A local service business in India should budget backwards from its target: divide your desired number of new customers by your realistic conversion rate to get required leads, then multiply by your cost per lead. In practice most single-location businesses need a minimum of roughly ₹20,000–30,000 per month per platform for the algorithm to gather enough data to optimise. Spending less than that usually produces noise rather than results.
Key takeaways
- Budget backwards from customers, never forwards from a number that feels comfortable.
- Below a certain threshold, ad platforms cannot optimise — tiny budgets waste money rather than saving it.
- Google captures existing demand. Meta creates it. They do different jobs and should not be judged by the same metric.
- Your real metric is cost per booked appointment, not cost per click or cost per lead.
- Do not advertise at all until you can answer every lead quickly. Ads into a leaking funnel are the most expensive mistake in local marketing.
Work backwards, not forwards
Most business owners pick a number that feels affordable. That is guessing. Do this instead:
1. New customers wanted per month → 10
2. Your lead-to-customer conversion rate → 20%
3. Leads required = 10 ÷ 0.20 → 50
4. Your cost per lead (test to find it) → ₹600
5. Monthly ad budget = 50 × ₹600 → ₹30,000
Then sanity-check it against value:
6. Average customer value → ₹8,000
7. Revenue from 10 customers → ₹80,000
8. Ad spend as a share of revenue → 37.5%
Whether 37.5% is acceptable depends entirely on your margin and whether those customers return. For a dental clinic where a patient returns for years, it may be excellent. For a one-off low-margin service, it is unsustainable.
Note step 2. If your conversion rate is 8% instead of 20% because leads wait four hours for a reply, you need 125 leads instead of 50 — and your budget becomes ₹75,000 for the same ten customers. This is why response speed is an advertising decision, not just an operations one.
What is the minimum viable budget?
Ad platforms need data volume to optimise. Below a threshold, the algorithm never exits its learning phase and performance stays poor.
As a practical rule for a single-location Indian local business:
| Situation | Practical minimum per month |
|---|---|
| Testing one platform, one campaign | ₹20,000–30,000 |
| Running Google and Meta together | ₹40,000–60,000 |
| Competitive category or metro city | Meaningfully higher |
| Below ₹15,000 | Generally not worth starting |
If your budget is below the threshold, do not split it across platforms. One platform funded properly beats two funded badly, every time.
Google or Meta: which should you start with?
They do fundamentally different jobs.
| Google Ads | Meta Ads | |
|---|---|---|
| Demand | Captures existing intent | Creates new interest |
| The customer is | Actively searching right now | Scrolling, not looking for you |
| Cost per click | Higher | Lower |
| Lead quality | Higher intent | Lower intent, higher volume |
| Best for | Urgent, searched-for services | Visual, considered, discretionary services |
| Typical fit | Dentists, emergency services, CAs, brokers | Gyms, interior designers, aesthetic services |
Start with Google if people already search for what you sell — "dentist near me", "CA for ITR filing", "3BHK in Whitefield". You are capturing a customer who has already decided.
Start with Meta if your service is discovered rather than searched, is visually compelling, or is an impulse or aspiration purchase — gym memberships, interior design, cosmetic treatments.
Most businesses eventually run both, with Google capturing bottom-of-funnel demand and Meta filling the top.
What should a lead cost?
There is no universal answer, but there is a universal ceiling:
Maximum sustainable cost per lead
= average customer value × gross margin × lead-to-customer rate × target ad share
Example for a gym:
- Annual membership value: ₹18,000
- Gross margin: 70% → ₹12,600
- Lead-to-member conversion: 25%
- Value per lead: ₹12,600 × 0.25 = ₹3,150
- Willing to spend 25% of that on acquisition → maximum ₹787 per lead
If your actual cost per lead is under that, scale up. If it is over, fix conversion before increasing spend.
The metric that actually matters
Cost per click is vanity. Cost per lead is incomplete. Cost per booked appointment is the only number that connects spend to reality — covered fully in cost per booked appointment.
A campaign delivering leads at ₹400 that never get answered is worse than one delivering leads at ₹800 that reliably convert into appointments.
When you should not run ads at all
Be honest about these. Advertising amplifies whatever system already exists — including a broken one.
- You cannot answer leads within an hour. Fix response first. Ads into an unanswered pipeline are pure waste.
- You have no tracking. If you cannot tell which campaign produced which customer, you are buying lottery tickets.
- Your Google Business Profile is unoptimised. Free local visibility is cheaper than paid. Do that first.
- You have five Google reviews. Ads will send people to a profile that does not convert.
- Your budget is under ₹15,000/month. Put it into fixing leaks instead — the return is higher and permanent.
A sensible 90-day ramp
| Phase | Spend | Focus |
|---|---|---|
| Days 1–30 | ₹20,000–25,000, one platform | Establish baseline CPL, gather conversion data |
| Days 31–60 | Same or slightly higher | Cut losing keywords/audiences, double down on winners |
| Days 61–90 | Increase 30–50% if CPL is under target | Add the second platform only once the first is profitable |
Do not change budgets and creative and targeting simultaneously. You will learn nothing.
Frequently asked questions
How much should a small business spend on Google Ads in India?
Calculate backwards from your customer target: required leads multiplied by your cost per lead. Practically, most single-location businesses need at least ₹20,000–30,000 per month for one platform, since smaller budgets do not generate enough conversion data for the algorithm to optimise.
Is Google Ads or Meta Ads better for a local business?
Google is better when customers actively search for your service, such as dentists or chartered accountants, because it captures existing intent. Meta is better for visual or discretionary services like gyms and interior design, where you need to create interest. Many businesses eventually run both.
What is a good cost per lead in India?
It varies widely by city and category, so the useful question is your maximum sustainable cost per lead — average customer value multiplied by gross margin and lead-to-customer conversion rate, then by the share of that you are willing to spend on acquisition. Anything below that ceiling is workable.
Should I run ads if I cannot answer the phone quickly?
No. Advertising into a funnel where leads wait hours for a reply wastes most of the spend, because the majority of customers buy from whoever responds first. Fix response speed before increasing budget — it is cheaper and improves the return on every rupee you subsequently spend.
How long before ads start working?
Expect 30 days to establish a reliable baseline and 60–90 days to reach stable, optimised performance. Judging a campaign in its first two weeks, while the platform is still in its learning phase, leads to abandoning campaigns that were about to work.
Spend less, convert more
fizaki tracks every rupee from click to booked appointment, and makes sure the leads your ads generate are answered in under 60 seconds.
Stop losing leads to a phone nobody answers.
Book a free 20-minute audit — we'll show you exactly how many leads you're leaking and what they're worth.