the tool proves itself before anybody has to build a custom metric.
The MCP server released in June 2026 lets Claude or ChatGPT query your governed metric definitions and get results from real queries rather than a model guessing at a pasted table.
Benchmarks compares your metrics against anonymised cohorts of similar companies by industry, size and revenue, which gives a small team outside context that a self-built dashboard will never have.
Cons
Data sources are the meter and each account or property counts on its own, so three GA4 properties swallow the whole three-source allowance on Pro before HubSpot or Google Ads is even connected.
Historical data stops at 24 months on everything below the $399 Growth plan and at 11 months on Free, which breaks year-over-year work for any client you have held longer than that.
Reviewers on Capterra and G2 describe connectors that demand reauthentication and widgets that blank out or pull the wrong ad account, and a dropped connector takes every metric and template built on it down with it.
Databox charts what happened and never models attribution, so buyers who arrive hoping to learn which touch created the deal end up wiring 15 connectors to answer a different question.
Layout snaps to a grid with a small set of colour schemes, and white-labelling is a paid add-on below the top tiers, so a client-facing report that looks like yours costs more than the sticker price.
Pricing
Free
$0/month 1 user, hard cap at 3 data sources
3 data sources, cloud integrations only
1 dashboard or report and 10 custom metrics
11 months of historical data
Daily sync and 50 AI credits a month
Sharing links, TV streams and notifications
Analyst
$64/month Billed annually, 1 user, 5-source hard cap
All integrations including databases and custom API
Unlimited dashboards, reports and custom metrics
24 months of history and hourly sync
Datasets and the MCP server
500 AI credits a month
Pro
$159/month Billed annually, unlimited users, $5.60 per extra source
3 data sources included and no upper limit
Unlimited users with role-based access
Goals, CSV export and Slack notifications
Datasets limited to spreadsheets, databases and custom sources
1,500 AI credits a month
Growth
$399/month Billed annually, unlimited users and sub-accounts
Unlimited historical data and 15-minute sync on up to 5 sources
Full datasets, OKRs and metric forecasting
Unlimited sub-accounts for client reporting
Dedicated customer success manager
4,000 AI credits a month
Custom
Talk to sales Everything in Growth with negotiated limits
White-labelling included
SSO, audit logs and session controls
Negotiated data source and AI credit limits
Guided account setup and priority support
Security and compliance review
How it fits your stack
Alternatives
At a glance
Around 130 native cloud integrations
SQL, NoSQL, Sheets and REST API sources
Unlimited users on every team plan
Scheduled reports, Slack digests and alerts
Genie AI analyst and MCP server
Sub-accounts and white-label for agencies
Goals, OKRs and metric forecasting
Free plan and a 14-day Growth trial
Read some of our client stories
GTM Workflows
Learn how to use GTM tools in workflows
Guides and tutorials to get the most from your workflows
Ready to get started?
By clicking Sign Up you're confirming that you agree with our Terms and Conditions.
