Small budgets fail for a predictable reason: people spread them thin. ₹300 a day split across four ad sets and six creatives gives each combination about ₹12 — not enough for the platform to learn anything, and not enough for you to conclude anything either.
The fix is concentration. One audience, few creatives, long enough to be real.
Before you spend anything
The 14-day test structure
One campaign, objective: sales
Not traffic, not engagement. Traffic campaigns deliver cheap clicks from people who do not buy, and they look great in a report.
One ad set, broad
India-wide or your top three cities. Age 22–50. No detailed interest targeting. At ₹300 a day, narrow targeting starves the system of the data it needs.
Three creatives, deliberately different
Not three colour variants. One product-on-white, one in-use lifestyle, one founder-to-camera talking about why the thing exists. You are testing message, not polish.
Run for 7 days without touching anything
This is the hard part. Every edit resets learning. Write down that you will not look before day 4 and mean it.
Day 8: cut the worst creative, keep the budget the same
Two creatives, same ₹300. Do not add new ones yet.
Day 14: decide
Three outcomes, below. Do not extend "just to be sure" — that is how a ₹4,200 test becomes a ₹15,000 one.
The four numbers, and what they mean
After 14 days you will have roughly ₹4,200 spent and some orders. Work these out:
| Number | How to get it | What good looks like |
|---|---|---|
| Cost per purchase | Spend ÷ orders | Below your contribution margin per order |
| Click-through rate | Clicks ÷ impressions | Above ~1% is healthy; under 0.6% means the creative is not landing |
| Landing page conversion | Orders ÷ landing page views | 1–3% is normal for cold traffic in India |
| Contribution margin per order | From your pricing maths | The number everything is judged against |
Reading the result
| What you see | What it means | What to do |
|---|---|---|
| Good CTR, good conversion, cost per purchase below margin | It works | Raise budget 20% every 3–4 days. Do not double |
| Good CTR, poor conversion | The ad promised something the page does not deliver | Fix the page first: photos, price clarity, delivery info |
| Poor CTR, good conversion | The product sells but the creative is invisible | Three new creatives, same audience |
| Poor on both | Wrong product, wrong audience, or wrong price | Stop. Do not spend more to find out |
| Good numbers, no profit | Margin is too thin to buy traffic | Raise price or increase order value before advertising again |
That fourth row is the one people refuse to accept. A test that says "no" has done its job and saved you ₹40,000.
What to make the ads about
At small budgets, the creative carries almost all the weight. Three things consistently work for small Indian stores:
- The maker, talking. Thirty seconds, phone camera, no script, explaining what it is and why you make it. It outperforms polished product films more often than anyone expects.
- The product doing its job, filmed plainly. The candle burning, the bag packed for a day out, the mug being filled.
- Before and after, or scale. Anything that shows size, texture or a real difference. Shot on a phone, in a real room.
What consistently fails: stock footage, text-heavy graphics, discount-first messaging before anyone knows who you are, and anything with a watermark from a template app.
Scaling without breaking it
Once the numbers work:
- Raise budget by about 20% every three to four days. Large jumps push the campaign back into learning and costs spike.
- Add creatives before adding audiences. Creative fatigue is the usual reason performance decays at small scale.
- Watch frequency. Above about 3 in a two-week window, the same people are seeing it repeatedly and cost per purchase rises.
- Keep a retargeting set small and always on — people who viewed a product or added to cart. It is cheap and it is not where growth comes from.
- Check margin monthly, not vibes. Ad platforms report revenue, not profit. A 3× ROAS on a 30% margin product is roughly break-even.
When not to run ads at all
- You have fewer than ten organic sales.
- Your contribution margin per order is under ₹200. The arithmetic almost never works.
- You cannot ship within your promised window at higher volume.
- You are out of stock on the product the ad features.
- You have a wedding, an exam or a holiday in the next two weeks. Ads generate messages, and unanswered messages are worse than no ads.
Frequently asked questions
What is the minimum budget for Instagram ads in India?
For a meaningful test, around ₹200–₹300 a day for at least 10–14 consecutive days. Below that the platform cannot optimise and you cannot distinguish signal from noise. Ten days at ₹300 beats thirty days at ₹100.
Should I boost posts or use the ads manager?
Ads manager, with a sales objective. Boosting optimises for engagement, which produces likes from people who will never buy. It feels like progress and is not.
How long before ads start working?
Give any campaign 5–7 days before judging it. The first three days are almost always the worst as the system learns. Most small budgets are killed on day two, which is why they never work.
Do I need a website, or can I sell through DMs?
You can start in DMs, and many Indian sellers do. But paid traffic arrives at all hours, and a DM-only funnel loses the buyer who wanted to check out at 11.40 pm without talking to anyone.
What is a good ROAS for a small store?
It depends entirely on your margin. At a 55% contribution margin you need roughly 1.8× to break even on the ad spend alone; at 30% you need over 3×. Work from your own number rather than a benchmark from somebody else's business.