Your website is getting visits, but the enquiries are not following. Before paying for more traffic or a redesign, check where people get stuck: finding the right service, understanding the offer, making contact or getting a response.
Start with the contact route. Can someone send a request, and can your team find it? Then check who is visiting and whether the page helps those people decide. A broken form, the wrong audience and an unassigned enquiry need different fixes.
A Perth example: visits without quote requests
Say you run a landscaping business in Perth. Your garden advice articles are attracting more readers, but quote requests have barely changed. This is a hypothetical example, not a client result.
First, separate those advice readers from people visiting your service page. Then check that a quote request actually reaches your team. If it does, look at whether the page explains the projects you take on, the areas you cover and what happens after someone enquires. Each check gives you a different place to improve.
1. Define what counts as a useful enquiry
Decide what you mean by an enquiry before judging the numbers. A phone-button click, a form submission, an answered call and a suitable project request are different outcomes.
Choose the result that matters to the page. For a quote request, that might mean a genuine person requesting a service you offer in an area you serve, with enough information for the team to take the next step. A useful educational article may have a different immediate purpose, such as helping someone understand whether that service fits their needs.
Record the stages separately:
| Stage | What it tells you | What it does not establish |
|---|---|---|
| Visit | A measured visit reached the website | The visitor needs your service |
| Contact interaction | Someone used a measured contact control | A call connected or a message arrived |
| Form event recorded | A form-related interaction or completion was recorded | The receiving system successfully stored and delivered it |
| Enquiry received | The business has a retrievable enquiry | It fits the offer or will become a sale |
| Enquiry assessed | The team established fit and a next step | The customer has agreed to buy |
This distinction can prevent the wrong repair. If genuine enquiries are arriving but nobody owns the follow-up, changing a button colour will not address that handover.
2. Check measurement and actual delivery
Before changing the page, confirm that the reported problem is real. Review the website's recorded actions alongside received enquiries, call records or booking records, using information the business is authorised to access.
GA4's enhanced measurement includes form interaction events such as form_start and form_submit. They record interactions with the form. To confirm an enquiry arrived, check the receiving inbox or customer record as well. A button-click event is an even earlier signal.
On a system you own or are authorised to test, follow one controlled enquiry through the full route. Confirm what the visitor sees, whether the receiving system stores it, where it is delivered and who can find it. Mark test records clearly so they do not inflate reporting or enter ordinary sales follow-up. Keep names, email addresses and message content out of Analytics event parameters.
Check what happens when a required answer is missing. Does the page explain how to correct it? Does a success message appear only after the intended submission has succeeded? Check the tracking alongside those states rather than assuming every interaction represents completion.
If people can contact you through phone, email or messaging, review those routes too. A click opening a messaging app is different from a message sent and received. Missing website events do not prove that no one contacted the business elsewhere.
Google's DebugView documentation explains how to inspect Analytics events during testing. Use it to check tracking alongside the receiving process.
3. Check who is visiting and what they came for
Look beyond the total visit count. Compare service pages with information pages, paid with organic visits, and relevant locations with places outside your scope. Our guide to what SEO involves explains why a service search and a general advice question need different pages.
For example, an article answering a general gardening question might attract people who want advice rather than a landscaping quote. That can be a useful content outcome, but it should not be judged as though every reader had searched for a local contractor.
For Google organic search, Search Console can show the queries and pages associated with impressions and clicks. Analytics can help examine what visitors do afterwards, where measurement is configured correctly. The two tools measure different things, so clicks and sessions should not be expected to match exactly.
Look for evidence of a mismatch:
- A promoted service is not explained on the page where visitors land.
- Much of the audience is outside the area the business can serve.
- A campaign suggests a simple quote, but the destination asks people to commit to a lengthy consultation.
- Educational pages account for most of the growth, while relevant service-page visits have barely changed.
Use those patterns to decide what to investigate. Compare the search data with the page, campaign and actual enquiries before changing how you attract visitors.
4. Check whether the page helps a suitable visitor decide
Ask someone unfamiliar with the business to read the relevant page and describe what they think is being offered. Observe where they hesitate. Agreement from a colleague alone does not establish that customers understand it.
Make the service and scope clear
A suitable visitor should be able to identify the service, intended customer, relevant location and next step. Explain exclusions or project requirements where they affect fit.
“Complete solutions tailored to your needs” offers less decision-making information than a truthful description of the projects you handle. Describe the work you actually do, with the details a prospective customer needs.
If price cannot be quoted without more information, explain what affects it and what the enquiry will establish. A realistic range may be useful where the business can support it. A made-up starting price can create the wrong expectations.
Show evidence people can assess
Relevant project examples, a clear process and genuine customer feedback can help someone evaluate a provider. Include enough context to understand what a result or testimonial actually describes. Display qualifications or memberships only when current, relevant and accurate.
The ACCC's guidance on online reviews emphasises genuine experiences and warns against misleading reviews. Use genuine work and feedback that relate to the service someone is considering.
Make the next action understandable
Describe what happens when someone acts. “Request a landscaping quote” communicates a different commitment from “Book a site assessment”. Use the label that reflects the actual process, with any material conditions made clear.
A campaign page may benefit from one clearly prioritised action. A wider service website can reasonably support different needs, such as calling, requesting a quote or checking project details. Prioritise the useful path without assuming every page must remove navigation or offer exactly one button.
Only offer an immediate response, free assessment or no-obligation appointment if your team can provide it.
5. Follow the contact journey on mobile and with a keyboard
Read and use the page in the conditions visitors are likely to encounter. On a phone, check whether navigation, notices, chat widgets or the on-screen keyboard cover the information and controls needed to enquire. Check the path through to completion, rather than stopping at the first screen.
Forms need clear labels and instructions. The W3C forms guidance explains labelling, and its instructions guidance explains why placeholder text alone is not a substitute for a label. A visitor should be able to identify the requested information after they have started typing.
For each question, ask what decision the answer supports. The GOV.UK Design System recommends knowing why each question is needed. That is useful design guidance, not a rule that every commercial form must have the same layout or number of fields.
Removing unnecessary questions may simplify the first contact. Removing information essential to route or assess the enquiry can create more work later. The aim is a proportionate request, with required and optional information clearly distinguished.
When an answer needs correcting, explain the problem and how to resolve it. W3C's notification guidance addresses understandable errors and confirmation of success. Preserve valid answers where possible so a correction does not force someone to start again. Test keyboard access as well as touch interaction. A quick check is useful but does not establish full accessibility compliance.
Performance deserves attention where it interferes with the task. Core Web Vitals cover loading, responsiveness and visual stability. Field and laboratory measurements help investigate different conditions. A good score is not proof that visitors understand the offer or that a conversion rate will improve by a particular amount.
6. Check what happens after the enquiry arrives
Once a genuine message is received, the question changes. Who owns it? Can they see the relevant context? Is there an agreed next step, and can the business tell whether that happened?
An acknowledgement can reassure someone that the request arrived. It should describe the real process and set an expectation the team can meet. Someone still needs to answer the question and take the next agreed step.
Keep a practical record of received enquiries, responsibility, status and outcome. This can begin with a simple controlled process. A CRM or automation may be useful where the work justifies it. Set ownership and expectations first. If the repeated handover is taking too much staff time, our comparison of AI agents and workflow automation explains the options and the decisions your team keeps.
Distinguish a page's ability to generate contact from the team's ability to assess and handle it. Both matter to the eventual result, but they need different changes and measurements.
Decide what to fix and how to evaluate it
Choose a change that addresses what you found. For the Perth landscaper, that could mean repairing delivery, clarifying project scope or assigning someone to each received request. Use this table as a decision aid:
| Finding | Appropriate next investigation or change | What to check afterwards |
|---|---|---|
| A valid request fails to arrive, or its notification is missing | Check storage, delivery and notification separately, then repair the failed step | The record can be retrieved and the responsible person knows it arrived |
| Visitors expect a different service or location | Review acquisition, destination and scope | Relevant visits and the fit of received enquiries |
| Suitable visitors cannot understand the offer | Clarify service, evidence and next step | Observed understanding and relevant enquiry outcomes |
| Visitors encounter unusable controls or errors | Repair the affected journey | Completion on the relevant devices and access methods |
| Received enquiries lack an owner | Define responsibility and next-step handling | Assigned, assessed and unresolved cases |
For reporting, decide what your rate actually measures. A session-based completion rate counts sessions with the defined outcome, divided by eligible measured sessions in the same cohort. Counting every event and dividing by sessions is a different calculation. One session can contain several events. Google's session key event rate definition illustrates that distinction.
If you cannot reliably connect a received enquiry with a measured session, report those counts separately and state the limitation. Do not invent a visitor-level connection to make a funnel look complete. Review valid enquiries as well as recorded submissions so a rise in spam does not appear to be success.
Record what changed and compare relevant audiences over a meaningful period. With few enquiries, a small numerical change can produce a large percentage swing. Different traffic sources, seasonality and simultaneous campaigns can also change the mix. A before-and-after improvement is evidence to examine, rather than proof that one edit caused it.
A/B testing needs a defined hypothesis, metrics and evaluation design. It can help answer a question when there are enough observations for the decision. Set the outcome and evaluation plan before judging the result. Do not call a test a winner after a handful of convenient outcomes. Repair and verify broken functionality as soon as possible rather than leaving it broken for an experiment.
The benefit of this review is a clearer priority: working contact routes, more useful information and a process the business can follow through. Check whether the change improves suitable enquiries and follow-through for your business.
If your website is attracting visits but the next step is unclear, talk to Ranking Rebels about the page, the audience and the point where the evidence stops. Those details provide a stronger starting point than a traffic total alone.