Quick answer: how do you optimise Advantage+ Shopping campaigns in 2026?
Advantage+ Shopping has been renamed and rebuilt as Advantage+ sales campaigns. To optimise one, keep Advantage+ budget, audience and placements on. Replace the removed existing customer cap with a customer-list exclusion or a capped second ad set. Size each ad set for about 50 purchases between edits, and judge results on new-customer profit.
Advantage+ Shopping is now Advantage+ sales: what changed
Advantage+ Shopping campaigns (ASC) no longer exist as a separate product. Meta has renamed and updated it as Advantage+ sales campaigns. In Ads Manager you now create a normal Sales campaign that runs in Advantage+ mode.
How legacy ASC was retired
- February 2025: industry coverage reported Meta testing one setup for all sales campaigns, with Advantage+ on by default. The rollout to developers followed from May 2025.
- 8 October 2025 (API v24.0): new ASC campaigns could no longer be created through the legacy APIs.
- 18 February 2026 (v25.0): creating, duplicating and updating ASC through the Marketing API was blocked.
- 19 May 2026: the block was extended to every API version.
- v26.0 (mid-2026): industry reports citing Meta's developer documentation say remaining legacy campaigns are paused or made uneditable at this version. Campaigns that used the existing customer budget cap can only be moved by duplicating them in Ads Manager. We could not confirm the v26.0 details on a Meta page.
Old structure vs new structure
| Feature | Legacy ASC | Advantage+ sales |
|---|---|---|
| Ad sets | 1 | No stated limit |
| Ads | Up to 150 creative combinations | 50 ads per ad set |
| Existing customer cap | 0-100% | Removed |
| Audience inputs | Location only | Hard controls plus suggestions |
| Custom audience exclusions | Not available | Available |
The three settings that keep a campaign in Advantage+ mode
A campaign runs in Advantage+ mode when Advantage+ campaign budget, Advantage+ audience and Advantage+ placements are all on. Narrowing the audience, excluding placements or giving ad sets their own budgets switches it off. Excluding existing customers does not.
Location, minimum age and language are hard limits. Age range, gender, detailed targeting and custom audiences only act as suggestions. Use hard limits for real business rules, such as the countries you ship to.
Industry reports citing Meta's developer documentation say migrating a legacy ASC into the new structure puts it back into the learning phase, and the cap does not carry over. Do it in a quiet fortnight. Avoid the run-up to Diwali, Ramadan or Black Friday and Cyber Monday (BFCM), when CPMs (cost per 1,000 impressions) are already climbing.
Replacing the existing customer cap: three structures
The cap that limited spend on past buyers is gone. You rebuild that control with an exclusion or a second ad set. You then check it with audience segments.
Set up audience segments before launch
Define your segments under Advertising Settings > Audience Segments. Existing customers should include your CRM list plus purchasers. Engaged audience should include site visitors and social engagers. Meta's Advantage+ sales campaigns page recommends segments for exactly this reporting.
Read them under Breakdown > By Demographics > Audience Segments. This shows spend, purchases and cost per result for New Audience, Engaged Audience and Existing Customers.
Pick the structure that fits your repeat rate
- Option A, pure acquisition: one ad set that excludes your customer list. Use it when email and WhatsApp already handle retention.
- Option B, split budget: two ad sets under campaign budget. One is aimed at existing customers and capped with a maximum ad set spending limit. The other excludes them.
- Option C, monitor only: one ad set with no exclusion and a weekly segment review. This suits newer brands with small customer lists.
Our rule of thumb: stay on Option C while existing customers take under 20% of spend. If they take more than 25% for 2 weeks running and new-customer cost per purchase is rising, move to Option B.
Worked example: capping repeat spend at ₹1 lakh a day
Step 1: Measure the existing-customer share
Example: a skincare brand spends ₹30 lakh a month, about ₹1,00,000 a day, in one Advantage+ sales ad set. Last week's segment breakdown shows existing customers took ₹35,000 a day.
Existing-customer share of spend
₹35,000 ÷ ₹1,00,000 = 35%Step 2: Split the budget and set the cap
35% is above the 25% line, it has stayed there for 2 weeks running and new-customer cost per purchase is rising, so the brand moves to Option B. It puts a maximum spending limit of ₹15,000 a day on the existing-customer ad set. The second ad set excludes the customer list and can take the remaining ₹85,000.
Step 3: Check both ad sets can still exit learning
At a new-customer cost per purchase of ₹1,100, ₹85,000 a day buys about 77 purchases a day. At ₹400 per repeat purchase, ₹15,000 a day buys about 37 a day. Both ad sets clear about 50 events in a couple of days.
Custom audiences are only suggestions, so the existing-customer ad set may still reach some new people. Check the segment breakdown after a week to see where the money really went.
Size the budget from your own CPA, not a flat daily rule
Give each ad set enough budget to reach about 50 optimisation events since its last significant edit. That should happen within roughly a week. The flat daily minimums quoted in many guides do not appear in Meta's documentation.
Weekly budget per ad set ≈ 50 × target CPA (cost per acquisition). Example: at a CPA of ₹1,200, that is ₹60,000 a week, or about ₹8,600 a day. A ₹10 lakh a month account (about ₹33,000 a day) can feed 2 or 3 ad sets at that CPA, but not 5.
Meta's wording is “around 50 optimization events since its last significant edit”, not “50 per week”. The week is only a working approximation.
Edits that send an ad set back into learning
Significant edits include:
- changes to targeting, creative, optimisation event or bid strategy
- adding a new ad
- large budget changes
- pausing for 7 days or more
Ad sets that cannot reach the threshold show as Learning Limited. Our guide on exiting the learning phase faster covers how to fix that.
Scale in steps
Our rule of thumb: raise budget by no more than 20% at a time, with 3-4 days between changes. Only scale while profit-based results hold. If CPA climbs and frequency rises before you have scaled, your bottleneck is creative, not budget.
Batch new creative weekly so learning resets once
Adding an ad to a live ad set is a significant edit. Release new creative in one weekly batch rather than one ad a day, so each ad set relearns once a week instead of constantly.
- Variety beats volume. The 50-ads-per-ad-set ceiling is far more than most brands need. Different hooks, offers and formats teach the system more than 30 versions of one idea.
- Cover vertical inventory. Include short vertical video for Reels and Stories, plus static images for Feed.
- Retire ads on evidence. Our rule of thumb: pause an ad only after it has spent at least 2× your target CPA. Pause it if it has no purchases at that point, or a CPA above 2× the ad set average.
- Check catalog defaults. Dynamic Media is now on by default for Advantage+ catalog ads created through the API. Make sure the videos and images in your catalog are ones you want shown.
If results fade over several weeks with the same ads, read our guide to spotting creative fatigue before you touch the budget.
Reading results: segments, attribution and incrementality
Platform ROAS (return on ad spend) is the weakest number in an Advantage+ campaign. Check three things instead: who the spend reached, how credit was counted, and whether the sales would have happened anyway.
Weekly segment check
Track the existing-customer share of spend and the new-customer cost per purchase every week. A rising existing share alongside flat new-customer orders explains most cases of “ROAS is great but growth is flat”.
The 2026 attribution change
In March 2026 Meta limited click-through attribution to link clicks. Other interactions moved to a 1-day engage-through column. ROAS benchmarks from before the change are not directly comparable, so compare like with like from the change date onward.
Proving the sales are extra
Meta recommends two built-in tools: an A/B test of Advantage+ against a manual setup, and Conversion Lift to measure incremental sales. Run one before you move most of your budget into Advantage+.
Before ScaleWin's AI assistant recommends a change, it pulls the last 7-10 days and breaks them down by ad set, ad, placement, region, device and hour. It checks learning-phase status, and it compares landing-page views with link clicks to catch page drop-off that looks like an ad problem.
Turn platform ROAS into profit in each market
Meta's reported revenue is usually larger than what you keep. Convert it before you judge the campaign, and compare the result with your break-even ROAS. The complete Meta Ads optimization guide covers value rules in full. Here is how they apply to Advantage+.
- India: send purchase value without GST and shipping. Example: a ₹2,360 order that includes 18% GST is ₹2,000 of revenue. For cash on delivery (COD), adjust value for RTO (return to origin, when the buyer refuses the parcel and it comes back). Cut RTO with OTP or WhatsApp COD checks, partial COD advances and COD limits by pincode or order value. Watch UPI checkout drop-off too. Send hashed phone numbers through the Conversions API (CAPI) and aim for an Event Match Quality of 6 or more.
- UK, EU and Gulf: prices usually include VAT, so strip it out of the value you send, every time. In the Gulf, apply the same COD and RTO logic as India.
- US: sales tax is added at checkout. Exclude tax and shipping so value matches product revenue. In Canada GST/HST is usually added at checkout, so handle it the same way. In Australia, prices include GST, so strip it out as you would VAT.
- Multi-country: run separate campaigns per region, with values in a consistent currency.
Refunds also skew the numbers. Shopify nets refunds out of revenue, but Meta never sees them unless you send them. Tax, shipping, discounts and duplicate events cause most other gaps. Our Pixel and Conversions API setup guide covers deduplication.
What most guides get wrong about Advantage+ Shopping
Most ranking guides were written for legacy ASC and have not caught up. These are the claims to ignore.
- “Set your existing customer cap to 20%.” The cap has been removed. Use an exclusion or a capped second ad set instead.
- “ASC has only one ad set.” Advantage+ sales allows several ad sets, with up to 50 ads in each.
- “Select Advantage+ shopping in Ads Manager.” Create a Sales campaign and keep budget, audience and placements in Advantage+ mode.
- “You only need 25 conversions a week.” We found no Meta source for this. Meta's guidance is about 50 optimisation events since the last significant edit.
- “Compare ROAS with last year's benchmarks.” The 2026 attribution change makes older ROAS figures hard to compare with today's.
- “You get no audience input.” You now have hard controls, suggestions and custom audience exclusions.
When a manual sales campaign is the better choice
Advantage+ sales is not the right tool for every goal. Use a manual campaign in these three cases.
- Retargeting-only goals: industry guides citing Meta's documentation say Advantage+ audience is not recommended for retargeting campaigns. Because custom audiences are only suggestions, delivery can go beyond your retargeting pool anyway.
- Restricted categories or strict age rules: when you need tight targeting, the controls you add will switch Advantage+ off anyway.
- Very low purchase volume: if no ad set can reach about 50 purchases between edits, consolidate first. Alternatively, optimise for a higher-volume event until purchases build up.
Your first 30 days, with one change window a week
- Week 0: confirm Pixel and CAPI, net-of-tax values and audience segments. Choose structure A, B or C.
- Weeks 1-2: hands off. Read segments and learning status, but do not edit.
- Week 3: release the first creative batch, retire clear losers and check the existing-customer share.
- Week 4: scale by up to 20% if profit holds, restructure if the existing share is above 25%, or start a Conversion Lift or A/B test.
Read next in this series
Key takeaways
- Legacy ASC is retired. You now build a Sales campaign that runs in Advantage+ mode, and migrating an old ASC puts it back into learning.
- The existing customer budget cap no longer exists. Rebuild it with a customer-list exclusion or two ad sets, with a spending limit on the existing-customer one.
- Size each ad set at about 50 × your target CPA per week, not a flat daily minimum.
- Add new ads in one weekly batch. Every new ad is a significant edit.
- Read audience segments every week and convert platform ROAS into profit after GST or VAT, shipping, refunds and RTO.
Meta ads optimization: frequently asked questions
Is Advantage+ Shopping still available in 2026?
What replaced the existing customer budget cap?
Does excluding existing customers turn Advantage+ off?
How many ads can I put in an Advantage+ sales campaign?
How much daily budget does an Advantage+ sales campaign need?
Will migrating my old ASC campaign reset learning?
Why is my Advantage+ ROAS high but new customer growth flat?
Can I use Advantage+ sales for retargeting?
Sources
- Meta: Advantage+ shopping ads (renamed to Advantage+ sales campaigns)
- Meta: Advantage+ sales campaigns
- Meta Marketing API changelog (via Releasebot)
Let ScaleWin read your ad data every day
ScaleWin checks your Meta and Google Ads daily, judges every ad on profit and tells you exactly what to change.
Start free, no card



