A BI (business intelligence) dashboard pulls data from the tools you already use, such as your CRM, ad accounts, accounting software and website, into one place, applies consistent definitions, and shows the handful of numbers that should change what you do next week. A report tells you what happened. A good dashboard tells you where to act.
Most businesses do not lack data. They lack one version of it that everyone trusts. This explainer covers how dashboards work under the hood, which metrics deserve screen space, and what it costs to get there.
What is the difference between reporting and BI?
Reporting answers fixed questions on a schedule; BI lets you explore why the numbers moved. You usually need both, but they are built differently.
- Reporting: a monthly sales summary, a weekly ad-spend sheet, an end-of-quarter profit statement. Same format every time, often emailed as a PDF or spreadsheet.
- Business intelligence: an interactive view where you can filter by region, campaign or salesperson and drill from a total down to individual records.
A small business often starts with reporting and grows into BI once the questions become "why did enquiries drop in one city?" rather than "how many enquiries did we get?"
How does a BI dashboard actually work?
Data flows through four layers: sources, storage, modelling and display. Problems in a dashboard almost always trace back to the first three, not the charts.
| Layer | What happens | Example tools or methods |
|---|---|---|
| Sources | Raw data lives in operational tools | CRM, Meta and Google ad accounts, accounting software, website forms, e-commerce platform |
| Extraction and storage | Data is copied on a schedule into one database or warehouse | API connections, scheduled syncs, a cloud database or data warehouse |
| Modelling | Raw tables are cleaned, joined and given agreed definitions | SQL views or a transformation layer that defines "lead", "customer", "revenue" |
| Display | Charts, tables and filters people actually look at | Looker Studio, Power BI, Metabase, Tableau, or a custom dashboard inside your own software |
The modelling layer is where trust is won or lost. If marketing counts a lead when a form is submitted and sales counts it when someone qualifies it, the dashboard will show two different truths. Agree the definitions in writing before anyone draws a chart.
Getting data out of each source relies on APIs. If that term is fuzzy, our plain-English guide to APIs explains it in a few minutes.
Which metrics should a business dashboard show?
Show the few numbers that lead to a decision, and hide everything that only makes people feel busy. A useful test: if this number doubled or halved, would anyone do something different?
A practical starting set for a service business:
- Cost per enquiry by channel and campaign. Tells you where to move budget.
- Enquiry-to-sale conversion rate by source. Cheap leads that never buy are not cheap.
- Average time to first response. Slow replies quietly kill conversion.
- Pipeline value by stage. Shows whether next month's revenue is already at risk.
- Revenue and margin by service line. Reveals which work actually pays.
- Overdue invoices. Cash flow problems show up here first.
Leave out raw impressions, page views and follower counts unless they connect directly to one of the above.
What does a decision-ready dashboard look like in practice?
It puts comparable numbers side by side so the outlier is obvious. Here is an example from the Meta ad accounts we manage, using lifetime figures under Meta's default attribution (7-day click, 1-day view).
| Account type | Result | Average cost per result |
|---|---|---|
| Clothing brand | 2,564 enquiries | ₹5.43 per conversation |
| Home-improvement (awnings) business | 153 conversations | ₹2.72 each |
| Wedding photography (5 accounts combined) | 1,130 enquiries | ₹58.55 per enquiry |
| Interiors business | 2 conversations | ₹423 each |
Seen together, the interiors account stands out immediately as a problem worth investigating, while wedding photography costs more per enquiry for reasons that make sense (higher-value, more considered purchases). A dashboard should make the question "what's different about this one?" impossible to miss. On a spreadsheet tab buried among twenty others, that same signal can sit unnoticed for weeks.
Context still matters. A ₹58 enquiry for a wedding booking can be far better value than a ₹5 enquiry for a low-priced item. That is why cost metrics belong next to conversion and revenue metrics, not on their own.
Should you use a BI tool or build a custom dashboard?
Use an off-the-shelf BI tool when your data sources have ready-made connectors; build custom when the dashboard needs to live inside your own software or act on the data. Many businesses use both.
Off-the-shelf BI tools suit you when:
- Your sources are mainstream platforms with existing connectors.
- The audience is internal and comfortable with a separate login.
- You mostly need to look, filter and export.
A custom dashboard makes sense when:
- It should sit inside your CRM or client portal, so people see numbers where they work.
- Different users need strictly different views, enforced by permissions.
- Clicking a number should trigger an action, such as reassigning a lead or sending a reminder.
Our guide to building a sales pipeline dashboard your team will actually use covers the design side in more depth.
How much does a BI or reporting setup cost?
It ranges from close to free for a single-source report to the price of a small software project for a multi-source, custom dashboard. These are typical market ranges, and scope drives the final figure.
| Setup | India (₹) | Australia (AUD) | Global (USD) |
|---|---|---|---|
| Free or low-cost BI tool connected to one or two sources, set up in-house | Tool subscription only | Tool subscription only | Tool subscription only |
| Professionally built dashboards with a small data pipeline | ₹1.5–5 lakh | $10k–30k | $6k–20k |
| Custom reporting module inside your own platform, several integrations | ₹4–12 lakh | $25k–80k | $15k–50k |
Budget roughly 15–25% of the build cost per year for maintenance. Source platforms change their APIs, and every change can break a sync.
What usually goes wrong with dashboards?
Most dashboards fail from neglect, not bad design. The common failure modes are predictable.
- Nobody owns it. Assign one person to check data freshness each week.
- Definitions drift. Someone changes a CRM stage name and conversion rates jump overnight.
- Too many charts. A dashboard with forty tiles gets ignored. Start with six.
- Stale data. If figures update monthly but decisions happen daily, people go back to asking colleagues.
- No link to action. Put the dashboard review on the agenda of an actual weekly meeting.
Accounting data is a frequent weak spot. Revenue and margin figures depend on a reliable sync with your accounting package, which we cover in accounting software integrations for Xero, QuickBooks and Tally. For where dashboards fit in your wider stack, see our guide to business systems that scale.
Frequently asked questions
Do small businesses really need a BI dashboard?
Not always. If all your data lives in one tool with decent built-in reports, use those. A dashboard earns its place once you are combining two or more sources by hand every week to answer the same questions.
How often should a dashboard update?
Match the refresh rate to how often you make decisions. Ad spend and lead flow usually benefit from daily or hourly updates, while profitability by service line can update weekly or monthly.
What is a data warehouse, and do I need one?
A data warehouse is a database designed for analysis, holding copies of data from many tools in one place. Small setups can skip it and connect a BI tool directly to sources, but once you have several sources or large volumes, a warehouse makes reports faster and more reliable.
Can a dashboard show data from Meta ads and my CRM together?
Yes. Both offer APIs, so ad spend can be joined with CRM outcomes to show cost per sale, not just cost per lead. The key is matching records reliably, usually through the lead ID or campaign details captured at form submission.
Who should own the dashboard inside the business?
Pick someone close to the decisions, often an operations or finance lead, rather than whoever built it. They check freshness, flag broken numbers and decide what gets added or removed.
Turn your numbers into a weekly habit
If your Monday meeting starts with someone stitching spreadsheets together, a focused dashboard can give that time back. Ask us for a fixed-price scope covering your sources, the metrics that matter and where the dashboard should live.