Examples are illustrative unless otherwise attributed.
A useful place to begin
Your analytics dashboard says people are arriving. Your bank account tells a quieter story.
It is tempting to respond with another campaign, a discount, or a homepage redesign. Each can help in the right circumstances. But they solve different problems. More visitors will not repair a broken checkout. A lower price will not explain an offer nobody understands. A prettier page will not make the wrong audience need your product.
When a website gets traffic but few sales, investigate five things in order: measurement, visitor intent, offer clarity, decision friction, and follow-up. The job is to find the earliest meaningful point where an interested person cannot make progress.
This guide focuses on the path from arrival to inquiry or purchase. If people already sign up but never reach value, use our signup journey guide.
1. Verify that “no sales” means no sales
Compare your website reports with your order system, payment processor, or CRM. A missing purchase event is a measurement problem. A form that says “thank you” without delivering the inquiry is an operational problem. Both can look like bad marketing.
Run a complete test on a phone and a computer. Submit the form. Confirm the message reaches the right person. Follow the booking link. Check confirmation emails. If you sell online, use a safe test purchase method and confirm that the order and analytics event each appear once.
Record four facts: the action you tested, what you expected, what happened, and who owns the fix. Resolve a confirmed failure before interpreting a conversion-rate trend.
Find the earliest meaningful point where an interested person cannot make progress.
2. Separate relevant visitors from everyone else
A total of 5,000 visits can contain shoppers, job seekers, existing customers, bots, and people looking for a free definition. Those visitors do not have the same likelihood of buying.
Review the landing page, traffic source, campaign promise, geography where relevant, and device. Ask what the visitor probably wanted when they arrived. Treat that answer as a hypothesis until search terms, customer conversations, or observed behavior support it.
For paid search, inspect actual search terms rather than assuming your chosen keywords describe every visit. Google's search terms report can reveal queries that call for tighter targeting or negative keywords. Source: Google Ads, search terms report.
Imagine a consultant selling onboarding strategy whose most popular article answers “what is onboarding?” The article may be useful, but many readers are not shopping for help. A more relevant next step could be a first-use checklist. An immediate request to purchase consulting may ask too much too soon.
3. Build the smallest useful funnel
Do not start with twenty dashboard widgets. Start with the steps between a relevant arrival and a commercial outcome.
| Step | Hypothetical monthly count | Question it helps answer |
|---|---|---|
| Relevant landing-page sessions | 1,000 | Did enough suitable people arrive? |
| Inquiry starts | 80 | Did the offer create interest? |
| Submitted inquiries | 40 | Could people complete the next step? |
| Qualified inquiries | 20 | Was the promise attracting the right fit? |
| Sales | 4 | Did suitable buyers move through the sales process? |
In this invented example, the submission rate is 4% of landing-page sessions. Half of submitted inquiries qualify, and one in five qualified inquiries becomes a sale. These figures are a teaching example, not benchmarks.
There is also a visible gap between starting and submitting the form. Investigate it. Perhaps the form is broken. Perhaps people discover a minimum budget they cannot meet. Those explanations imply very different changes.
For a real business, join leads and sales by cohort where possible. A sale closed this month may have begun with a visit three months ago. Dividing unrelated monthly totals can produce an attractive but misleading funnel.
4. Check whether the offer answers a buyer's questions
Read the page as someone who has never met you. Can they explain who it is for, what problem it addresses, what they receive, and what happens next?
“Strategic solutions for ambitious businesses” leaves almost every buying question unanswered. “We review one trial onboarding journey, implement one agreed improvement, and hand over a measurement plan” gives the reader something concrete to evaluate.
Put relevant proof beside the claim it supports. An example deliverable can demonstrate what the customer will receive. A testimonial can describe a real experience, with permission. Neither should imply a financial result that the evidence does not establish.
If the page attracts interest but generates unsuitable inquiries, make fit clearer. A narrower offer can reduce form submissions while improving qualified conversations.
5. Look for the cost of saying yes
Price is only one cost. A buyer may also anticipate a long call, a difficult implementation, a contract they cannot escape, or the embarrassment of choosing badly.
Name what actually happens after the click. If it is a free fit call, explain its length and purpose. If a purchase requires a separate subscription or setup work, make that visible. If delivery takes several weeks, say so before checkout.
For ecommerce, inspect shipping costs, delivery expectations, returns information, available payment methods, and checkout errors. For services, inspect scope, suitability, access requirements, scheduling, and who will do the work.
Observe a few people attempting the task. Ask them to show you what they would do next and explain what they expect. “I like the website” tells you much less than “I cannot tell whether this includes implementation.”
6. Follow the inquiry after the website ends
A submitted form is not a sale. Assign an owner, a response expectation, and a useful first reply. Explain the next step without making the person repeat everything they already told you.
If the lead asks about scope, answer scope. If the lead asks whether you support their platform, resolve that question. A generic “just checking in” does little to help someone make a decision.
Our companion guide, A Lead Nurture Email Sequence That Helps People Decide, gives you five messages to adapt.
7. Choose one change and a fair way to judge it
Write a short decision brief: “We observed X. We think Y may explain it. We will change Z. We will watch this outcome and these possible downsides.”
For the hypothetical form problem, you might remove nonessential fields while preserving one useful qualification question. Measure qualified inquiries per eligible landing-page session, form completion, and the team's time spent handling poor-fit inquiries.
With limited traffic, use direct observation to resolve usability problems and report quantitative changes cautiously. A handful of extra leads does not establish a dependable conversion lift. A before-and-after comparison may also reflect seasonality, audience changes, or a different offer.
Your seven-day starting plan
Day one: test the transaction or inquiry path. Day two: separate traffic by intent and source. Day three: map the funnel with counts. Day four: observe customers or prospects completing the task. Day five: choose one supported change. Day six: implement and check it. Day seven: review whether the change works technically and set an appropriate outcome-review date.
Seven days can establish a better process. It does not guarantee enough evidence to declare a revenue result.
Put this to work: Use the free Growth Leak Finder to choose a starting point. If you want help investigating an acquisition or inquiry journey, describe it to Ashley. Start with the specific step where people seem to get stuck.
Sources & further reading
Google Ads — Search terms report ↗Background references are distinguished from our original examples and proposed exercises.


