SolutionsPricingWorkshopsBlogNewsAboutLet's Talk

How to Build a Working BI Dashboard for Your Business

Practical guide to setting up a BI dashboard: what to connect, how to choose metrics, and how long it takes. Includes comparison table of dashboard types.

AutomushPublished 9 min read

What is a BI dashboard and why does a business need one

A BI (Business Intelligence) dashboard is a single screen that displays the metrics that matter to your business in real time or with automatic updates. Instead of opening three files, logging into the CRM, and asking accounting for a report, the business owner sees the current state at a glance: how many leads came in this week, how many quotes went out, what the sales volume is, and where there are delays.

We build dashboards that connect existing systems - CRM, Google Sheets, accounting systems like Morning (Green Invoice), and internal databases. The dashboard does not replace these systems; it pulls data from them and displays it in one place. When data updates at the source, the dashboard updates automatically - usually every hour, but you can set a different frequency as needed.

A good dashboard answers questions the business owner asks every day. If you open the same file every morning to check the same number, that’s a metric that should be on the dashboard. If you wait for someone to prepare a report once a week, that’s a metric that should be on the dashboard.

Which data should appear on a BI dashboard

An effective BI dashboard displays between five and ten core metrics, no more. More than that makes the screen crowded and makes decision-making harder. We choose metrics based on three criteria: does the business owner check it at least once a week, does the number influence a concrete decision, and can it be pulled automatically from an existing system.

In a business that sells services, the dashboard typically includes: number of new leads (by source - website form, WhatsApp, phone), conversion rate from lead to quote, conversion rate from quote to closed deal, monthly sales volume versus target, and average response time to leads. If there is inventory or open projects, we add metrics that fit those.

In a business that works with subscribers or repeat customers, it’s relevant to add: number of active customers, churn rate, monthly recurring revenue (MRR), and number of open support tickets. These metrics help identify problems before they become crises - for example, an increase in support tickets can point to a problem with the product or work process.

Every metric on the dashboard needs to be clearly defined. “Number of leads” could be every inquiry, only inquiries that passed initial screening, or only inquiries entered into the CRM. We define this during the planning stage, then build the connection accordingly. If the definition changes after a month, we can change the logic in the workflow that feeds the dashboard.

How to connect the dashboard to existing systems

A BI dashboard receives data from systems already working in the business. The connection is made in three main ways: direct API, automatic export to a shared sheet, or webhook that updates the dashboard when an event occurs in the source system.

When the system supports an API (like most CRM systems, accounting systems, and the WhatsApp Cloud API), we build a workflow in n8n that pulls data on a fixed schedule - usually every hour. The workflow asks the system “what changed since the last hour,” processes the response, and updates the dashboard. If the API is limited in the number of calls per day, we schedule the workflow accordingly - for example, four times a day instead of every hour.

When the system does not support an API but allows automatic export (for example, a report sent by email every morning or a file that updates in Google Drive), we build a workflow that reads the file, parses it, and updates the dashboard. This is less elegant than an API but works with the same reliability.

When immediate update is needed (for example, an alert when a new lead comes in), we use a webhook - the source system sends a message to our workflow the moment the event occurs, and the workflow updates the dashboard within seconds. Not every system supports this, and not every metric needs immediate update, but when it’s relevant it saves a lot of time.

We publish the final dashboard in a BI tool (like Power BI or Google Data Studio) or in an interactive sheet that updates automatically. The choice depends on what the business already uses and the complexity of the display. Sometimes a simple sheet with clear graphs works better than a sophisticated dashboard that requires training.

How long does it take to set up a BI dashboard for a business

Setting up a simple BI dashboard - five to seven metrics, two data sources, display in a sheet - takes between one and two weeks. This includes a planning meeting, building the connections, testing, and initial fixes. The longest stage is usually not the building but the clarification - where the data sits, how it’s defined, and who is responsible for each system.

A more complex dashboard - ten or more metrics, four or more data sources, display in a BI tool with breakdowns and filters - takes between three and five weeks. Here we also include decisions about permissions (who sees what), handling edge cases (what happens when a system doesn’t return data), and sometimes also a data processing layer that unifies numbers from different sources.

After the initial setup, the dashboard requires minimal maintenance. If a system changes its API, we need to update the connection. If we add a new metric, it usually takes a day of work. If we change the definition of an existing metric, it takes an hour to two hours. We monitor the workflows through a monitoring board, so if something breaks we see it before the business owner notices.

The longest time we’ve seen for setting up a dashboard was eight weeks, in a business with seven data sources that hadn’t communicated with each other before, including an old system without an API that required a custom solution. The shortest time was four days, in a business where all the data was already in Google Sheets and they just wanted an organized display with graphs.

What is the difference between an operational dashboard and a strategic dashboard

An operational dashboard answers the question “what’s happening now” and updates at least once a day, usually every hour. It displays metrics that change quickly and require immediate response: how many leads came in this morning, how many inquiries are waiting for handling, what’s the status of open orders. The operations manager or business owner looks at it several times a day and makes short-term decisions - who to pass a lead to, which task to push, where there’s a bottleneck.

A strategic dashboard answers the question “where are we going” and updates once a day or once a week. It displays trends, comparisons to previous periods, and cumulative metrics: monthly growth in sales, average conversion rate in the last quarter, customer acquisition cost versus customer lifetime value. Management looks at it in a weekly or monthly status meeting and makes long-term decisions - whether to increase the advertising budget, whether to change the pricing structure, which product to develop next.

We build both types on the same infrastructure - the same workflows, the same data sources - but the display and logic are different. The operational dashboard displays raw numbers and real-time status. The strategic dashboard calculates averages, displays trend graphs, and compares to previous periods or targets.

In small and medium businesses, we usually start with one operational dashboard that serves everyone, and after a few months add a separate strategic view. In larger businesses, sometimes there’s an operational dashboard for each department and one strategic dashboard for management. The technology is identical; the difference is in what we choose to display and at what frequency.

Dashboard Type Update Frequency Who Uses Type of Decisions
Operational Hourly or real-time Operations manager, business owner Immediate response, task allocation
Strategic Daily or weekly Management, owners Planning, budget, business direction
Hybrid Hourly with trend view Everyone Both operations and planning

What happens when a system doesn’t return data

When a source system is unavailable or doesn’t return data, the BI dashboard needs to continue working and display the last information it successfully pulled. We build the workflows with error handling: if an API call fails, the workflow tries again after a minute. If the second attempt also fails, the workflow saves the last data it successfully pulled and sends an alert to whoever is responsible.

The dashboard itself displays next to each metric the time of the last update. If the metric updated an hour ago, everything is fine. If the metric hasn’t updated in three hours, the user sees that and knows the number isn’t current. This is better than a dashboard that displays an error or a dashboard that displays an old number without saying it’s old.

If the source system changes its data structure (for example, a field whose name changed or a date format that changed), the workflow may stop working. Therefore we build checks that run the workflow every day and notify if something changed. In most cases, such a fix takes an hour to two hours - we identify the change, update the logic in the workflow, and check that everything is working again.

When working with critical systems, we build a backup: the workflow saves a copy of the data in a sheet or separate database, so even if the source system disappears completely, there’s history. This happened once when a CRM system underwent an upgrade and the data wasn’t available for two days - the dashboard continued to display the data up to the upgrade day, and users knew it was a temporary situation.

How to choose a tool for displaying the dashboard

The tool for displaying a BI dashboard depends on three things: how complex the display is, who needs to see it, and what already exists in the business. If the dashboard is simple - a table with numbers and a few graphs - a Google Sheets sheet that updates automatically works excellently. If we need breakdowns, filters, and different views for different users, a dedicated BI tool like Power BI or Google Data Studio is more suitable.

We prefer to start simple. A first dashboard in a shared sheet gives results quickly, costs less, and is easy to change. After a month or two, when it’s clear what works and what doesn’t, we can upgrade to a more sophisticated tool. The transition is simple because the workflows that pull the data remain the same - we just change the destination they write to.

If the business already has a license for Power BI or another BI tool, we build the dashboard there. There’s no point adding a new tool when there’s an existing tool that does the job. If there’s no existing tool and the dashboard is simple, we start with a sheet. If there’s no existing tool and the dashboard is complex, we evaluate Google Data Studio (free) or Power BI (paid, but with more advanced capabilities).

The most important tool is not the one that displays the dashboard but the workflows that feed it. If the workflows are built correctly, we can switch the display tool within a day of work. If the workflows aren’t built correctly, even the best tool won’t help. Therefore we invest most of the effort in correct building of the connections and logic, and less in choosing the display tool.

More information about building reporting and BI systems for businesses is available on our solutions page. If you’re considering adding an automation layer that feeds the dashboard automatically from leads coming through WhatsApp or forms, it’s worth reading our guide on CRM lead automation.

Frequently asked questions

Can you build a BI dashboard without changing existing systems?

Yes. A BI dashboard pulls data from existing systems without changing them. We build workflows that connect to the system's API or read files it exports, and transfer the data to the dashboard. The source systems continue to work exactly as before, and the dashboard simply displays the information already sitting in them.

What happens if I want to add a new metric to the dashboard after it's already working?

Adding a new metric to an existing dashboard usually takes a day of work. We build a new workflow that pulls the data from the appropriate system, connect it to the dashboard, and check that it works correctly. If the new metric comes from a system that's already connected, it can take only a few hours. If we need to connect a completely new system, it can take two days.

Can the dashboard send alerts when a metric goes outside a certain range?

Yes. We build alerts as part of the workflow that feeds the dashboard. For example, if the daily number of leads drops below a certain threshold, the workflow sends a message on WhatsApp or email to the manager. The alert can be immediate (the moment the metric goes outside range) or summarized (a daily report of all metrics that went outside range). This works especially well for operational metrics that require quick response.

How much does it cost to set up a BI dashboard for a small business?

The cost depends on the number of data sources and the complexity of the display. A simple dashboard with two to three sources and five metrics usually costs between $2,000 and $4,000 for one-time setup, with no fixed monthly costs unless you choose ongoing maintenance. A more complex dashboard with five sources and ten metrics costs between $5,000 and $9,000. This includes planning, building, testing, and training.

Interested?reporting and business intelligence solutions

From the newsroom

All news

Which of your processes could run on its own?

A free 30-minute discovery call, no commitment. We map your processes together and tell you honestly whether and how automation can help.

Try a call with agent now