Find Out If Ads Work Before Your Shop Spends Big

Posted on 22nd Oct 2026 12:14:10 in Digital Growth Tips

Tagged as: paid ads, google ads, small business advertising, ad budget, digital growth tips

Every week someone tells a shop owner to "run ads". The advice usually arrives without a price, a target or a way to tell whether the money came back. So the owner boosts a post for Rs 500, watches the likes climb, gets no orders, and concludes that advertising does not work for small shops.

That conclusion is usually wrong, and so is the leap of faith that produced it. Advertising is a testable expense. You can spend a fixed amount, over a fixed number of days, on one clearly defined action, and then do the arithmetic to see whether it paid for itself. If it did, you spend more. If it did not, you stop, fix the weak link, and test again.

This guide sets out a two-week test a shop can run on a small budget: the numbers to write down, the mistakes that waste the most money, and the cases where ads are simply the wrong answer. It applies to Google search ads and to boosted Instagram or Facebook posts, because the arithmetic is the same for both.

What Advertising Can and Cannot Do for a Shop

Paid ads buy attention from people you have not reached yet. That is the whole product. Everything else, from the price to the stock to the voice on the phone, has to work already.

Done properly, ads do three useful things for a local shop. They put you in front of people who are already searching for what you sell, including searches that carry your locality or "near me". They reach a defined area rather than a whole country. And they fill a specific gap, such as a quiet Tuesday afternoon, a new category you want to test, or a festive week when customers are actively looking.

What they cannot do is repair a broken front door. If your Google Business Profile has no hours or photos, if nobody answers the number painted on your board, if your product photos are dark, or if your price sits well above the market, ads will simply deliver more people to that problem, and you will pay for each one. Google makes the same point from the other side: the quality of your ad and the usefulness of your landing page both feed into what you pay, so a page that does not match the ad tends to cost more.

Pick One Action That Counts as a Result

Before you spend a rupee, decide what a win looks like in one sentence: "A win is a WhatsApp enquiry from someone within five kilometres who asks about a price." Not likes, not reach, not awareness.

For most Indian shops the realistic action is one of four: a phone call, a WhatsApp message, a direction request on Google Maps, or an order form on the website. Choose one. If you count all four at once, you will not know which one the ad actually produced.

Then make the action countable. Google Ads can record calls placed from your ads, calls to a number on your website, and clicks on a phone number on a mobile site, and some local actions are recorded without any tracking code at all. If your action is a call, switch on call reporting before the test starts and note the minimum call length in your settings. If it is WhatsApp, send paid clicks to a page or catalogue with a working click-to-chat button, and note the date in your chat list so you can count the new conversation threads later.

Next, work out what one enquiry is worth to you. If four enquiries produce one sale on average, and a sale earns Rs 800 in gross margin, then an enquiry is worth about Rs 200 to you. Any cost per enquiry below that figure leaves room to grow. This single number is the difference between judging an ad by its clicks and judging it by its profit.

Set a Budget You Can Afford to Lose

Google Ads asks for an average daily budget per campaign, and the word average matters. Google may spend up to twice your daily budget on a day when traffic is busy, and your monthly ceiling works out to 30.4 times the daily figure.

• Rs 150 a day: up to Rs 300 on any single day, capped at about Rs 4,560 in a month.
• Rs 300 a day: up to Rs 600 on any single day, capped at about Rs 9,120 in a month.
• Rs 500 a day: up to Rs 1,000 on any single day, capped at about Rs 15,200 in a month.
• A 14-day test at Rs 300 a day: plan for about Rs 4,200 in total, and accept that one or two days may run higher.

Treat that total as money spent on information, not on guaranteed sales. If losing Rs 4,200 would hurt the shop, the test is too big, so drop to Rs 150 a day for two weeks. If it would not hurt at all, the test may be too small to produce enough data to judge.

Meta works the same way in principle. You set a budget, and campaign or account spending limits cap what can leave the account. Meta shows estimated reach and estimated results before you publish, and tells you at creation if the ad type you picked carries a minimum spend.

Build the Test in Six Steps

• 1. Send clicks to one page that answers the ad. A product or category page, a Google Business Profile, or a WhatsApp catalogue. It must load quickly on a phone, show a price or a price range, and carry one obvious button. A click that lands on a page with no price and no button is a wasted click.
• 2. Write down five to ten real searches. Use the words a customer would type: "tailor near Andheri West", "cake shop Dadar", "almirah repair near me", "wedding card printer Pune". Keep them specific. A broad word like "furniture" attracts people who are not buying.
• 3. Set the area tightly, then check the setting. Google lets you target a radius around a location. By default, targeting also reaches people who have merely shown interest in your area rather than being there, which suits travel and real estate but wastes money for a shop that needs walk-ins. Switch to presence only, and keep the radius small enough that a customer can drive to you.
• 4. Write two versions of the ad. Give each one a headline naming your locality or your speciality, plus one concrete line about price, delivery or timing. Skip superlatives like "best in the city". Nobody checks them, and they do not answer the question the customer is asking.
• 5. Start with clicks, not conversions. Cost-per-click bidding means you pay only when someone clicks through to your shop. Target cost-per-action bidding needs conversion history before it can work well, so leave it for the second month.
• 6. Mark your own traffic. Ask every caller "were you looking for us on Google?" and keep a separate WhatsApp greeting for ad enquiries. It is manual and unglamorous, but in the first fortnight it is more reliable than most dashboards.

Read the Numbers After Two Weeks

Do not judge an ad on day three. Google states that automated bidding needs a standard learning phase of seven to fourteen days, and that frequent manual changes to budgets, targets or conversion goals reset that window. Give the test its full run, and leave the settings alone in the middle.

At the end, write six numbers on one page: total spend, clicks, enquiries, sales, revenue, and gross margin per sale. Then do three sums.

• Cost per click is spend divided by clicks. It tells you what attention costs in your category.
• Cost per enquiry is spend divided by enquiries. Compare it with what one enquiry is worth to you.
• Cost per sale is spend divided by sales. This has to sit below your gross margin per sale, or the campaign is buying customers at a loss.

Take a worked example. A shop spends Rs 4,200 over 14 days, gets 60 clicks, nine enquiries and two sales, with Rs 1,500 gross margin on each sale. Attention cost Rs 70 a click, an enquiry cost about Rs 467, and a sale cost Rs 2,100. The two sales returned Rs 3,000 against Rs 4,200 spent, a loss of Rs 1,200. Break-even needed three sales, at Rs 1,400 each.

A second shop spends Rs 9,000, gets 12 sales, and earns Rs 1,200 on each one. Revenue is Rs 14,400, profit after ad spend is Rs 5,400, and the cost per sale is Rs 750. Similar effort, very different arithmetic, only because the margin and the conversion rate were better.

Notice what the first example does not prove. It does not prove that ads fail in that category. It proves that at a 3.3 percent click-to-sale rate and a Rs 1,500 margin, Rs 4,200 was not enough to find three buyers. The fix might be the page, the price, the radius or the offer. Blaming the platform would end the test before it taught you anything.

Five Ways a Small Ad Budget Gets Wasted

• Targeting too wide. "All India" at Rs 300 a day buys a thin sprinkling of clicks everywhere and no pattern you can learn from. Start with your own locality and the two or three next to it.
• Bidding on your own shop name. People searching your name already know you. Paying for that click usually buys a customer you would have had for free.
• Changing the budget every morning. Each adjustment restarts the learning phase, so the campaign never settles and you end up judging noise instead of results.
• Running ads into a stock-out or a festival rush. Ads during a week when you cannot pack orders convert badly and leave a bad first impression. Pause them, or point them at what you can actually deliver.
• Measuring nothing. Without call reporting, a tracked WhatsApp link or a simple "how did you find us" question at the counter, you will be guessing again, and the next decision will be as emotional as the first.

When Ads Are the Wrong Answer

There is a short list of things to fix before any ad money is spent, because each one is free and each one leaks customers all day.

• Your Google Business Profile is unclaimed, or lacks hours, photos and a category.
• Your phone number is wrong somewhere customers look, or nobody answers it during shop hours.
• You have fewer than ten reviews, or the newest one is a year old.
• Your product photos are dark, cropped, or borrowed from a supplier's catalogue.
• You cannot take an order without a conversation, because there is no price list, no catalogue and no delivery option.

Fix those first, then test. Advertising multiplies whatever is already there, including the leaks.

Your One-Page Test Sheet

Keep this on paper. It takes ten minutes to fill and it stops the conversation from drifting into opinion.

• The start and end dates, and the total budget set aside.
• The one action that counts: call, WhatsApp, direction request or form.
• The area targeted, with the radius or the pin codes.
• The five to ten searches you are bidding on.
• Your break-even numbers: gross margin per sale, and enquiries needed per sale.
• The results: spend, clicks, enquiries, sales, revenue and cost per sale.
• The decision: scale, fix and retest, or stop.

What to Do With the Result

If the test worked, raise the budget in steps of 20 to 30 percent rather than doubling it. Keep the ad and the landing page unchanged while you scale, so the only new variable is money. Watch the cost per sale for a fortnight. If it climbs sharply, you have found the ceiling of local demand at your price.

If it failed, resist the urge to blame the platform after one attempt. Change one thing, whether the landing page, the offer or the radius, and run the same test again for another two weeks. Two honest tests tell you more than a year of occasional boosting.

Either way, keep the arithmetic. One line a month in your ledger, spend against revenue from ads, turns advertising from a gamble into a line item you can manage. That is the real prize: not a viral post, but knowing what a customer costs you and what one is worth to you.

Sources

• Google Ads Help - Understanding bidding basics
• Google Ads Help - About average daily budgets
• Google Ads Help - About advanced location options
• Google Ads Help - About conversion measurement
• Google Ads Help - Manage your spend in Google Ads
• Meta Business Help Centre - How much it costs to advertise on Meta technologies