Your website report shows visits and conversions. Your calendar tells a different story. Some inquiries become work, some are a poor fit, and others are still waiting for a decision.
Review the requests your business actually received, then connect each one to what happened next. That gives you a more useful question than “Did traffic go up?”
For an Arizona service business, a short weekly review can reveal unclear service information, missed follow-up, or a measurement problem. It does not need to become another dashboard nobody opens.
Start with the inquiries your team received
Bring together the form records, relevant calls, and booking requests your team actually handled. Use one agreed place to review them, whether that is a spreadsheet or an existing customer system.
Define an inquiry before counting it. A request for an estimate is different from a phone-link tap. A customer calling again about the same estimate usually belongs to the existing inquiry record.
Keep spam, tests, and duplicate submissions identifiable. Exclude them consistently from your genuine-inquiry count, while preserving enough information to investigate a broken form or repeated spam.
A returning customer asking about a new job can be a new inquiry. The point is to count the work being discussed consistently, rather than treating every message as another potential customer.
Our tracking checks before advertising explain how to verify the event behind a conversion count. Here, start with the business record and use the analytics report as supporting evidence.
Agree on what each outcome means
Choose a few labels that reflect how you sell. Define them with the person who answers inquiries, not just the person preparing the report.
“Inquiry received” means you have a genuine request. “Suitable opportunity” might mean the service, location, and timing fit what you offer. “Booked work” needs a clear confirmation rule.
An appointment to discuss a job is not necessarily a sale. An accepted estimate is not necessarily completed work or money received. Keep those distinctions if they matter to your business.
- Received
- Genuine requests, with duplicates and tests handled consistently.
- Suitable
- Requests that fit the service, area, and timing you can offer.
- Booked
- Work confirmed under your agreed booking rule.
- Still open
- Requests awaiting a decision, with the next step recorded.
Do not add these stage counts together as if they were separate people. One inquiry may have progressed through several stages.
Record a useful reason when work does not proceed. “Outside service area,” “Date unavailable,” and “Customer chose another provider” suggest different next steps. Keep “Reason unknown” when you do not know; guessing makes the report less useful.
Keep customer details in your restricted working system. A summary for a broader meeting can use totals and reasons without copying names, phone numbers, or message contents.
Separate this week's activity from inquiry outcomes
There are two useful views of the same work.
The first asks what happened during the week: new inquiries received, estimates sent, and jobs booked. This helps you understand workload.
The second follows inquiries that arrived during a particular period and asks what became of them. This helps you understand how requests progress.
Imagine an inquiry arriving at the end of one week and becoming a booking the following week. Dividing the second week's bookings by its new inquiries mixes different groups.
What arrived, moved forward, or booked during these dates?
Useful for workloadWhat happened to the requests received during the chosen period?
Useful for progressA booking this week can belong to an inquiry from an earlier week.
For an inquiry-to-booking rate, divide the number of inquiries that became booked work by the total inquiries in that same group. State the review date, and keep undecided requests visible. A group with many open estimates is not finished yet.
Compare similar review windows and note how long requests have had to develop. A recent group has had less time to book than an older one. With only a few inquiries, read the actual reasons before treating a percentage change as a pattern.
Use source reports without forcing every record to match
Where available, keep the recorded campaign or source alongside the inquiry. If you also ask customers how they found you, store their answer separately. A remembered recommendation and a tracked website visit can describe different parts of the same journey.
Check which report you are comparing. Google Analytics' Traffic acquisition report uses session-based source dimensions and covers new and returning visitors. Its User acquisition report answers a different question about new users. Google explains the Traffic acquisition report.
Ask your reporting partner to label the date range, source field, and measured event. “Google leads” is too vague if one report counts sessions and another counts submitted requests.
Attribution is the assignment of credit to interactions along a customer's path. Google Analytics uses attribution models to decide how that credit is allocated. Treat it as a reporting method, not a complete account of why a customer chose you. Read Google's attribution overview.
Leave an unknown source unknown until you have evidence. Do not assign it to the channel you hope is working or quietly spread it across other sources.
Also allow for reporting delay. Google says Analytics processing can take 24–48 hours, during which report data may change. Record when you pulled the report and avoid treating an unfinished day as a final comparison. That window is not a guarantee that every figure is permanently settled. See Google's data-freshness guidance.
Find the problem behind the number
Use the review to choose something concrete to investigate.
Imagine a Phoenix cleaning business receiving repeated requests for a service it does not offer. The next step may be clearer service wording, not more traffic. Our service-page guide can help with that explanation.
If suitable inquiries stall before anyone replies, review the handoff. Our lead follow-up guide covers responsibility, reminders, and recovery when a connection fails.
If the team's records look steady but the analytics count changes suddenly, check measurement before judging the offer. Confirm what changed in the form, consent setup, event definition, or reporting view.
If the team cannot take the requested dates, record capacity as the reason. A website cannot create availability your business does not have.
Look at the actual requests with the person handling them. A total can point to a question; it rarely answers it on its own.
End the review with one clear action
Write a short note: what you observed, what you think might explain it, what you will check, and who owns the next step.
Keep observation and explanation separate. “Several requests mention a service we do not offer” is an observation. “Our opening headline is causing it” is something to investigate.
Change the relevant part of the page or process, record the date, and review the affected inquiries afterward. Note other changes too, such as a new offer, closed dates, or a different campaign. Otherwise, you may credit one change for something else.
Keep checking older open inquiries until they have an outcome. If the weekly review only looks at newly received requests, unfinished work disappears from view.
A useful review ends with a decision your team can carry out. It should help you improve the website, the handoff, or the report itself.
Our technical SEO and analytics service can help connect the website's measurements with useful business questions. Tell Brandon what your reports still cannot explain.
