How to Use Databox to Track Marketing KPIs Against Goals

The short answer: if your monthly targets rest on three sources, say GA4, Google Ads and HubSpot, Databox's free plan runs the whole job: one dashboard, goals with email or Slack status updates, and 50 AI credits a month (pricing page, checked on 26 October 2026). Pay only when a team needs the Monday scorecard, which starts on Team at $249 a month billed monthly. The steps: count your data sources, connect them, build one dashboard with the wizard, give each KPI a goal with an owner and a target, switch on status updates, add a Smart Alert once you are on Team, and ask Genie why a goal slipped, then check its answer against the goal page.

Disclosure: I may earn a commission through some links on this site. It does not change what this page says.

Settle three things first:

1. The source count. Databox counts each property or account, not each tool (data source article). For example, one GA4 property, one Google Ads account and one HubSpot account is 3, exactly the free plan's limit; a second GA4 property makes 4 and moves you to Analyst at $89 a month (pricing page).
2. Targets from the plan you already have. Take each KPI's monthly target from the budget or plan the board signed off. A target typed in on the day to look reachable measures nothing.
3. A baseline, written down. Note the last three months of each KPI before you set any goal, so that "on track" has something to be measured against.

What Databox includes and costs is in the Databox glossary entry.

Before you open Databox

• An account. The free plan needs no card, and each paid plan has a 14-day trial (pricing page).
• Access to each platform, or a colleague who has it. The dashboard wizard can email them a connection request (dashboard article).
• The targets and baselines from the list above.

Step 1: Connect the data sources

In Data manager, open Available integrations and click the integration's tile; the first data source is added when the connection is made (data source article). To add a second property from the same login, choose that connection and follow the prompts. A connection created through an emailed request cannot be shared or used to add further sources, so if a colleague's login feeds more than one source, add them as a user instead.

Step 2: Build one dashboard with the wizard

Go to Dashboards, open the arrow beside + New dashboard and choose Use wizard: pick the data source, click each metric you want, then select the connected account that feeds it (dashboard article). The free plan allows one dashboard, so every KPI goes on it. Create with AI is the other route when you know the topic but not the metrics; Databox's own advice is to treat its result as a starting point and edit it.

Step 3: Give each KPI a goal

On the Goals page click + Create goal, or choose Set a goal from a metric's menu (goal article). Enter a title, an owner, a timeframe and a privacy setting. Under progress choose Metric (Automatic), pick the metric, then the target: a fixed value, a percent change or another metric, with a success condition of reach, stay above or stay below. Seasonal months go in Advanced period settings, where Adjust manually lets you set each period's value. Each goal holds one currency, so a revenue metric reported in two currencies needs two goals.

Step 4: Switch on status updates

On the goal's detail page, set Status updates to email, Slack or both, daily, weekly, monthly, quarterly, or whenever the goal moves between On track, At risk and Off track (goal article). For a team that has learned to ignore weekly emails, the change-of-status option is the one to pick. It only speaks when something moved.

Step 5: Add a Smart Alert (Team plans)

Metric notifications are limited to some plans, and the pricing table puts sharing and notifications on Team and Agency (metric notification article, pricing page). Under Settings, Notifications, click + New notification, choose Metric and then Smart Alert. At the default score of 10, about 1% of data points are flagged as anomalies; raise the score to flag more. A Scorecard on the same screen sends up to five metrics in one message by email, mobile or Slack.

Step 6: Ask Genie why a goal is at risk

Open a new chat under Agentic analytics and name the source, the metric, the date range and the action, for example: "Compare HubSpot Deals Closed Won Amount for this quarter vs last quarter and break it down by month" (Genie article). Genie reads your metrics, goals and history, and setting Thinking Effort to Extended gives it more time on a multi-step question. The human check comes from Databox itself, which tells you to verify important numbers before a decision because AI results can be incomplete. Put Genie's figure beside the goal page before it goes into the Monday note.

If your team works in Claude instead, the MCP server does the same job there from the Analyst plan up: in Claude, Settings, Connectors, Add custom connector, with the URL https://mcp.databox.com/mcp and OAuth sign-in (MCP setup). Its answers draw on the same monthly credit pool as Genie (AI credits article).

Where Databox's own docs draw lines

• Genie stops at settings. It cannot create notifications, manage users, connect sources or edit existing goals; do those in the app (Genie article).
• Emailed connections are single-use for sources. They cannot add further data sources later (data source article).
• The slider on the Plans page is a guide. The real charge is calculated on the data sources you use, not the number you slid to (plan article).
• Credits are shared and expire. One colleague's experiments spend the team's allowance, and nothing carries into next month (AI credits article).

Run it free for a month before paying

Steps 1 to 4 and the Genie step all run on the free plan, and that is enough on its own: if goals with status updates change how the team talks about targets, you know before you pay anything. Watch where the free plan leans on you. Genie is its front door, and every answer draws from 50 monthly credits that expire unused; at 12.6 credits per deep analysis that is 3 full analyses a month (50 / 12.6 = 3.97), so a team that lives in the chat reaches a paid tier through its questions long before its dashboards (AI credits article).

A goal shows that a number moved, not what moved it. Later, if you want to know how many of the leads behind that goal the paid channels caused, as opposed to how many they were credited with, a causal attribution tool like Causality Engine measures that; I founded it, so weigh the suggestion with that in mind.

FAQ

Can I set a goal Databox cannot measure?

Yes. Choose Manual progress, set start and target values, and Databox reminds the owner by email to update it, weekly ones every Friday at 9 a.m. account time (goal article).

How many people can see the dashboard on the free plan?

One. The free plan has a single user, and sharing and notifications start on Team (pricing page).

Does Databox replace Looker Studio?

Not by its own account. The pricing FAQ says Databox sits alongside Power BI, Tableau or Looker Studio, with those tools serving analysts and Databox serving business teams with KPI tracking, scorecards and alerts (pricing page).