SEO Leads Singapore: Paying for SEO but No Enquiries?

Short answer: our own tracker logged 41 "leads", yet only 1 of 33 real rows left contact details. So if you pay for SEO and see almost no enquiries, check how your SEO leads Singapore report is counted before you judge the SEO. A WhatsApp tap is not an enquiry, and raw rows can count one enquiry twice.

Most talk about SEO leads in Singapore jumps straight to "more traffic" or "better keywords". That skips a step. If the number you are shown is wrong, every decision built on it is wrong too: you may fire an agency that was working, or keep one that was not.

We know this because we checked our own numbers first. Below is what our tracker logged for singrank.com, three counting faults we found in our client work, and five questions you can put to any agency this week.

Why your lead number may be wrong before SEO is

Many owners who pay for SEO end up unsure whether it is working. That doubt can be a counting problem, not only a ranking problem.

The count can go wrong in three ways:

• The wrong thing is counted. A button press is logged as a "lead", even when no message was ever sent.
• The same thing is counted twice. One enquiry becomes two rows.
• The credit goes to the wrong page. The tracker records where the button was pressed, not the page that brought the visitor in.

Search Console will not settle it either. It counts a click whenever someone follows a link from Google to your site (Google Search Console Help). A click is a visit, not an enquiry.

What our own tracker logged

Between 21 August and 30 October 2026, the lead tracker on singrank.com stored 41 rows. When we opened them, this is what they really were:

What the row really was	Rows
Another SingRank site that shared the same tracking tag	6
Our own tests on a local computer	2
WhatsApp button taps on singrank.com	32
Form submits on singrank.com	1
Total "leads" in the tracker	41

So 8 of the 41 rows did not come from real singrank.com visitors at all. Of the 33 that did, 32 were WhatsApp taps, and 3 of those taps were repeats from the same device within 10 minutes. Only 1 of the 33 left contact details.

If you want to check our counting, the anonymised rows are here. They hold no IP addresses, names, emails, phone numbers or message text.

A tap tells us someone pressed the button. It does not tell us they sent a message. Our tracker cannot see inside WhatsApp, so we do not know how many of those 32 taps became real conversations. We would rather say that than round it up.

Two more things we did not like:

• None of the 41 rows recorded the page the visitor entered on. We know the button was pressed on the homepage for 16 rows and on our pricing page for 9, but not which article or search brought those people in.
• The tag was shared with another site, which is how 6 rows from that site ended up in our count.

For the 33 singrank.com rows, the browser reported an AI assistant as the referring site for 11, Google for 3, our own pages for 9, and nothing for 10. That is the referrer of the page where the tap happened, not of the first visit. In other words, it is a last-touch view: it shows the last step before the tap, not how the visitor first found us.

If we reported "41 leads", we would be fooling ourselves. That is the point of this article: ask what is behind the number. This is also one small site over 71 days, so treat it as an example of how raw counts mislead, not as a benchmark for yours.

A tap, an enquiry, a qualified lead: four different counts

Google's own Analytics documentation splits leads into stages. Its recommended event generate_lead fires when a visitor sends in a form or asks to be given information. A separate event, qualify_lead, marks a lead judged to fit the criteria for a qualified lead, and close_convert_lead marks one that has become a customer (Google Analytics Help).

A WhatsApp tap fits none of those on its own. Keep the four counts apart:

Count	What it means	Where you check it
Contact taps	Someone pressed WhatsApp, call or email	Your tracker or GA4 event
Enquiries	A message or form with a way to reply reached you	Your inbox, WhatsApp, call log
Qualified leads	The enquiry fits what you sell, where you sell it	You or your salesperson decide
Customers	The enquiry paid	Your invoices

An honest report shows all four, or says which ones it cannot see. A report that shows only the first count and calls it "leads" is not exactly lying, but it answers a different question from the one you asked.

Three counting faults we found in our client work

These come from our own work setting up and checking lead tracking for client sites. No client is named, and the figures are counts, not anyone's details.

1. One enquiry, stored twice. On one client site, September 2026 showed 175 raw lead rows. Once we removed duplicates, there were 113 unique contacts. Every form message had been stored twice, in the same second. Most of the extra rows were WhatsApp repeats: 56 WhatsApp rows followed another WhatsApp row on the same page within 10 seconds.

On the same site we later found two tracking scripts running at once. The older one wrote a "WhatsApp click" every time someone clicked inside the contact widget, even just to open or close it. Between 3 September and 5 October, 148 rows were these bare clicks, against 5 labelled WhatsApp taps.

2. The entry page was never recorded. On another client site, the tracker only started recording the entry page (the page a visitor first lands on) on 3 September 2026. The hundreds of leads recorded before that date only carry the page where the button or form was used. They can never be traced back to the article that brought the visitor in. Comparing "leads per article" before and after that date would compare two different measuring tools, not two levels of performance.

3. A conversion rate that could not be real. Our own internal work queue once showed conversion rates for three pages that were higher than the number of visitors could allow. The cause was simple: it divided leads from every channel by clicks from Google organic search only. Those are two different groups of people, so the figure had no ceiling and was not a true rate at all. We do not use those rows to rank work, and you should be wary of any lead rate that looks too good to be true.

None of these was an SEO problem. But each one could have changed what an owner decided about their SEO.

SEO leads Singapore: how to track them without fooling yourself

You do not need an agency for this part. All it takes is a month of careful counting.

1. Count enquiries by hand for one month. Include every WhatsApp chat, form and call from a real person asking about your service. Note the date and, if you can, what they said brought them.
2. Set key events in Google Analytics 4. Google describes key events as a way to measure the actions that matter to your business and to judge which marketing channels lead people to take them (Google Analytics Help). Make a form submit and a WhatsApp tap two separate events, never one.
3. Know which channel gets the credit. GA4's attribution paths report needs a key event, because without one there is no path to show. Google has also removed several older attribution models from GA4 since November 2023, first click and linear among them (Google Analytics Help). Ask your agency which model its report uses.
4. Look at AI referrals separately. GA4 has an AI Assistant channel for visitors who come from tools such as ChatGPT, Gemini, Deepseek, Copilot or Grok (Google Analytics Help). On our own site, AI assistants were the largest recorded referrer class for taps, so do not lump them into "direct". More on this in the AI traffic your reports miss.
5. Match the tracker to your inbox. If the tracker says 30 and your inbox says 6, the tracker is counting something else. Find out what before you read any trend.

If your contact route is WhatsApp only, the gap between taps and conversations matters even more. We looked at that choice in a WhatsApp-only contact button, and at how Singapore sites let people reach them in how Singapore websites let people get in touch.

Five questions to ask your agency about its lead count

Google's own guide on hiring an SEO suggests asking: "What kind of results do you expect to see, and in what timeframe? How do you measure your success?" (Google Search Central). For leads, make it concrete:

1. What exactly do you count as a lead? A tap, a sent message, or a message with contact details?
2. Can I see one week of raw rows, with duplicates and test rows marked?
3. Does your tracker record the entry page, or only the page where the button was pressed? Since when?
4. Which channel gets the credit, and by what rule?
5. How do your numbers match my inbox, WhatsApp and invoices?

A good agency answers these without getting defensive, and says plainly where its numbers stop. If you want the wider picture, see what a monthly SEO report should show, how to prove SEO pays and a 30-day test for an agency that is not delivering. Rankings still matter as an early sign; here is how to check your rankings for free.

FAQ

Does a WhatsApp click count as a lead?

Not on its own. A click shows someone pressed the button; it does not show a message was sent. Google's lead event, generate_lead, is meant for a visitor who sends in a form or asks to be given information (Google Analytics Help). On our own site, 32 of 33 tracked "leads" were taps and only 1 left contact details. Count taps, but report them apart from enquiries.

Why don't Search Console clicks match my lead count?

They measure different things. Search Console counts clicks from Google to your site. A lead count, if it is honest, counts people who reached you, from any channel. Our own tracker shows how big the gap can be: of 33 real rows, only 1 left contact details. If a page shows more leads than clicks, the two numbers are measuring different groups of people.

Can an SEO agency guarantee enquiries?

No honest one will. Rankings, AI mentions and enquiries depend on things no agency controls, from competitors to how people search. We do not guarantee them either. What an agency can promise is clear work, a clear way of counting, and a straight answer when the count is low or the tracker cannot see something.

How SingRank counts leads

We count a lead when a person reached us, not when a button was pressed, and we show taps separately. Our plan includes a live dashboard that shows which pages bring enquiries, not just rankings, and our pricing page says what we promise: "We do the agreed work and show you honestly what changed, including the enquiries we can track." What we do not offer is "Any guaranteed ranking, AI mention, enquiry or sale" (SingRank pricing).

The plan is S$750 per month, 3-month minimum, then month to month with 30 days' written notice. If you are not ready for that, a one-hour consultation is S$150.

Send us your last lead report and your website on WhatsApp. We will tell you which numbers are taps, which are enquiries and what is missing. We reply within 1 working day. Or see the full plan on our pricing page.

---

Method

We exported every row stored for singrank in our lead collector, from 21 August to 30 October 2026, and labelled each by site, event type, button page, referrer class, whether contact details were filled in and whether an entry page was recorded. Every figure in this article was recomputed from that anonymised export.

Limits

This is one small site, 71 days and 41 rows: an illustration, not a benchmark. A tap shows a button was pressed, not that a message was sent. Repeat taps were flagged by device, which cannot be reproduced from the export beyond the flag, and referrers show the page where the tap happened, not the first visit.

Cite this

SingRank, "SEO Leads Singapore: Paying for SEO but No Enquiries?", 31 October 2026, https://singrank.com/blog/seo-leads-singapore/