5 min read

Product

₹1,075 in Google Ads Got Me 33 Signups. One Person Paid ₹49.

A 17.5% click-through rate looked brilliant on day 12 of my first Google Ads campaign for Astrika. Then I opened GA4 and counted how many people paid. Here is the funnel and why I let the prepaid balance run out.

google ads conversion trackingcost per signup vs cost per purchasegoogle ads ctr misleading

₹1,075 in Google Ads Got Me 33 Signups. One Person Paid ₹49. · Products Decoded
By Debashis Nayak · Products Decoded27 Oct 2026

On day 12 of my first Google Ads campaign for Astrika, the click-through rate was 17.5%. I would have framed that screenshot if the next tab, GA4, had not shown me how many people actually paid. One. For ₹49.

I had spent ₹1,074.92 to find that out. This post is the read I did that day, the numbers as they came, and why I decided not to put more money in.

What did the Google Ads dashboard say?

One campaign was doing all the work, an intent campaign for India. The Ads UI, all-time:

MetricValue
Impressions559
Clicks98
CTR17.5%
Spend₹1,074.92
Average CPC₹10.97

Eleven rupees a click, and nearly one in five people who saw the ad clicked it. If you stop reading here, this is a campaign you scale.

The keywords split in a way I did not expect. Marriage queries took 90 of the 98 clicks and ₹1,015.80 of the spend. Career queries got 8 clicks for ₹59.12. So roughly 92% of my clicks came from people asking about marriage, and the money followed them.

The brand campaign had 1 impression and 0 clicks in 12 days. Nobody was searching for the product by name, which tells you where a young product sits. People did not know Astrika existed, so every click had to be a stranger.

What did GA4 say about the same clicks?

I pulled the Google / cpc traffic in GA4 from 7 October, when I started tracking.

  • 33 users signed up. That works out to about ₹32.6 per signup.
  • 32 of them asked a free question.
  • 1 got as far as starting checkout.
  • 1 purchase, ₹49.

Read that list again slowly, because the funnel is lopsided in a very specific way. Signing up worked. Asking a free question worked. The step where money changes hands is where 32 of 33 people stopped.

I should say what that does and does not prove. It is 12 days and 98 clicks. A sample that small can swing a lot, and I would not publish a conversion rate off it. What it does show is the shape: people will happily take a free answer from a stranger’s product, and almost none of them paid on day one.

Why are cheap signups the metric that lies most?

A signup costs the user almost nothing. An email address and a few taps. So a signup tells you the ad was interesting, the landing page did not scare anyone off, and the free offer was attractive. It tells you nothing about whether the product is worth money.

₹32.6 per signup looks healthy next to the ad benchmarks people quote. But a signup is only a cost if it can turn into revenue, and the revenue side here was ₹49 against ₹1,075 spent. I think the honest name for the 33 is “free users I paid for”.

CTR has the same problem one step earlier. 17.5% mostly measures how well the ad copy matches what someone just typed. Marriage questions are emotional, and my guess is that an ad saying “we answer that” gets clicked on instinct. Whether the click is worth ₹11 depends entirely on what happens after the click, and CTR cannot see that.

My rule after this: judge an ad on cost per purchase, and treat cost per signup as a diagnostic only. If those two numbers are far apart, the gap is the product’s problem, not the ad’s.

Why did the Ads UI show 0 conversions?

Google Ads said zero conversions while GA4 clearly had a purchase from Google / cpc. The GA4 import was not crediting conversions back to the campaign.

That matters more than it sounds. If I had only looked inside Ads, I would have concluded the campaign produced nothing, which is wrong, and if Google’s bidding had been learning from that zero it would have been optimising blind. I never found out how much of the problem was setup and how much was timing, so treat the import as unfinished business. For this read I trusted GA4 and cross-checked what I could.

If you run ads on a small budget, check this before you trust any dashboard. Open both tools side by side on the same date range and make sure the purchase count matches.

Why did I let the prepaid balance run out?

About ₹425 of prepaid balance was still sitting in the account. My recommendation to myself was to let it run out and not top up.

The reasoning is plain arithmetic. At roughly ₹33 a signup and a free-to-paid step that had produced one ₹49 order, more ad money would buy more signups, and signups were never my problem. Buying 100 more of them would test the same broken step at higher cost, and I already had 32 free-question askers who had seen the paid step and walked past it.

So the money should go to whatever sits between “asked a free question” and “paid”. An ad cannot fix that. Putting more spend on a leaky funnel is how a ₹1,000 experiment turns into a ₹10,000 habit.

I will also admit the uncomfortable half. A single payer could mean the price is wrong, the offer is unclear, the checkout is clumsy, or that this audience simply does not pay for this at the first visit. I do not know which one yet, and 33 signups cannot tell me. That is the whole reason to stop spending and go and look at the 32 people who did not buy.

What would I check before spending on ads again?

  1. Is the conversion event imported into Ads, and does the purchase count match GA4 on the same dates?
  2. What is the cost per purchase, not per signup? If it is above what a customer pays, stop.
  3. Do the clicks come from the intent I want? Mine were 92% one topic, which is worth knowing before writing more ad copy for the other.
  4. Is anyone searching for the product by name? If the brand campaign shows zero, every click is cold.
  5. Did I set a stop rule before I started? Mine was the balance running out, which turned out to be a decent one.

A 17.5% CTR cost me ₹1,075 and taught me where the funnel breaks. Cheap tuition, honestly. I just would not have learned it from the Ads screen.