Start with one action that matters
A conversion funnel is a short sequence of steps a visitor might take before completing a goal. For a small organization, that goal could be submitting an enquiry, signing up for a newsletter or completing a donation. The funnel helps answer a more useful question than “How much traffic did we get?”: where do interested visitors stop moving forward?
Choose one goal before opening an analytics report. If your organization wants more service enquiries, the completed enquiry form is the conversion. Work backward through the likely journey: a visitor views a service page, selects the contact link, opens the form and submits it. This is a starting model, not a claim that everyone follows the same route. Someone might arrive directly on the contact page, so decide whether your funnel will include those visitors or focus on people who started on a service page.
Turn the journey into measurable steps
Write down what counts as completing each step. A page view can represent an entry page or form page. A tracked link selection can represent intent to contact. A successful submission should be recorded only after the form has actually been accepted, rather than when someone presses the submit button. Otherwise, validation errors may make unsuccessful attempts look like conversions.
Keep the first funnel short. For the service-enquiry example, you might measure service-page views, contact-link selections, form views and successful submissions. Give each event a clear name and record which page or form it relates to. If several service pages lead to the same form, decide whether to review them together or separately. Document these choices in a simple note so that the funnel still means the same thing when you revisit it.
Before interpreting any results, test the journey yourself. Follow it on a phone and a desktop device, submit a test form if appropriate, and check that each step appears once in your measurement tool. Make sure a reload does not accidentally count as another submission. Use the same reporting period and the same visitor or session basis across steps; mixing different counting methods can create a misleading picture.
Read drop-off without jumping to conclusions
For each step, compare the number of visitors who reached it with the number who continued to the next one. The gap is the drop-off at that point. A funnel report may also show a continuation rate: the share of people at one step who reached the next. Check what your tool uses as its denominator, since a rate based on all site visitors answers a different question from one based only on people who viewed the service page.
A large gap does not automatically mean a page is broken. Some visitors may be researching, looking for opening hours or deciding that the service is not a fit. Treat the funnel as a prompt for investigation. If many people reach the form but fewer complete it, try the form yourself: are instructions clear, required fields reasonable and errors easy to correct? If fewer people select the contact link from a service page, check whether the next action is easy to find and whether the page answers basic questions before asking for an enquiry.
Look at useful segments when there is enough activity to make them meaningful. Comparing phone and desktop journeys, or different entry pages, may reveal a problem hidden by the overall view. Avoid drawing strong conclusions from tiny groups or brief fluctuations. Also check whether visitors commonly take another valid path, such as calling instead of using the form. A funnel measures the path you defined, not every way your website helps someone.
Make a small change, then measure again
Pick one plausible obstacle and one change you can explain. If people reach the form but abandon it, you might clarify a confusing field label or remove a question that is not needed for an initial enquiry. If visitors rarely move from the service page to the form, you might make the contact option more visible and describe what happens after someone gets in touch.
Record what changed and when, then compare the same funnel over a reasonable period before and after the edit. Keep in mind that changes in traffic sources, seasonal demand or tracking setup may affect the comparison. The aim is not to prove that every movement in a report came from your edit. It is to see whether the evidence supports keeping the change, investigating further or trying a different approach.
Keep the measurement useful and proportionate
A basic funnel should be easy to maintain. Review event names when pages or forms change, and periodically repeat a test submission so you know the final step still records correctly. Note any known gaps, such as phone enquiries you cannot connect to a website visit. This helps others read the report with the right expectations.
Collect only the information needed to understand the journey. Funnel events generally do not need the contents of a visitor’s message, email address or other personal details. Follow your organization’s privacy practices and the requirements that apply to your site when choosing measurement tools. A modest, reliable funnel is more valuable than an elaborate dashboard nobody trusts or uses.
