Google’s setup flow will walk you helpfully through building an account that is wrong for your budget.
Broad match keywords, automated bidding from day one, Display and Search Partners switched on by default, and a gentle suggestion to create a separate campaign for each of your services. Every one of those is reasonable advice for somebody spending Rs 8 lakh a month. At Rs 30,000 they are the mechanism by which the money disappears.
Here is the number that decides everything else.
Indian agencies publishing 2026 benchmarks put the average Google Search click at roughly Rs 24, which I went into when comparing Google and Meta costs in India. At that rate, Rs 30,000 of media buys about 1,250 clicks in a month. Forty two clicks a day.
If your landing page converts at a respectable 3%, that is around 37 conversions in the month.
Now read that against how automated bidding works. Google’s own Smart Bidding material sets no hard minimum for Target CPA, but the figure everybody uses, including Google’s own guidance on when to judge the strategy, is 30 conversions in the preceding 30 days. You are at 37. One campaign, barely clearing it.
Split that across six campaigns for six services and each one has six conversions a month. That is not data. That is noise, and you have asked an algorithm to learn from it.
At this budget the scarce resource is conversion signal, not money. Build the account to concentrate it.
Everything below follows from that.
One campaign, two or three ad groups
Not one per service. One, total.
Inside it, group keywords by what the searcher wants rather than by what you call your departments. A plumber does not need separate campaigns for leak repair and bathroom fitting. They need one campaign, an ad group for emergency intent and an ad group for planned work, because those two want different ad copy and nothing else differs.
If a service genuinely deserves its own budget and its own target, it deserves its own campaign. Almost nothing does at Rs 30,000.
Exact and phrase match only, for the first six weeks
Broad match at this budget is an expensive way to discover that Google’s idea of a related search is generous.
Start with exact match on the eight to twelve phrases you are confident about, and phrase match on the same terms. Two match types, one small list. You will get less volume than the forecast promised and the volume you get will be relevant.
Add broad match later, in its own ad group with its own budget cap, once you have a search terms history worth mining. Not before.
The negative list, written before you launch
Most advice tells you to check your search terms report weekly and add negatives. That is correct and I have written about it as the first thing to open when conversions disappoint, but it is a repair, and it runs a week behind the spend.
Build the list before the first impression. Ten minutes, and in India it is close to the same list every time.
Free, cheap, sasta, low cost. Course, training, diploma, certification, institute, tutorial, how to, DIY, jobs, vacancy, salary, internship, resume. Wikipedia, Quora, Reddit, YouTube. Your own brand name if you are running a separate brand campaign, and every competitor’s brand name unless you have deliberately decided to bid on them.
Then the category-specific ones. A dentist adds “dental college” and “BDS”. An interior designer adds “images”, “designs”, “ideas” and “drawing”, because those pull students and browsers. Anybody selling a service adds “franchise” and “dealership”.
That list will save more money in month one than any bid adjustment you make in month three.
Conversion tracking before launch, not after
This is the one that gets skipped and it invalidates everything else.
If you launch without conversion tracking, you spend the first month buying clicks you cannot evaluate, and automated bidding has nothing to optimise towards, so it optimises for clicks. Set it up first, test it by submitting your own form, and confirm the conversion registers in the interface before you spend a rupee.
And be specific about what counts. A form submission counts. A page view does not. If your only conversion action is “visited the contact page”, you are measuring curiosity.
The India part: calls and WhatsApp
For most Indian service businesses the enquiry does not arrive as a form. It arrives as a phone call or a WhatsApp message, and both are routinely untracked.
Calls are the easier half. Use the call asset and Google’s call reporting so a call lasting longer than a chosen duration registers as a conversion. Pick the duration honestly, because thirty seconds is somebody checking you exist.
WhatsApp takes a small trick. Track the click on your WhatsApp button as a conversion event, which gives you the count, and then accept that attribution stops at the handover, because the conversation happens somewhere Google cannot see.
The trick is to put a pre-filled message in the link. A wa.me link accepts a text parameter, so the message arrives already written: something like “I saw your ad for emergency plumbing”. Now whoever answers the phone knows the source without asking, and you can tally real enquiries against the click count at the end of the month.
That one change does more for an Indian small business account than most optimisation work, because it closes the gap between what the dashboard says and what actually happened.
The weekly fifteen minutes
Not daily. Daily tinkering at 42 clicks a day is reacting to randomness.
Once a week, in this order. Open the search terms report and add negatives for anything irrelevant. Check the conversion count against your own record of enquiries received, because a divergence there matters more than any metric in the account. Glance at the ad schedule if you are only reachable during office hours. Then close the tab.
Change one thing at a time and give it two weeks. At this volume you cannot read the result of two simultaneous changes, and people who make four changes a week never learn which one worked.
One note on the money: advertising services in India carry 18% GST, recoverable as input credit if you are registered, so the sticker figure and the real cost differ. The detail is in what to budget and your CA will know your position.
Before lunch
An hour, in order, before you launch anything.
Write the negative keyword list. Use the generic block above and add six of your own.
Then set up and test the conversion action by submitting your own form and watching it appear.
Then add the pre-filled wa.me link to your landing page, and send yourself a message through it to check the text arrives.
Only then build the campaign, with one campaign, two ad groups, exact and phrase match, and manual bidding until you have a conversion history worth handing over. Enhanced CPC was retired in March 2025, so the choice is manual CPC or an automated strategy, and at month one manual is the honest answer.
If you’d like a second pair of eyes on an account before you turn it on, drop me a line on email, WhatsApp, phone or LinkedIn and we can have a quick chat, and my notes on performance marketing cover the rest. I’m taking on a small number of contracts at the moment, so I’ll tell you straight whether it’s something I could help with.