Build a Marketing Dashboard in Data Studio

Choose the business decisions
Define measures and sources
Build a small verified report
Check calculations and filters
Share with appropriate access

What should the dashboard help someone decide?

Begin with a short list of recurring questions. An owner might need to know which campaigns generate suitable inquiries, whether the receiving team is keeping up and whether current spending matches the approved plan. A dashboard answering those questions is more useful than a collection of every metric available in a connector.

Write the decision beside each proposed chart. If a graph changes, what would the viewer investigate or do differently? A chart with no clear use may belong in a deeper analysis rather than on the main page. Keep a small executive view and provide detail where it helps explain the result.

Define the audience. A campaign specialist may need account-level detail, while an owner needs an understandable summary. Customer names, email addresses and private notes usually do not belong in a broad marketing report. Decide what viewers need before connecting a detailed operational dataset.

Which definitions should be agreed before connecting data?

Create a measure dictionary with the name, meaning, source, date basis and any exclusion. For example, inquiry could mean a successfully received request, while qualified inquiry means the team confirmed it fits the service scope. A phone click, answered call and suitable conversation should have different labels.

Record the time zone and currency for each source. A campaign’s activity date and a sale’s completion date answer different questions. If the dashboard compares them, explain that relationship instead of assuming matching calendar labels make the records equivalent.

Choose a consistent cost definition. Advertising media spend differs from a total that includes management, production or supporting software. Both can be useful, but cost-per-lead calculations should state which costs and which leads they include. A visual cannot resolve an ambiguous business definition.

How do you build the first report?

Google’s documented workflow starts by creating a report and adding a new or existing data source. Use an account authorized for the task, select the intended connector and verify the specific property, account or dataset before adding charts. Similar account names make it easy to connect the wrong source.

Start with one source and a small number of components: a date control, an accurately labeled total, a trend and a table that explains the total. Confirm field types and aggregation. A date stored as text, a numeric identifier treated as a measure or an unexpected default date range can produce confusing output.

Check the first report against the source system using the same date range, filters and definitions. Resolve differences before decorating the page or connecting additional datasets. A reusable template can save layout work, but its formulas and filters still require review for your data.

What would a useful small-business example look like?

Imagine a hypothetical commercial laundry service evaluating two campaigns. During a defined period, campaign A has $900 of media spend and nine qualified inquiries, while campaign B has $600 and three. Under those invented assumptions, their media cost per qualified inquiry is $100 and $200 respectively. This is arithmetic illustration, not a Dappr result or a benchmark.

The combined figure is $1,500 divided by twelve qualified inquiries, or $125. Averaging the two campaign rates would produce $150, which answers the wrong aggregation question. The dashboard should calculate the overall ratio from the relevant totals, with a defined treatment when the denominator is zero.

The owner also needs context: whether the inquiries concerned services the company can deliver, whether the qualification criteria were consistent and whether the campaigns had comparable opportunities. The lower observed cost does not automatically make campaign A the right place for every additional dollar. The report should support investigation rather than substitute a calculation for business judgment.

When should you combine sources?

Combine data only when you can explain the relationship between the records. Google’s blend documentation describes joining information from multiple sources. The technical ability to join tables does not establish that their fields are suitable keys or that their rows represent the same level of detail.

Suppose an advertising table has one row per campaign per day, while an inquiry table has several records for the same campaign and day. A careless join can repeat the daily spend across those records and inflate the total. Agree on the intended level of detail and validate totals before and after the join.

When the relationship is uncertain, separate panels may be more honest than a forced combined table. Display observed campaign conversions, website requests and qualified opportunities with their own definitions. Do not label them a fully reconciled funnel until the data actually supports that connection.

Keep a small test dataset with known expected results for important calculations. A zero-inquiry day, an unmatched campaign and a duplicate key are useful cases to inspect. The test should reveal whether the dashboard behaves as intended, not simply whether it displays without an error message.

How do you explain freshness and missing data?

Data Studio can reuse previously fetched results, and freshness behavior depends on the source and configuration. Its refresh setting does not eliminate delays in the underlying advertising, analytics or operational system. A recently refreshed dashboard can still contain incomplete recent outcomes.

Show the reporting period and note known latency. Distinguish a true zero from unavailable data. An expired connector credential, quota limit or failed import should not silently become a zero-performance day that alarms the owner or changes a budget decision.

Google also distinguishes refreshing report data from refreshing data-source fields after the dataset’s structure changes. If someone adds or renames a spreadsheet column, review the source schema and affected charts. Treat a structural change as a report-maintenance task rather than assuming the dashboard will automatically interpret it correctly.

What should you check before sharing?

Review the credential mode and the intended recipients. Google explains that owner credentials can allow report viewers to see data without their own underlying dataset access, while viewer credentials require their own access. That choice affects what sharing the report exposes. Do not assume a private source remains private simply because the recipient cannot open it directly.

Use the organization’s approved access process and test the report as an intended viewer. Check whether they can see the correct information, whether controls behave as expected and whether any drill-down reveals unnecessary detail. Avoid publishing an unrestricted link to a report containing confidential campaign or customer information.

Keep editing rights limited to people responsible for maintaining the report. A viewer who needs to review performance does not automatically need permission to change formulas or sources. Document who owns the connections so the report can be maintained when an employee or vendor leaves.

How can the design make the numbers easier to interpret?

Use plain labels and include units. A number called conversions is not sufficiently clear if three sources use different definitions. Show comparison periods explicitly and explain whether a percentage is a rate, a relative change or a share of total. Do not make viewers infer the meaning from a color.

Keep important qualifications near the chart they affect. If recent outcomes are incomplete, place that note beside the relevant trend. If a cost figure excludes management fees, make the exclusion visible where the number is used. A distant methodology page is useful but should not carry all the burden of interpretation.

Use consistent colors and readable text, and check the report at the screen sizes the audience uses. The main page should be scannable without hiding essential detail. More charts do not necessarily create more insight; a concise view with reliable definitions is easier to discuss.

Who maintains the dashboard after launch?

Assign owners for the business definitions, data connections and report design. Schedule a review of source access, changed fields and reconciliation with the underlying systems. Record material changes to tracking, conversion definitions and campaign structure so later comparisons have context.

Collect the questions viewers ask during reviews. If a repeated question cannot be answered, decide whether it requires another chart, better data or a separate analysis. Do not add a calculated field that implies precision the source cannot support merely to satisfy a request for one more number.

Dappr can discuss the website and inquiry measurement behind your reporting needs. Bring the decisions you need to make, approved data sources and current definitions. Connector compatibility and dashboard scope should be confirmed before implementation; a reporting plan does not itself establish that every system is integrated.

Questions before you begin

Is Looker Studio still the current product name?

Google returned the product name to Data Studio in April 2026. Older Looker Studio documentation and links may redirect. Use current documentation when following setup instructions or reviewing available features.

Should I average the cost-per-lead figures from several campaigns?

For an overall cost per lead, calculate from the relevant combined cost and lead totals. A simple average of campaign ratios can misrepresent the overall result when campaign volumes differ. Keep definitions consistent.

Does refreshing a report guarantee real-time data?

No. The source system may have processing delays, and connector freshness or caching can also affect what appears. Document the relevant delays and avoid presenting incomplete recent data as final.

How to Build a Marketing Dashboard in Data Studio