How to Launch a Meta Sales Campaign in India: Setup, Audiences and Learning
Shreya Singh, Paid Media Analyst13 min read
Build one sales campaign with Advantage+ on: Campaign budget, Advantage+ audience and Advantage+ placements, with location, minimum age and custom audience exclusions as your hard limits. Indian D2C brands put 65% of Meta spend into Advantage+ sales campaigns, which averaged a 15% lower cost per purchase than manual sales campaigns in accounts running both. Start each ad set at the ₹2,750-a-day benchmark, aim for Meta's threshold of about 50 purchases a week and avoid significant edits while it learns. 38.5% of new ad sets are still Learning limited after 7 days, and those that exit take 6 days on average.
What has changed in Meta sales campaigns since 2025?#
Sales campaigns now open with Advantage+ on, so the old choice between an Advantage+ shopping campaign and a manual campaign has gone. You build one sales campaign and decide how far to narrow it.
| When | What changed |
|---|---|
| 6 February 2025 | Meta announced one streamlined setup, an Advantage+ "on" label and a new name: Advantage+ sales campaigns |
| 29 May 2025 | The streamlined setup reached sales campaigns, then rolled out to all advertisers over the following months |
| 2025 | The existing customer budget cap was withdrawn for new campaigns |
| 18 February 2026 | Meta's API began blocking new or edited old-style Advantage+ shopping campaigns, in every API version from 19 May |
| March 2026 | Click-through attribution for website purchases began counting link clicks only |
| By April 2026 | A Customer lifecycle strategy setting appeared in sales ad sets |
| 29 July 2026 | Marketing API v26.0 launched, the version from which old-style campaigns can't be edited |
What turns Advantage+ on#
A sales campaign shows Advantage+ on when it uses all three of these:
- Campaign budget, also known as Advantage+ campaign budget
- Advantage+ audience, in at least one ad set
- Advantage+ placements, in every ad set
Every manual feature is still there. If you narrow the budget, audience or placements too far, that section just shows Advantage+ off. Meta says a campaign with Advantage+ off still benefits from the broadest settings you can keep.
Budget, audience and placement changes#
- Budget: Campaign budget keeps Advantage+ on, with one ad set or several, and with optional ad set spending limits. Several ad sets with their own budgets turn it off.
- Audience: most settings are now suggestions, not limits. The original audience options sit behind "Further limit the reach of your ads", and using them as hard limits turns Advantage+ off.
- New customers: the Customer lifecycle strategy setting lets a sales ad set exclude the people you've defined as existing customers.
- Placements: excluding any placement, or opting out of mobile or desktop, turns Advantage+ off. If you exclude placements, "Allow limited spend to excluded placements" lets Meta spend up to 5% of the budget on each one. On a ₹2,750 daily budget, that's up to ₹137.50 a day per excluded placement.
If you still run an old Advantage+ shopping campaign#
Duplicate it into the new setup. Meta's developer documentation says old-style campaigns become uneditable from Marketing API v26.0. Those using the existing customer budget cap are paused at that point.
Open the campaign in Ads Manager and follow the prompt to duplicate it. The copy is a new campaign, so it goes through learning again.
What Indian brands are seeing#
Indian D2C brands put 65% of their Meta spend into Advantage+ sales campaigns: ₹65,000 of every ₹1,00,000. These benchmarks come from ProfitBox360 research, 2025 (client data and shopper surveys).
In accounts that ran both, Advantage+ sales campaigns averaged a 15% lower cost per purchase than manual sales campaigns. For every ₹1,000 a manual campaign paid per purchase, an Advantage+ sales campaign paid about ₹850. Part of that period used the old Advantage+ shopping campaign name. It's an average across accounts, so test it on your own.
How do you set up a new sales campaign?#
Prepare three account settings, then build one campaign with one ad set, Campaign budget and Advantage+ on.
Before you build#
- Check that purchases reach Meta. Your Meta pixel and Conversions API should send a Purchase event for every order.
- Define your customers once. In Ads Manager, open All tools, then Advertising settings, then Audience segments. Choose custom audiences for Engaged audience and Existing customers.
- Set account controls for rules that never change. In Advertising settings, Audience controls cover locations, a minimum age and brand-safety exclusions. Placement controls cover placements you never want.
Existing customer sources can include your website, app activity, catalogue, customer list, offline activity, shopping and lead forms. For a Shopify store, a customer list export plus a website purchasers audience covers both routes.
Build the campaign#
- In Ads Manager, click + Create, choose Sales and click Continue. Advantage+ shows as on at the top right.
- Under Budget, choose Campaign budget. Start with one ad set at ₹2,750 a day, the benchmark starting budget per ad set for Indian D2C brands.
- In the ad set, choose Website as the conversion location. Pick "Maximise number of conversions" as the performance goal and Purchase as the event.
- Choose a Customer lifecycle strategy: all audiences, or new customers only.
- Set the audience controls: location first, then minimum age, custom audience exclusions and languages if you need them.
- Leave Placements on Advantage+ placements.
- Add several clearly different ads before you publish. Meta advises a wide variety of creative for Advantage+ sales campaigns.
- Check the campaign score, if Meta shows one, then publish.
Move to "Maximise value of conversions" later, once purchase values arrive with each order. Launch new ad sets at least a week before a festive sale, so learning is over before traffic peaks.
Which audience controls can you use in India?#
You get four hard controls: location, minimum age, custom audience exclusions and languages. Everything else is a suggestion, unless you switch it to a hard limit and accept Advantage+ off.
| Setting | Where you set it | How it works |
|---|---|---|
| Location | Ad set, Audience, Controls | Hard limit |
| Minimum age | Show more controls | Hard limit |
| Custom audience exclusions | Show more controls | Always a hard limit |
| Languages | Show more controls | Hard limit |
| Age range, gender, detailed targeting, custom audience and lookalike inclusions | Suggest an audience | Suggestion |
| The same settings with "Use as a suggestion" unticked | Further limit the reach of your ads | Hard limit; turns Advantage+ off |
| Account locations, minimum age, brand-safety and employer exclusions, placements | Advertising settings | Hard limit for every campaign |
| New customers only | Customer lifecycle strategy | Excludes the audiences you choose |
Location#
Choose India, or only the states and cities you can deliver to. You can enter up to 250 cities, and exclude cities inside a state you've chosen.
When you pick a city or region, "Reach more people likely to respond to your ads" may be ticked by default. Untick it if you can only deliver there. Meta doesn't guarantee location accuracy, so a few orders may still come from outside your areas.
Minimum age and languages#
With Advantage+ audience, the minimum age can't go below 18. Raise it for age-restricted products; any minimum up to 25 keeps Advantage+ on. A maximum age can only be a suggestion unless you make it a hard limit.
Set languages only when your ads, website and customer support work in just those languages. Otherwise, leave the field empty and let delivery find buyers across India's languages.
Exclusions and suggestions#
Exclusions stay hard limits even in Advantage+ audience. Exclude recent buyers when a campaign's job is new customers. You can't exclude a lookalike audience while using Advantage+ audience.
Suggestions guide Meta's AI but don't fence it in. If you suggest women, your ads may still reach men when the system expects them to buy. Suggestions never go outside your controls.
Account controls#
Account controls apply to every campaign in the ad account and don't turn Advantage+ off. Changes to existing campaigns can take up to 48 hours. Use them for fixed rules, such as a legal minimum age.
Existing customers and new-customer campaigns#
Your Existing customers and Engaged audience definitions do two jobs. Sales reporting can split results into new audience, engaged audience and existing customers. The same custom audiences are the ones to pick when Customer lifecycle strategy asks what to exclude.
Customer lifecycle strategy has two options:
- Get conversions from all audiences: choose this for the lowest cost per result.
- Acquire new customers: exclude existing customers, and optionally engaged audiences. Meta warns this may raise cost per result, and says to judge it on new-customer CPA rather than overall campaign CPA.
Meta retired the existing customer budget cap for new campaigns in 2025, and its API dropped it with v25.0 on 18 February 2026. In its place, Meta suggests excluding existing customers, or giving them a separate ad set with its own maximum spending limit under Campaign budget. How much they should get is a retargeting decision that needs its own test.
What should you do when an ad set keeps learning?#
Leave it alone for its first week, then fix Learning limited by combining ad sets, raising the budget or widening the audience.
An ad set usually leaves learning after about 50 results in the week after its last significant edit. Ads Manager marks it Learning limited when it's unlikely to reach about 50 optimisation events in that week.
In Indian sales campaigns, 38.5% of new ad sets are still Learning limited after 7 days. That's roughly 2 in every 5. Ad sets that do leave learning take 6 days on average.
Edits that restart learning#
| Always significant | Significant if the change is large |
|---|---|
| Any change to targeting | Budget amount |
| Any change to ad creative | Ad set spending limit |
| Any change to the optimisation event | Bid control, cost per result goal or ROAS goal |
| Adding a new ad to the ad set | |
| Pausing the ad set for 7 days or longer | |
| Changing the bid strategy |
Meta gives no percentage for a large budget change. Its example is that a 1% rise is unlikely to restart learning, while a tenfold rise may.
With Campaign budget, changing the campaign budget or bid strategy can restart learning for several ad sets at once. Adding a new ad set, or editing one ad set, doesn't restart the others.
So batch your changes. Add new ads in one go, and raise budgets in a few planned steps.
Meta's advice while an ad set learns#
Wait to edit until the ad set leaves learning, because results during learning are less stable. Make only edits you expect to improve results. Keep ads and ad sets few, since combining similar ad sets combines their learning. Set a budget large enough to get enough results.
Why ad sets get stuck in Learning limited#
Meta lists these causes:
- a small audience
- a low budget
- a low bid or cost control
- high auction overlap with your other ad sets
- an optimisation event that happens rarely
- too many ads running at once
Learning limited isn't a penalty. It means the budget isn't being spent as well as it could be.
How to fix Learning limited#
Work through Meta's fixes in this order:
- Combine ad sets and campaigns that target similar people.
- Raise the budget, so the ad set can buy about 50 purchases a week.
- Widen the audience by removing hard limits you don't need.
- Raise or remove a bid or cost control that's too low.
- As a last step, optimise for a more frequent event, such as add to cart instead of purchases.
To track progress, open Columns, click Customise columns and tick Optimisation events. Add the Last significant edit column too. An ad set that gets enough events since its last significant edit moves from Learning limited to Active.
How much daily budget does a new ad set need?#
Budget for about 50 purchases a week: multiply your cost per purchase by 50, then divide by 7. The benchmark starting budget for Indian D2C brands is ₹2,750 a day per ad set.
Daily budget needed = 50 × cost per purchase ÷ 7
Worked example: one new ad set#
Illustrative example: one new ad set in a sales campaign, optimising for purchases.
- Meta CPM of ₹210, link-click CPC of ₹11.25 and click-to-purchase rate of 4.1%: the Indian averages
- Cost per Meta-reported purchase: ₹11.25 ÷ 4.1% = ₹274.39, about ₹274
- Target: Meta's threshold of about 50 purchases in 7 days
- Meta-reported purchases include COD orders, some of which will later be refused
50 × ₹274.39 ÷ 7 = ₹1,960 a day, or ₹13,720 a week.
At the ₹2,750 benchmark budget:
| Step | Per day |
|---|---|
| Impressions: ₹2,750 ÷ ₹210 × 1,000 | 13,095 |
| Link clicks: ₹2,750 ÷ ₹11.25 | 244 |
| Purchases: 244.44 × 4.1% | 10.0 |
That's about 70 purchases a week, so the ad set reaches 50 in about 5 days. The ₹2,750 leaves ₹790 a day of headroom over the ₹1,960 needed.
Meta's own rule of thumb lands in the same place. It says a daily budget should be at least 10 times the average cost of your performance goal. Here, 10 × ₹274.39 = ₹2,744.
Why splitting the budget backfires#
Split the same ₹2,750 across three ad sets and each gets ₹916.67 a day. That buys about 3.3 purchases a day, or 23.4 a week.
Each ad set would take 15 days to reach 50 purchases, so all three would show Learning limited. Running three ad sets at the benchmark needs ₹8,250 a day.
When your cost per purchase is higher#
The budget has to rise with your cost per purchase:
| Cost per purchase | Purchases a week at ₹2,750 a day | Daily budget for 50 a week |
|---|---|---|
| ₹274 | 70.2 | ₹1,960 |
| ₹400 | 48.1 | ₹2,857 |
| ₹550 | 35.0 | ₹3,929 |
Take your cost per purchase from your last month of Meta results. Then set each new ad set's daily budget at 50 × that cost ÷ 7, or more.
FAQ#
Can you still run a manual sales campaign on Meta?#
Yes. Every manual setting is still inside the Sales objective. Narrowing the budget, audience or placements just shows Advantage+ off on that section. Manual control suits a hard business rule, such as a delivery area. Keep other settings broad, because Meta says campaigns with Advantage+ off still benefit from broad settings.
What happens to an old Advantage+ shopping campaign?#
Meta is retiring the old format. Its developer documentation says old Advantage+ shopping campaigns can't be edited from Marketing API v26.0, launched on 29 July 2026. It adds that those using the existing customer budget cap are paused. Open the campaign in Ads Manager and follow the prompt to duplicate it into the new setup.
Does adding a new ad restart Meta's learning phase?#
Yes. Meta counts a new ad in an ad set as a significant edit, so the ad set learns again. Any change to an ad's creative counts too. Add new ads in batches, perhaps every week or two, rather than one at a time. Each batch then costs one learning period instead of several.
Should a sales ad set optimise for add to cart?#
Only when purchases can't reach about 50 a week, even after combining ad sets and raising the budget. Add to cart happens more often, so the ad set leaves learning sooner. The catch is that Meta then finds people who add to cart, not buyers. Switch back to Purchase once volume allows.
Does Learning limited mean a Meta ad set is failing?#
No. Meta says the status isn't a penalty. It means the ad set probably won't reach about 50 optimisation events in a week, so delivery can't settle. The ads may still sell at an acceptable cost. Fix it by combining ad sets or raising the budget, then judge the results.
How soon can you judge a new Meta sales campaign?#
Give it about two weeks. Indian ad sets that leave learning take 6 days on average, and results during learning are unstable. Meta can also credit a purchase up to 7 days after a link click. Judge cost per purchase on the week after learning ends, and compare it with store orders.
Shreya Singh
Paid Media Analyst, ProfitBox360Shreya writes about Meta and Google advertising for ecommerce brands. Her articles cover campaign setup, creative testing and performance measurement, connecting advertising decisions to orders, acquisition costs and revenue.
How this research is checked, and corrected