Zomato & Swiggy Ads: How to Rank and Convert on Food Aggregators

How aggregator visibility actually works

Placement on the food apps is a mix of paid and earned. Ads buy you higher or more prominent placement, but the platform still weighs relevance and quality. Your rating, order volume, acceptance and delivery reliability all feed how you show up, paid or not. That means a badly-run kitchen can’t simply buy its way to the top for long. Ads amplify a listing; they don’t rescue a weak one.

The levers that move the numbers

Once the fundamentals are in place, these are the dials worth turning:

• Ad placement and targeting — when and where you bid, tuned to your delivery radius and peak hours.
• Hero photography — the top few dishes carry most of the click; professional food shots earn their cost quickly.
• Menu structure and pricing — bestsellers and combos surfaced first, priced for margin, not just volume.
• Ratings and reviews — the single biggest conversion lever, and the slowest to build.
• Offers — used deliberately, not as a permanent discount that trains customers to wait.

How we approach an aggregator account

1. Fix the page first — photos, menu order, descriptions and pricing before a rupee goes to ads.
2. Read the economics — know the contribution margin per dish so we promote what makes money.
3. Bid to peak intent — concentrate spend on the hours and areas that convert.
4. Build the rating engine — consistent quality plus a nudge for reviews from happy customers.
5. Measure to profit — judge ads on margin after commission and delivery, not gross orders.

What most restaurants get wrong

The number-one mistake is scaling ad spend on a listing that isn’t ready, poor photos, a cluttered menu, a mediocre rating. You’re paying to send more people to a page that doesn’t convert, which just loses money faster. The second is discounting as a habit: a permanent offer erodes margin and teaches customers to only order when you’re cheap. The third is judging success by order count instead of margin after the platform’s cut and delivery cost. A busy kitchen losing money on every order is not a win.

The contrarian take: more ad spend won’t fix a bad menu page

The instinct when aggregator sales dip is to raise the ad budget. Usually that’s the wrong lever. If your rating is average and your top dish has a dark, badly-lit photo, more visibility just means more people bouncing. The highest-ROI ‘ad optimisation’ on Zomato and Swiggy often isn’t the ad at all, it’s a better hero image, a cleaner menu, and a rating you’ve earned. Spend on the page before you spend on placement.

What tends to improve, a realistic picture

• Business type: a delivery-first restaurant or cloud kitchen leaning on aggregator ads.
• Common problem: rising ad spend with flat or unprofitable orders.
• Typical approach: rebuild the menu page and photography, price for margin, concentrate ad spend on peak intent, and build a review habit.
• What tends to improve: conversion on the visits the ads already buy, and clearer margin visibility, which can lift return on the same spend. Outcomes vary by cuisine, city and competition.

Audit your aggregator listing in 10 minutes

1. Are your top three dishes shot professionally, or on a phone in bad light?
2. Is your bestseller and a combo visible without scrolling?
3. What’s your rating, and when did you last actively ask for reviews?
4. Do you know the margin on each promoted dish after commission and delivery?
5. Are you running a permanent discount that’s become invisible to customers?

Close those gaps before touching the ad budget. It’s the same discipline in our cloud kitchen marketing and wider restaurant marketing, and it pairs with building direct orders that dodge commission.

One order, taken apart rupee by rupee

Most operators know their commission percentage and stop there. That single number hides everything. Take one order apart, line by line, because it is the only way to see the account clearly. The figures below are illustrative, so open your own settlement report and run the same exercise with the real ones, ideally on your three highest-volume items rather than an average that flatters everything.

Take a ₹500 order. Platform commission comes off first, and say that is ₹100. You are running a promotion that discounts twenty percent, another ₹100, funded by you rather than the platform. Payment gateway charges take a small slice, perhaps ₹10. Packaging on a two-dish order with gravy might be ₹30, and you are absorbing the delivery charge on orders above a threshold, another ₹25. Food cost at thirty percent of the menu price is ₹150. That leaves ₹85 against your rent, your kitchen staff, your utilities and your ad spend, on a ticket you booked at ₹500.

Now put the ads in. If your ad spend across the month works out to ₹40 on every order it influenced, that ₹85 becomes ₹45. The order is still profitable, barely, and it stops being profitable the moment you raise the discount to thirty percent to chase a ranking bump. That is the trap. Volume rises, the dashboard looks healthier, and the bank balance does not move.

The useful output of this exercise is a floor. Work out the discount level and the ad cost per order at which a given item stops paying for itself, write those two numbers on a card. Then hold the line. Especially during a festival push, when everyone around you is doing exactly that.

The four ratios that locate your actual problem

Aggregator dashboards give you impressions, menu opens and orders. Three numbers, four ratios. Between them they will tell you which part of the listing is broken far more reliably than any hunch.

First, impressions to menu opens. This is a shopfront question, decided almost entirely by your cover image, your rating, the delivery time shown and whatever offer badge is attached. If this ratio is weak, more ad money buys more impressions that nobody opens, which is the most common way a restaurant wastes a promotion budget. Second, menu opens to cart additions. Now the customer is inside and looking at your dish photographs, your descriptions and your prices, so a drop here is a menu-page problem and no amount of bidding fixes it.

Third, cart to order. Losses at this step are almost always about money appearing late: delivery charge, packaging charge, taxes and a minimum-order threshold that all reveal themselves at checkout after the customer had a number in their head. Fourth, orders to repeat orders within sixty days. That one measures the food and the packaging, nothing else, and it is the only ratio on this list that advertising cannot influence at all.

Fix them in order. There is no point buying impressions for a menu page that loses people at the second step, and no point improving the menu page if the food does not bring anybody back. Diagnose top down, spend bottom up.

Ads, discounts, or photographs, given the same ten thousand rupees

Assume you have ₹10,000 spare this month and three obvious places to put it. They behave very differently over time. That difference matters more than the immediate return.

For a listing that has been live a year with a decent rating and tired photographs, the third option is almost always the right call, and it is the one operators pick last. It feels like a cost. Campaigns feel different. Ads make sense when your ratios are already healthy and you simply need more people at the top, or during the first eight weeks of a new outlet when the platform has no performance history to rank you on and paid placement is genuinely the only route to being seen at all.

Discounts are the one to be careful with. A promotion running continuously for six months is not a promotion, it is your price, and going back to the original number reads to your customers as an increase. If you discount, discount in defined windows with a stated end. Then actually end it.

A 30-day plan for an account that has stalled

Four weeks, one focus each, no heroics. This sequence assumes an existing listing that has gone quiet rather than a new launch.

Week one is measurement and menu hygiene. Pull ninety days of data, calculate the four ratios above, and identify your ten highest-revenue items along with the five that generate the most complaints. Delete the dead weight. A menu with 140 items where 30 sell is slowing every customer down and quietly wrecking your kitchen timings, which then shows up as a longer delivery estimate on the very screen where customers were deciding whether to open your listing at all.

Week two is the shopfront. New cover image, fresh photographs for the ten items that matter, descriptions rewritten so they say what the dish actually is instead of repeating its name in different words. Week three is the money layer: check what your packaging and delivery thresholds look like at checkout, decide whether a minimum-order value is protecting your margin or costing you the small orders that become regulars, and set your discount floors from the arithmetic in the earlier section rather than from what the account manager suggests on a call.

Week four is where ads finally enter. Now that the listing converts, put a controlled budget behind your two strongest dishes in the two or three pin codes where your delivery times are genuinely competitive, and leave the distant areas alone regardless of how many impressions they promise. Judge it after a fortnight against cost per order, not against impressions. If the number does not work at that budget, it will not work at double, and doubling is the reflex the platform will encourage.

One last discipline. Change one thing at a time. An account where photographs, prices, discounts and ad budgets all moved in the same week teaches you nothing at the end of the month, and you will be back to guessing, which is precisely how the account stalled in the first place.

Renting demand forever, or building some of your own

Every improvement described so far makes a rented channel work better. That is worth doing. It is also worth understanding that the commission applies to your hundredth order from the same customer exactly as it applied to the first, which means a regular who has ordered from you every fortnight for two years has been paying a distribution fee for a discovery that happened long ago and never needs to happen again.

The direct channel is where that stops. In practice, for most Indian outlets, it means a WhatsApp Business number with a catalogue, UPI collection through a QR code or a payment link, and your own delivery arrangement or a hyperlocal partner charging per drop rather than a percentage of the bill. Setup is cheap and the running cost is close to flat, so the margin advantage widens with every order rather than shrinking. None of this replaces the aggregators. It reduces how completely you depend on them.

Payment behaviour is the part to plan for rather than fight. UPI dominates repeat ordering and settles instantly, which is a genuine improvement on waiting for a platform payout cycle, and it is the reason a direct channel helps working capital as much as it helps margin. Cash on delivery still shows up, mostly on first orders and in specific pockets of every city, and it brings the usual problems of refused deliveries and riders carrying change. Keep cash for new customers. Then nudge repeat ones onto UPI with something small attached.

Start with the people already loyal. Insert a card with a QR code into every aggregator order going to a returning address, offer a genuine reason to order direct next time, and accept that the conversion rate will be modest. That is fine. Even a tenth of your regulars moving across changes your monthly arithmetic permanently, and unlike an ad budget, that shift keeps paying after you stop working on it.