Small-Business Marketing Metrics to Review Each Week

For a weekly small-business marketing review, connect visits, unique inquiries, qualified opportunities, sales, and clearly defined costs. Reach and likes can describe distribution or engagement, but they do not establish that customers were acquired. Start with consistent definitions and a record showing what happened after the initial contact.

Count unique opportunities, preserve source information where it is known, and follow the same group through the sales process. A cost or conversion comparison is meaningful only when its numerator, denominator, and period are understood.

Define the lead and sale before building a dashboard

Write the team’s lead definition. It might be a received form containing the necessary information or a conversation requesting an offered service. A support message from an existing customer is not automatically a new sales opportunity.

Separate a contact, a qualified opportunity, and a completed sale. Decide how repeated interactions map to an opportunity. A person using the form and then sending a message may create two interactions while representing one request.

Define which stage counts as a concluded sale for your business. Do not silently change between acceptance and payment to improve a week’s apparent result. If the definition needs revision, document it so earlier comparisons remain understandable.

Use a small dashboard with explicit questions

Add notes for interrupted campaigns, holidays, staffing changes, form edits, and measurement incidents. Those events may explain a variation better than declaring the channel good or bad from one total.

Measure actions according to what they mean

Google explains how to measure important actions as key events. Choose an event representing the intended stage, such as a completed form submission, rather than naming every button interaction a received lead.

Validate the implementation with an authorized, identified test. Confirm the event, receiving record, and absence of unintended repeated counts. Record which implementation produces the event so another person can review it later.

A website can measure a click opening a messaging route without proving that a conversation began. A qualified opportunity or sale needs evidence from the business process or an appropriate integration. Marking the click as a key event does not change that meaning.

Keep campaign naming consistent

The GA4 campaign-parameter documentation explains source, medium, and campaign information in tagged URLs. Establish a small naming convention before distributing links.

For an illustrative newsletter, the team might use source newsletter, medium email, and an agreed campaign identifier. Keep the same convention throughout that action and verify the destination.

Do not put a visitor’s name, phone number, or email address in campaign parameters. Where a reliable association exists, preserve useful source context in the commercial record. If the origin is unknown, leave it explicitly unknown instead of guessing a channel to make the report look complete.

Separate activity this week from a cohort’s outcome

Google’s lead acquisition report depends on implemented lead-generation events. It does not create a meaningful qualification stage automatically; the underlying events must represent the actual business process.

A proposal can take several weeks to close. Sales completed now may belong to older inquiries. Dividing those sales by only this week’s new leads can mix unrelated groups and produce a misleading rate.

Maintain two views: activity within the current period and the later progression of a defined entry group. A simple record can support both if it retains the entry date, source, status, and relevant closing date.

Use the same qualification and sale definitions across comparisons. If they differ between periods or channels, explain the limitation before presenting a ranking.

Use a transparent numerical example

Consider an illustrative campaign spending $600 and receiving 30 unique inquiries, of which 10 meet the business’s qualification criteria. Spend per inquiry is $20: $600 divided by 30. Spend per qualified inquiry is $60 because the denominator changes to 10.

Suppose two opportunities from that same group later become sales. Media spend per sale is $300 in this simplified example. It is not automatically total customer acquisition cost: staff time, tools, and other costs may be excluded.

These dollar amounts are hypothetical US-facing examples, not a currency conversion, a market benchmark, or a promised result. The point is to make the calculation’s scope visible.

The example also does not establish profit. Revenue, refunds, taxes, and delivery costs have not been supplied. Do not let a precise division create a conclusion the available inputs cannot support.

When visits rise but inquiries do not

First check measurement and the contact route. Open the page on mobile and use an authorized test to confirm that the intended inquiry reaches its owner. Increasing spend before excluding a failed form adds traffic to an uncertain process.

Then examine whether the content and offer fit the visitors’ intent. An informational article can be useful without being designed for immediate quotation requests. Compare sources and destinations separately instead of expecting the same rate from every session.

Review the offer’s clarity, necessary information, and form effort. Choose one practical hypothesis and record the change. Altering the ad, page, pricing, and response process together makes causal interpretation more difficult.

When inquiries arrive but sales remain low

Use loss categories the team can apply: outside the target need, no response, timing, price, or another documented reason. Labeling everything did not close provides little basis for action.

Inspect an appropriately authorized sample of opportunities for duplicate entries and outdated statuses. A concluded sale left open in the CRM can distort the dashboard just as a duplicated lead can.

Different causes need different changes. Requests outside scope call for another investigation from qualified inquiries waiting too long for a response. Do not assign responsibility to a person based on a small unexplained fluctuation.

Keep uncertainty visible. A difference of a few opportunities may not support a broad strategic decision, especially when the observation periods or follow-up time differ.

Build a simple starting record

Useful fields include an internal opportunity identifier, entry date, known source, campaign, stage, loss reason, first-response date, sale date, and confirmed value. Keep personal details in the appropriate business system rather than analytics parameters.

Maintain spending by channel and period separately from the contact rows. If a campaign spent $600, do not sum another $600 for every related lead. Use the period total once in the corresponding calculation.

Document how repeated contacts are matched and when a support conversation becomes a separate commercial request. The rule should be consistent enough that two team members do not count the same history differently.

Make the weekly review produce an action

An illustrative routine is to check collection, update opportunity states, and then review totals together. The result can be a verified task: repair a form, follow up an unanswered proposal, or investigate a channel attracting requests outside scope.

For a meaningful deviation, record the evidence, hypothesis, action, owner, and review date. If no new evidence justifies a change, do not invent an intervention merely to give the meeting an outcome.

Save the version of the numbers used for the decision and note late updates. Platforms and CRM records can recognize stages at different times. Two screenshots are not necessarily equivalent datasets just because both describe a recent period.

Preserve a measurement boundary

Document which website, property, and event definitions a report covers. Separate regional operations or sites where the measurement plan requires it rather than mixing totals and trying to reconstruct the distinction later.

This guide does not install tags or change your Analytics setup. For a controlled improvement, first establish the missing definition or implementation, then verify its evidence before claiming the reporting gap is solved.

Keep an explicit data-cutoff time for a weekly review. If a salesperson adds an older closed deal after the meeting, update the record without pretending that the original dashboard already contained it. The change may affect a cohort outcome while leaving this week’s inquiry count unchanged. Recording both the correction and its timing prevents an apparent disagreement between two screenshots from becoming an unsupported claim that one tool is broken.

Also distinguish missing information from a measured zero. An unavailable spend total is not zero spend, and a period with no valid denominator cannot produce a meaningful rate through ordinary division. Mark the calculation unavailable and assign the missing input. This is especially useful when comparing a new channel with an established one: different tracking coverage can create an apparent performance difference before the business has comparable evidence. Repair the definition or collection gap before making a broad ranking from that comparison.

Frequently asked questions

Are likes useless?

No. They can describe an aspect of interaction or distribution. They do not prove a sale.

Which conversion rate should I target?

Define the action, audience, channel, period, and cohort first. A generic rate without those details is a weak operating target.

Must I automate everything?

No. Start with a small consistent record, then automate repeatable steps while preserving identity, deduplication, and source checks.

What makes the dashboard useful?

The team can explain each number, trace its origin, and identify the decision it supports.