The first ad budget question every small business asks — 'Google ya Facebook?' — has a real answer, and it isn't a platform: it's a match between how your customers buy and what each platform sells. Google sells intent (people searching for a solution now); Meta sells discovery (interrupting the right people with something they didn't know they wanted). Ten thousand rupees spent against that logic teaches and earns; spent against it, it just teaches. Here is the decision framework and the exact setups.
The decision rule
- Choose Google Search when customers already search for what you sell: emergency/needs services (plumber, CA, clinic), high-intent local queries ('company registration Lucknow'), and comparison purchases — search volume exists; you're buying position in an existing queue
- Choose Meta (FB/Instagram) when customers must discover you: new/visual products (D2C foods, fashion, decor), impulse and aspiration categories, and audiences definable by interest/behaviour — no queue exists; you're creating demand
- The quick test: search your own money keywords — real volume with competitor ads means Google-first; crickets mean Meta-first
- Both-worlds businesses (restaurants, salons, coaching): usually Meta for discovery + Google Business Profile (free) for capture — paid search joins later
The ₹10k Google setup (intent capture)
One campaign, tightly caged: Search only (decline Display/Performance Max expansions at this budget — they eat small budgets illegibly), exact and phrase match keywords only (5–15 money terms; broad match at ₹10k is donation), geo-fenced to your actual service radius, negative keywords loaded day one ('free', 'jobs', 'course', 'salary' — the intent you don't want), ads written as answers (keyword in headline, offer + proof + CTA), and landing on a page that converts (the WhatsApp-button service page — never the homepage). Budget rhythm: ₹300–350/day for 30 days beats ₹10k in a week — algorithms and learnings both need days. Expect in most service categories: ₹30–150 CPCs, ₹150–600 per lead — which means ₹10k buys 20–60 leads and, more valuably, the data on which keywords produce customers.
One campaign, conversion-focused: objective = Leads or Sales (never Awareness at this budget — reach without response is decoration), audience = your radius + broad-with-good-creative (Meta's targeting now works best lightly constrained), creative is 80% of results — 3–5 variations (a process video, a customer-proof image, an offer card; vertical, native-feeling, hook in second one), destination = WhatsApp (click-to-WhatsApp ads are Indian SMEs' highest-converting format — conversations beat forms). Same rhythm: ₹300+/day, 3–4 weeks, kill losing creatives weekly and feed winners. Expect: ₹5–25 CPMs' worth of learning, ₹30–200 per conversation by category — and the creative insights that are the real asset.
Measurement: the part that makes it an investment
- Before spending: conversion tracking live (GA4 + platform pixels; WhatsApp click events at minimum), and the 'kahan se aaye' question institutionalised at your counter/DMs
- The only dashboard: spend → leads/conversations → customers → revenue, per platform — cost-per-lead means nothing until cost-per-customer joins it
- Judge at 30 days, not 3: early CPMs/CPCs settle, learning phases end, and week-one panic kills campaigns that week-three vindicates
- The honest kill rule: if 30 disciplined days produce cost-per-customer above your unit economics, the problem is usually the offer or landing experience — fix those before blaming platforms
After the first ₹10k: the scaling gates
The sequence that compounds: double down on what converted (the winning keywords/creatives get the next ₹10k — diversification comes after domination), fix the leaks the data exposed (slow responses, weak pages, pricing objections), add remarketing (₹2–3k retargeting site/profile visitors — the cheapest conversions in advertising), and only then add the second platform. The mindset that separates businesses that scale ads from those that swear off them: the first budget's job is learning economics (what a customer costs, what they're worth), not miracles — and ₹10k that proves a repeatable ₹200-per-customer machine is worth infinitely more than ₹10k that vaguely 'got some enquiries'.
How Aidwish helps
Aidwish runs first-budget ad programmes for client businesses — platform selection by category, the caged campaign setups, tracking installation and the 30-day read — so early ad money buys a working acquisition machine, not just impressions.