Reach is up. Clicks are up. The report is so green it could apply for agricultural funding.
Then someone asks, “Should we spend more next month?”
You scroll back to the top.
A useful marketing dashboard helps you make a decision. It connects the campaign's job to evidence, shows the limits of that evidence and identifies what should happen next. Another chart will not fix uncertainty about what counts as a lead.
Here is how to turn a monthly reporting meeting into a working review, with a calculation you can borrow for your own campaigns.
Decide what the number means before you celebrate it
Suppose a catering company reports one hundred leads. Are those contact-button clicks, submitted forms or distinct genuine requests?
The same person might submit a form, call and follow up on WhatsApp. Counting each interaction as a new prospect can make marketing look unusually good at multiplying people.
Write working definitions with the people who handle enquiries:
| Metric | Example definition |
|---|---|
| Received enquiry | A distinct genuine request, with contacts about the same job connected. |
| Qualified enquiry | A request the business can realistically serve under its agreed criteria. |
| Booking | A confirmed order under the business's stated rule. |
| Media cost per qualified enquiry | Media spend divided by qualified enquiries attributed using the stated method. |
The definitions must fit your operation. Keep missing qualification information unresolved until checked. Define how cancellations and refunds affect reported bookings or revenue.
GA4 allows an event to be marked as a key event because the business considers it important. That label does not transform a WhatsApp click into a conversation or a form into a qualified prospect.
Write cost definitions too. “Acquisition cost” should not silently mean media spend alone if the decision also needs production, agency, and sales costs.
The cheapest lead can be the more expensive option
Imagine that caterer comparing two campaigns. These figures are invented to demonstrate a decision, not client results or industry targets.
Assume the campaigns share a period, cost basis, and enquiry definitions. All requests have been deduplicated and assessed for qualification.
| Measure | Campaign A | Campaign B |
|---|---|---|
| Media spend | KES 20,000 | KES 20,000 |
| Received enquiries | 100 | 40 |
| Media cost per enquiry | KES 200 | KES 500 |
| Qualified enquiries | 10 | 20 |
| Media cost per qualified enquiry | KES 2,000 | KES 1,000 |
On raw enquiry cost, A looks cheaper. On qualified enquiry cost, B is cheaper and produced twice as many suitable opportunities.
That changes the conversation. It doesn’t yet establish which campaign is more profitable.
For A, inspect the unsuitable requests. Did the ad imply a service area or event size the caterer cannot support? Was the offer unclear? Give the campaign manager a specific question to investigate, rather than an instruction to “optimise performance”.
For B, inspect the suitable requests and whether sales progressed them. If quotes are waiting on someone internally, raising the budget may simply enlarge the queue.
A useful report makes both pieces of work visible.
Follow the same enquiries through to an outcome
Extend the fictional example after enough time for the defined groups to conclude:
| Measure for those same enquiry groups | Campaign A | Campaign B |
|---|---|---|
| Confirmed bookings | 2 | 4 |
| Media cost per booking | KES 10,000 | KES 5,000 |
| Assumed contribution per booking before marketing | KES 8,000 | KES 4,000 |
| Total contribution before marketing | KES 16,000 | KES 16,000 |
| Contribution after media spend only | -KES 4,000 | -KES 4,000 |
B produced more bookings at a lower media cost each. Both leave the same negative contribution after media in this simplified scenario. The booking values and delivery costs changed the story.
The assumed contribution is revenue less the relevant variable fulfilment costs, before marketing. Fixed overhead, production and agency costs are excluded here. Include applicable costs when assessing the actual business.
The practical question becomes whether the offer economics, targeting or acquisition costs can improve enough to justify continuation. “More leads” is no longer an adequate answer.
If outcomes remain open, leave them open. Do not compare this month's new enquiries with this month's sales when those sales came from older conversations. Follow the same enquiry group over time and show what remains undecided.
Bring Attention Lasso a report and the decision it still cannot help you make. We can identify whether the gap sits in tracking, lead records, financial definitions, or interpretation.
Do not count the same customer three times
A customer sees a social ad, searches your name later, and submits a form. More than one platform may claim a conversion under its own rules. Adding those totals can overstate the number of customers.
Google's attribution guidance explains how credit is assigned to recorded interactions, including the effect of attribution models and lookback windows. Compare compatible definitions before comparing totals.
Keep business outcomes tied to unique underlying records where possible. Use platform reports for their delivery and optimisation questions, and business records for the orders or bookings actually confirmed. Neither view captures every influence on a buying decision.
Attribution also differs from incrementality. It asks how to assign credit; incrementality asks what happened because the marketing ran. Google's Conversion Lift guidance describes comparing exposed and control groups. Eligibility and study design matter, so do not assume every account can run that tool.
Without a suitable test, describe an attributed or associated outcome honestly. You can make a considered operating decision without pretending to have proved a causal effect.
Check the data before inventing the explanation
If enquiries fall, first confirm that the form works. A persuasive explanation about consumer sentiment will not repair an email integration.
Before the review, check:
- The same definitions, currency, time zone, and cost inclusions.
- Working forms, events and enquiry receipt.
- Duplicate, test and spam handling.
- Comparable periods and a record of campaign or site changes.
- How much of the recent data remains incomplete.
Google's data-freshness guidance says GA4 processing can take 24 to 48 hours, with report figures changing during processing. Date the extract and label provisional information.
Keep small counts beside percentages. A change from one booking to two is a 100% increase and one additional booking. Both descriptions are correct; only one makes the scale immediately obvious.
AI commentary needs evidence too. “Clicks fell” is an observation. “Customers lost interest” is an explanation requiring investigation. Ask the writer or system to separate what the data shows from what it thinks caused it.
Replace the slide tour with a decision note
Keep detailed charts available. Open the meeting with a short note people can act on:
Goal: Generate suitable catering enquiries at a sustainable cost.
Finding: Campaign B generated more qualified enquiries and bookings, but the illustrative contribution calculation is negative for both campaigns after media spend.
Next action: The campaign manager will inspect unsuitable requests from A. Sales and operations will check actual booking values and fulfilment costs before any budget increase.
Review: Named owners will bring the updated figures to the next agreed budget meeting. Any budget test needs a spend limit and a stopping rule.
*That note interprets our invented figures. It is not advice to change a live account.
For your own review, name the decision, evidence, owner, and review date. Start the next meeting by checking the previous action. If nobody knows whether it happened, the reporting process has another useful place to improve.
Frequently asked questions
Which marketing metrics should I put on my dashboard?
Choose the metrics needed for the decision. Lead-generation reviews need enquiry quality, cost and outcomes. Creative reviews may need viewing behaviour and response. Keep delivery indicators for diagnosis, but do not expect impressions alone to settle a sales-budget decision.
Are impressions and engagement vanity metrics?
Not automatically. They can help assess distribution and creative response when those are the campaign's jobs. They become misleading when presented as proof of an unrelated outcome, such as profitable sales. Explain what each measure can tell you.
Why do Google, Meta and my CRM show different conversions?
They can use different event definitions, attribution rules, time settings, and counting methods. Check those before assuming one is broken. Reconcile actual customers or orders through underlying records, and do not add platform totals as though every conversion is unique.
What is the difference between a lead and a qualified lead?
A lead signals interest; a qualified lead meets agreed service-fit criteria. Define both for your business. A genuine catering enquiry outside your delivery area is different from one you can fulfil, even though both people submitted a form.
Can I measure marketing ROI from revenue alone?
No. Revenue does not tell you what the sales cost to fulfil or acquire. Include relevant costs and distinguish attributed sales from incremental impact. A revenue-to-ad-spend ratio can be useful, but it is not a complete profit or ROI calculation.
How often should I review the dashboard?
Review operational failures promptly and business performance on a rhythm suited to the buying cycle. A live display does not require constant budget changes. Allow enough complete evidence for the decision, then give actions an owner and review date.
Make the next report earn the meeting
Choose one decision your current report should support. If the necessary definitions or outcomes are missing, that is the brief for improving it.
Share that decision and a sample report with Attention Lasso. We can help connect what you measure to the next useful action.
Keep pulling the thread
Related reading
Daily AI video ads: a production system or an expensive content treadmill?
Read next
How to choose a GEO agency without buying a very expensive screenshot
Read next