A KPI dashboard is a live display of the metrics that tell you whether your business, team, or campaign is on track. Done well, it replaces the daily ritual of pulling data from five different tools and gives every stakeholder a single view of performance. Done poorly, it is a collection of numbers that no one trusts and that does not inform decisions.
This guide covers how to build KPI dashboards that actually get used: what belongs on one, how to connect your data sources, which tools to use, and how to avoid the most common design mistakes.
What a KPI Dashboard Is (and Is Not)
A KPI dashboard is a curated view of your key performance indicators, updated automatically from connected data sources, and designed to surface the metrics that drive decisions. The defining characteristics:
- Automated: data pulls from live sources without manual export or copy-paste.
- Curated: a dashboard with 40 metrics is not a dashboard, it is a data dump. An effective KPI dashboard shows 5-15 metrics that matter for a specific audience.
- Audience-specific: what an executive needs to see differs from what a campaign manager needs to see. Build separate dashboards for separate audiences rather than one dashboard that tries to serve everyone.
- Decision-oriented: every metric on a dashboard should answer a question someone is actually asking. If no one is making decisions based on a metric, it does not belong on the dashboard.
A KPI dashboard is not a reporting tool for producing PDF exports, a data exploration interface (that is what a BI tool is for), or a substitute for the analysis that interprets what the numbers mean.
Choosing the Right KPIs
The most common dashboard mistake is including metrics because they are available rather than because they drive decisions. Before building any dashboard, answer these questions for each candidate metric:
- Who uses this metric, and how often do they need to see it?
- What decision would they make differently if this number changed?
- Is this a leading indicator (predicts future performance) or a lagging indicator (reports past performance)?
- Do we have reliable, automated data for this metric?
KPIs by Business Function
Common KPI categories by function:
Marketing KPIs: leads generated, cost per lead by channel, marketing qualified leads (MQLs), website sessions by source, organic keyword rankings (directional), email open and click rates, paid ad spend vs. pipeline generated.
Sales KPIs: pipeline value by stage, deals in each stage, close rate, average deal size, average time-to-close, quota attainment by rep, new MRR or ARR booked.
Revenue KPIs: MRR, ARR, churn rate, net revenue retention, customer lifetime value (LTV), LTV-to-CAC ratio, average revenue per account.
Customer success KPIs: net promoter score (NPS), customer satisfaction score (CSAT), time-to-first-value, product adoption rate (feature activation), support ticket volume, escalation rate, renewal rate.
Executive KPIs: revenue vs. target, gross margin, runway, headcount vs. budget, LTV:CAC ratio, net new ARR.
Leading vs. Lagging Indicators
Lagging indicators (revenue, churn, closed deals) tell you what already happened. Leading indicators (trials started, demos booked, proposal sent) tell you what is likely to happen. A good KPI dashboard has both, but leading indicators are more actionable because you can still influence the outcome.
Example: if your dashboard shows that revenue is down, you cannot change it. But if it shows that demos booked dropped six weeks ago, you can investigate the cause and adjust. Include enough leading indicators that your dashboard serves as an early warning system, not just a scorecard.
Data Sources: What to Connect
The data sources most marketing and revenue dashboards need to connect:
- Web analytics: Google Analytics 4 (sessions, conversions, source/medium), Google Search Console (clicks, impressions, rankings).
- Paid advertising: Google Ads (spend, clicks, conversions), Meta Ads (spend, reach, conversions), LinkedIn Ads if applicable.
- CRM: HubSpot, Salesforce, Pipedrive, or similar. Source of pipeline, deal stage, close rate, and revenue data.
- Email marketing: Mailchimp, ActiveCampaign, or similar. Open rates, click rates, list growth.
- Product analytics: Mixpanel, Amplitude, or similar for SaaS products.
- Finance: QuickBooks, Stripe, or similar for ARR, MRR, and churn data.
The challenge is that each of these lives in a separate system with a separate login. Dashboard tools solve this by connecting to each via API and pulling data on a refresh schedule (typically every 15 minutes to 24 hours depending on the tool and data source).
Dashboard Tools: Which One to Use
Google Looker Studio (formerly Data Studio)
Looker Studio is free, connects natively to all Google products (GA4, Google Ads, Google Search Console, Google Sheets), and has a library of community connectors for non-Google sources (HubSpot, Facebook Ads, Mailchimp, Stripe, etc.). It is the right starting point for most small and mid-sized businesses because the price is zero, the Google connectors work reliably, and the report builder is intuitive enough that non-technical users can maintain it.
Looker Studio limitations: some community connectors require paid subscriptions (Supermetrics is the most common), cross-source joins require a blended data source that is fiddly to configure, and it is not designed for real-time data (refreshes are typically every 15 minutes at best).
Databox
Databox is a purpose-built KPI dashboard tool with over 100 native integrations, a mobile app for on-the-go monitoring, and pre-built dashboard templates for common use cases (marketing performance, sales pipeline, SEO). Its goal tracking and scorecard features make it well-suited for teams that want to track KPIs against targets, not just raw numbers. Pricing starts around $47/month for small teams.
Klipfolio
Klipfolio is a veteran dashboard tool with strong data connectivity and a flexible component-based builder. It is more powerful than Databox for custom metric calculations and transformations but has a steeper learning curve. Suited for teams that need complex custom metrics or want to build formula-driven KPIs from raw data.
Tableau and Power BI
Tableau and Microsoft Power BI are enterprise BI tools that can function as KPI dashboards but are primarily designed for exploratory data analysis and ad hoc reporting. They are the right choice if your organization already uses them for other purposes or if you need to build complex calculated metrics from warehouse data. For straightforward KPI monitoring, they are significantly over-engineered.
HubSpot Dashboards
If your team lives in HubSpot, its native dashboard builder covers marketing, sales, and service KPIs using HubSpot data. It is fast to set up and requires no connector configuration. The limitation is that it only shows HubSpot data, so you cannot blend it with Google Ads spend or GA4 behavior data natively.
Dashboard Design: What Makes a Dashboard Actually Usable
The most common design mistake is building a dashboard for yourself rather than for the person who will use it. Before you finish any dashboard, ask: if someone opened this dashboard without context, would they know what it is showing and what to do with the information?
One Audience, One Dashboard
Build a marketing performance dashboard for the marketing team. Build a revenue dashboard for the executive team. Build a campaign dashboard for the channel manager running paid ads. These three audiences need different metrics at different levels of granularity.
The executive does not need to see keyword-level ROAS. The paid search manager does not need to see gross margin. Combining them produces a dashboard that serves no one well.
Organize by Decision, Not by Data Source
A common layout mistake is organizing a dashboard by where the data comes from: a Google Ads section, a Facebook section, a GA4 section, an email section. This mirrors your tool stack rather than how you make decisions.
A better structure: organize by the question the dashboard answers. For a marketing performance dashboard, those questions might be: How much pipeline did marketing generate this month? Which channels are producing the most qualified leads? Where are we vs. target on leads and MQLs?
Use Targets and Comparisons
A metric without context is noise. 1,247 leads this month: is that good or bad? Add a target (1,500) and a prior period comparison (same month last year: 890) and it becomes meaningful: below target, but strongly up year-over-year. Configure your dashboard to show current value, target, and comparison period for every primary metric.
Limit Colors and Chart Types
Use color to signal status (green = on track, yellow = at risk, red = below target) rather than to decorate. Avoid pie charts for anything with more than 3 categories. Use bar charts for comparisons, line charts for trends over time, and scorecards for single current-value metrics. A clean dashboard with 8 scorecards and 2 line charts is more useful than a visually complex dashboard with 15 chart types.
Common KPI Dashboard Mistakes
- Too many metrics: if everything is a KPI, nothing is. Ruthlessly limit each dashboard to the 8-12 metrics that drive decisions for that audience.
- Vanity metrics: website sessions, social media followers, and email list size feel like progress metrics but rarely correlate with revenue. Include them only if they directly inform a decision.
- Stale data without a visible refresh timestamp: always show when the dashboard data was last updated. A dashboard that might be 24 hours old without labeling it creates false confidence in current-state data.
- No owner: every dashboard needs an owner who is responsible for keeping connectors working, updating targets when they change, and deprecating metrics that no longer matter. Unowned dashboards decay.
- Metrics without context: a number without a target, a trend, or a comparison period is not a KPI, it is a data point. Every metric needs the context required to evaluate whether it represents good performance or bad performance.
Attribution: Connecting Dashboard Metrics to Revenue
The hardest KPI dashboard problem is connecting marketing metrics (leads, MQLs, conversion rates) to revenue metrics (MRR, ARR, closed deals). The tools that track these live in different systems, and the lag between a marketing touchpoint and closed revenue can be weeks or months.
The practical solution for most teams: capture lead source at the form-submission level using first-party attribution, store it in the CRM on the contact record, and build a closed-won-by-source report in your CRM. This creates a revenue-by-channel view that connects your marketing dashboard to your revenue dashboard without requiring a data warehouse.
Tools like Sales Provenance automate the UTM capture and CRM write steps, so the source data is available on every lead for inclusion in CRM reports and dashboards.
Getting Started: A Practical Dashboard Build Sequence
- Define the audience and the 3-5 questions the dashboard must answer. Write them down before opening any tool.
- Identify the 8-12 metrics that answer those questions. Confirm each one has a reliable automated data source.
- Choose a tool based on your data sources and team. Looker Studio if your stack is mostly Google. Databox if you want mobile access and pre-built templates. HubSpot dashboards if you live in HubSpot.
- Connect the data sources and build the layout. Organize by decision, not by data source. Add targets and prior period comparisons to every primary metric.
- Share with the target audience and get feedback. The first version will need adjustment. Build iteration time into the launch plan.
- Assign an owner and a refresh cadence. How often will this be reviewed? Weekly? Daily? Who is responsible for keeping it accurate?
A working KPI dashboard is not a project that gets finished. It is infrastructure that evolves as your business changes and as you learn which metrics actually predict the outcomes you care about.