PanditG is ByteFlow's own product. That matters for this case study, because it means we paid for our own mistakes in cash rather than billing someone else for the learning.
The measurement was wrong before the funnel was wrong
The first instinct on a bad account is to optimise. We audited attribution instead, and found the thing that makes most small-budget accounts unfixable: the reported cost per lead was not the real one.
In this category most enquiries arrive as a phone call. A call that is not tracked does not exist in the ad platform, so the platform optimises toward whatever it can see — usually form fills from lower-intent traffic. Every decision made on that data pushes the account further from the customers actually converting.
So nothing was optimised until calls were counted. This is unglamorous work, and it is the single highest-leverage thing on most Indian service accounts, because the entire market runs on the phone.
Two things follow from getting it right:
- Bids move toward the queries that produce real enquiries, not the ones that produce measurable ones.
- You find out which half of the spend was never working — which is uncomfortable, and cheap compared to not knowing.
Then the funnel
The audit found the ordinary failures: a large share of spend on broad-match queries, and those clicks landing on a homepage that answered none of them.
Landing page per intent. General puja, premium specialisation, and NRI live puja are three different buyers with three different objections. One page cannot answer all three, and a page that tries answers none of them well. Each paid funnel got a page written for its query.
Ad copy that disqualifies. The cheapest click is the one you never pay for. Copy was rewritten to state scope and positioning plainly up front, so the wrong visitor self-selects out before the click rather than after it.
Programmatic SEO underneath. A city × puja matrix, schema-marked and internally linked, with local proof and question-shaped FAQ blocks. The point is not volume — it is that paid and organic stop being two teams guessing at the same market and start running off one intent map.
Why this transfers
The specifics are religious services, but nothing about the method is. Any Indian business where the enquiry arrives by phone — clinics, manufacturers, education, real estate — has the same measurement gap sitting under its ad account, and usually has not looked.
We run this account with our own money. The discipline it forces is why we are willing to run someone else's.