Last Updated on by Amninder Singh
Industry: restaurant (pizza, wings, fast food) · Market: a city in north Punjab and the towns around it · Channel: Facebook and Instagram · Period: February to March 2026 · Budget: ₹893 in total · Client name withheld
This is the smallest budget I have written up. A restaurant spent ₹893 across five weeks: two reels boosted from its own page and one campaign for page likes. I am publishing it because most restaurant owners I meet have this much to spend, and they want to know what it buys.
The numbers
| Campaign | Spend | Reach | Cost per 1,000 impressions | Result | Cost per result |
|---|---|---|---|---|---|
| Reel 1, brand jingle (February) | ₹142.50 | 10,070 | ₹13.92 | 1,495 full views | ₹0.10 |
| Reel 2, customer going for pizza (March) | ₹371.95 | 42,456 | ₹7.78 | 8,742 full views | ₹0.04 |
| Page likes, wings video and image (February) | ₹378.55 | 5,003 | ₹63.19 | 95 page likes | ₹3.98 |

“Full views” are what Meta calls ThruPlays: the video was watched to the end, or for at least 15 seconds.
What worked
1. The second reel: 42,456 people for ₹372
The reel was a casual customer-style clip with the caption that someone is going to the restaurant for pizza, and it called the place the best in the city. It was boosted for full views to everyone aged 18 to 65 within 40 km of a nearby town, with Meta allowed to widen the audience.
| Measure | Reel 1 | Reel 2 |
|---|---|---|
| Spend | ₹142.50 | ₹371.95 |
| Video plays | 9,925 | 46,376 |
| Full views | 1,495 | 8,742 |
| Share of plays watched in full | 15% | 19% |
| Average watch time | 5 seconds | 8 seconds |
| Post engagements | 4,215 | 21,448 |
| Cost per full view | ₹0.10 | ₹0.04 |
The second reel held people three seconds longer and cost less than half as much per full view. It also had wider targeting and Meta’s audience expansion switched on, where the first reel was limited to ages 18 to 44 in a fixed list of towns. Both changes happened together, so I cannot say how much of the gain belongs to the video and how much to the settings.
2. A video earned likes at nearly the same price as an image, from far fewer people
The page likes campaign ran two ads to people interested in meat dishes in seven surrounding towns.
| Ad | Spend | Impressions | Click rate | Likes | Cost per like |
|---|---|---|---|---|---|
| Wings video, “Crispy outside. Juicy inside.” | ₹283.26 | 1,426 | 8.35% | 73 | ₹3.88 |
| Image ad | ₹95.29 | 4,565 | 1.12% | 22 | ₹4.33 |
One in twenty people who saw the wings video liked the page. For the image it was one in two hundred. Meta charged almost ten times more per impression for the video, so the cost per like ended up close. Food that moves on screen gets a stronger reaction than a still photo, and Meta sent three quarters of the budget to the video without being told to.
What did not work
1. Nothing connects the views to tables
10,237 full views and 95 new followers are counts of attention. The restaurant has no offer code, no “show this reel for a free drink” and no call or directions button on these ads, so the account cannot say whether one extra customer walked in. For ₹893 that is acceptable. For ₹8,930 it would not be.
2. The restaurant’s own city is not named in the targeting
All three campaigns list towns around the restaurant: Gurdaspur, Dasua, Kathua, Jammu, Samba, Kangra and Dharamshala. Some of the radius settings will cover the home city, and people do drive in from these towns. A person in Jammu is still much less likely to come for wings than someone ten minutes away. I would have put a 10 km circle around the restaurant first and added the outer towns second.
3. Page likes are a weak goal for a restaurant
₹378 bought 95 likes. Meta shows a page’s posts to only a small share of its followers, so those 95 people will rarely see the next post without another boost. The same money on the second reel’s settings would have bought about 8,900 more full views.
4. One impression per person
Average frequency was 1.02 on the first reel and 1.13 on the second. Almost everyone saw the restaurant once. A single 8-second view is easy to forget by Saturday evening. A smaller area shown the reel three or four times would do more for footfall than 42,000 people shown it once.
What I learnt
- Boost the reel that looks like a customer made it. The casual pizza clip beat the brand jingle on watch time and on cost.
- At this level, full views are the cheapest thing to buy. Four paise per view means ₹100 a week keeps a restaurant visible to its town.
- Change one setting at a time. Two reels with two different audiences leave a question that a single extra test would have answered.
- Tight circle first. For food, distance decides more than interest.
- Give people something to show at the counter. It costs nothing and it is the only way a small restaurant can count results.
What I would run next
- ₹100 a day for seven days on a new customer-style reel, 10 km around the restaurant, ages 18 to 45, aiming for each person to see it about three times.
- A line on the reel: show this video for a free drink with any pizza this week. Count the redemptions.
- A second ad with a “Get directions” or “Call now” button on Friday to Sunday only.
- No more page like campaigns.
Figures are from the Meta ad account and were not adjusted.
More Meta Ads case studies
- Spoken English classes, India: 1,120 leads at ₹5.98 each
- Video editing institute, Punjab: 5,787 WhatsApp enquiries
- Hebrew teacher, Israel, UK and US: leads from $0.97
- Bill comparison site, UK: 395,179 people reached for ₹6,378
- Local businesses, Punjab and Delhi NCR: eight campaigns under ₹2,200 each
- Mental health counselling: leads from $2.09 each
If you want to learn to run and read campaigns like these, see my Meta Ads course. If you want them run for you, see my Meta Ads agency page. All of my Meta Ads case studies are on the Meta Ads case studies page. To help reels get found without paying, use the Instagram SEO checklist. You can look up any advertiser’s live ads in the Meta Ad Library.
