Case studies / Student accommodation — paid search

The Google Ads dashboard said ₹243 a lead. The real figure was ₹1,100. Here is what we found and fixed.

A season of Google Ads looked efficient on the dashboard — ₹243 to ₹453 per conversion. Checked against real named leads, it was about ₹1,100. Roughly 40% of the recorded conversions were fraudulent clicks, call tracking had been dead for seven weeks, and doubling the budget made results worse. We found ₹11,300 of low-value spend and rebuilt the account at ₹500 a day.

Radiance Residency · Indore, India · radianceresidency.com

Built and run in-house — this is our own property, not a client engagement, which is why every number can be produced on request.

~40%
of recorded conversions were fraud: ₹11,300
low- or zero-value spend identified: −53%
conversions/day after a budget doubling: 147
negative keywords now shielding spend

Source: Google Ads API pulls logged in the property's ads tracker, June – September 2026. The conversions column is quoted only as the thing that was wrong.

What was the problem?

The campaign's own numbers said it was working. Cost per conversion of ₹243–453 looked like a bargain for a booking worth months of rent, and the obvious move was to spend more.

The owner did not trust it, and spotted what the data could not: the three “winning” search terms were the fraud. The owner was right, and the dashboard was wrong.

What did we build?

Lead-level truth instead of the conversions column

A named lead was redefined as a form email carrying the Google-ad flag and a real name and number. Measured that way, the real cost was around ₹1,100 per lead — two to four times what the dashboard claimed.

Fraud evidence Google could not argue with

One day on the Search Partner network showed a 49.10% click-through rate — 164 clicks from 334 impressions — against 7.8–14.4% on every other day. That anomaly, which needs no reference to our own site, became the basis of a ₹2,678.78 refund claim. Search Partners and Display were switched off for good.

Fixing what the account had broken itself

The phone-call conversion goal pointed at a conversion action that did not exist, so seven weeks of calls were never recorded. Four exact- and phrase-match duplicates of the best keyword were bidding against it and spent ₹2,674 for nothing. Both fixed, and 147 negative keywords added from cost-ordered search terms — never impression-ordered, which once hid 75% of a week's spend.

Budget rules written down from the evidence

On Maximise Conversions, a doubled budget (₹1,200 → ₹2,500/day) produced 17% more clicks at 23% lower cost per click — and 53% fewer conversions per day. Cheaper traffic, worse traffic. The rule now: check whether the constraint is budget or ad rank first, never raise more than 50% in one step, and judge every change on cost per named lead.

What changed?

₹11,300 of low- or zero-value spend identified line by line, with the recoverable ₹2,678.78 claimed from Google and the rest turned into account rules.
The September refill sprint ran at ₹500/day inside a 15 km radius: ₹1,416 over four days, 27 clicks, well under its ₹3,000 kill line.
The campaign was paused on 17 September because the property was full.

What is not working yet?

Attribution was broken for most of the season — no click IDs on the June and July rows, call tracking dead for seven weeks — so we cannot honestly say how many bookings the ads produced. The season billed ₹50,767, and half of August's spend landed after the season was effectively decided. We publish that because the next account we audit will have the same problems, and this is how we know where to look.

Common questions

How do I know if my Google Ads conversions are fake?

Match a week of recorded conversions against real, contactable people. If the dashboard's cost per conversion is far below your cost per named lead, or one day shows a click-through rate several times your normal, the conversions column is contaminated. Search Partners is the usual source.

Should I increase my Google Ads budget if impression share is lost to budget?

Only if budget is the bigger constraint. Here, 58–69% of impressions were lost to ad rank and often under 14% to budget — so more money bought cheaper, worse clicks. Fix ad quality and landing pages first.

Google Ads APIGA4Form-level ad flagsPython audit scriptsDaily waste watch

Could this work for your business?

It probably fits if:

Your ads dashboard reports conversions you cannot match to real customers
You are about to raise a budget because the cost per conversion “looks good”
You run on Maximise Conversions with Search Partners or Display switched on

What this used

Visibilityfrom $499

Carefrom $79

More case studies

Student accommodation