Local Inventory Ads Default 2026: What to Change

On August 31, 2026 Google stopped asking for permission: local inventory ads default behaviour is now baked into every Shopping campaign. The Campaign.ShoppingSetting.enable_local field has no effect any more, and on Google Ads API v25.1 and later, explicitly setting it to false returns OPERATION_NOT_PERMITTED_FOR_CONTEXT. Shopping campaigns were simply aligned with Performance Max for retail, where local offers have been serving for a long time.

For pure online sellers with no local feed, there is no practical impact. For anyone with physical stores, this is a real change: one campaign now contains two fundamentally different kinds of demand — “ship it to me” and “I’ll pick it up today” — with different attribution, different average order values and different business value. Below: what to check, how to split the channels, and what your reports will start doing.

The core takeaway: if you have stores and a single target ROAS on the whole Shopping campaign, then since August 31 you are almost certainly optimising toward a blended metric where cheap local conversions drag the average around and the bid strategy has no idea. Splitting channels is not about tidy reports — it is about bidding on comparable numbers.

What changed: local inventory ads default with no off switch

The mechanics are simple. Local inventory ads used to be a campaign-level option: turn it on and products from your local feed served with a “available nearby” treatment; turn it off and only online offers ran. Since August 31, 2026 there is no campaign-level switch — the local channel is always considered enabled.

What Google gains: a complete view of demand and the ability to route a shopper to whichever fulfilment path converts better. What advertisers gain: more reach where the local feed is genuinely maintained, and metric confusion where it was set up once and forgotten.

Who this actually affects

Advertiser type	What happens	What to do
Online-only store, no local feed, no stores	Nothing — there is no local data to serve	Nothing, beyond confirming no local data source exists in the account
Retailer with stores and a working local feed	Local offers switch on automatically; reach and conversion mix change	Split campaigns by channel, recalculate target ROAS/CPA
Retailer with a stale local feed	The worst case: ads for items not actually in stock, and unhappy customers	Refresh stock, hours and addresses immediately — or close the local channel with a filter
Multi-brand seller via a marketplace	Depends on the feed source	Check which data sources supply local availability attributes

How to separate online from local now

The switch is gone, but control is not — it moved into inventory filters and campaign criteria.

In the UI: campaign settings → Inventory filter. This restricts the campaign to the channel you want: online offers only, or local only.
Via the API: instead of enable_local, use CampaignCriterionService with a listing scope and the product_channel dimension set to ONLINE. The same mechanism builds a mirrored LOCAL campaign.
Via structure: two campaigns with separate budgets — one for online sales, one for local demand. For retailers with stores this is the recommended setup: different goals, different bids, comparable metrics inside each.

If you or your developer hit the Google Ads API directly, the code needs editing: on v25.1+ an explicit enable_local=false fails the operation, and on older versions the value is silently treated as true. While you are in there, check your scripts and bulk-upload templates for legacy references to that field — the wider feed-tooling shift is covered in the piece on migrating Shopping feeds to the Merchant API.

What local offers require to actually serve

Enabling the channel shows nothing by itself. A local offer only reaches the auction when all of the following hold:

A local product feed in Merchant Center — either your own availability export or a local feed partner (LFP).
Verified store locations linked to your Business Profile, with correct addresses and opening hours.
Matching identifiers between the primary and local data sources: if the id values disagree, the local offer never attaches to the product.
Current stock data. Stale availability is not “slightly less accurate targeting” — it is advertising an item the store does not have, which costs you the click and the customer.

The most common technical failure here is local offers getting disapproved over price, availability or GTIN mismatches. Diagnosis is the same as for regular products: read the disapproval reasons and fix the source rather than re-uploading the feed on a loop. Full detail in the article on product disapprovals and feed diagnostics.

What breaks in reporting and bidding

Average order value and conversion rate blend together. Local demand usually produces more clicks and a different basket: people check nearby availability and then buy in store. In the campaign report this reads as a falling site conversion rate alongside rising clicks — a familiar pattern for anyone who worked through the Merchant Center cross-platform reporting changes.

Store pickup is attributed differently from an online purchase. Part of the value moves offline and, without offline conversion uploads, is invisible to the campaign. The bid strategy then optimises against what it can see — an incomplete picture.

Budget gets diluted. One budget across two channels means the algorithm allocates between them based on visible conversion value. If you do not feed offline value back, the local channel stays undervalued — or gets overvalued off the back of cheap micro-conversions.

Target ROAS stops being comparable. A target set from purely online data now applies to a blended sample. At minimum recalculate the target; ideally split the campaigns and give each its own.

A seven-step action plan

Freeze the before/after baseline. Clicks, impressions, conversions, revenue, AOV, ROAS for the 30 days before August 31 and the 30 days after. It is the only way to separate the change from seasonality.
Check whether you even have a local data source. Merchant Center → data sources. No local feed means the change does not touch you.
Refresh local data. Stock, prices, opening hours, addresses. If you have no way to keep them current, closing the local channel with a filter is more honest than advertising availability you cannot vouch for.
Split campaigns by channel. An online campaign with product_channel = ONLINE, a separate local one, each with its own budget and goal. For product-level segmentation inside them, lean on Merchant Center custom labels.
Feed offline value back. Without it the local channel will always look worse than it is. Mechanics in the piece on offline conversion import, and prioritisation between conversion types in the article on conversion value rules.
Recalculate targets. Set new ROAS/CPA targets from blended or split data depending on the structure you chose. Move them in steps of no more than 15–20% so the strategy does not drop back into learning.
Review geography. Local demand is sensitive to the radius around stores and to how presence-versus-interest is configured; the logic is unpacked in location targeting: presence vs interest.

How to measure the effect honestly

The local channel almost always looks worse in the ad account and often looks better at the till. Three ways not to fool yourself:

Offline sales upload. Weekly is enough, partial store coverage is enough, loyalty-card matching is enough. Incomplete data beats zero data.
Geo experiment. Run the local channel in one group of regions and compare total revenue against holdout regions — that measures incremental lift rather than reallocation.
Channel-split reporting. Look at online and local offers separately; comparing blended averages between them tells you nothing.

If your offline business sells services rather than products, do not confuse these mechanics with the Local Services Ads move, which has its own logic — see the article on migrating Local Services Ads into Google Ads.

Common mistakes

“We have no stores, so this doesn’t apply” — without checking. Local data sources sometimes get connected by an agency or a developer years earlier and quietly stay. Verify the data sources yourself.

Keeping one budget across two channels and then panicking about ROAS. That is not campaign decay, it is a change in sample composition. Split first, conclude second.

Trying to fix the feed by re-uploading it. Price and availability disapprovals are fixed at the data source. The full working order is in the fundamentals piece on Google Shopping product feeds.

Setting an aggressive ROAS target immediately after splitting. New campaigns start in learning; a hard target on day one nearly guarantees underspend and choppy pacing.

How this fits the rest of 2026

Turning the local channel on by default belongs to the same series as moving Display campaigns into Demand Gen and shipping text guidelines for AI-generated assets: Google removes narrow switches and leaves advertisers control at the level of data and constraints. In that world the winners are the accounts with a clean feed, honest value signals and a sanely separated campaign structure.

If you need help launching this kind of setup on a stable account, look at Google Ads agency account rental and the rest of the PPC Rebels services.

Three structures that work

There is no universal answer: the choice depends on how many stores you have and how different your online and offline economics really are.

Structure	Best for	Upside	Risk
One blended campaign	1–3 stores, minimal local demand	Simple, needs no rebuild	Blended metrics, imprecise ROAS target, no control over the local channel
Two campaigns: ONLINE + LOCAL	Retail with a store network and meaningful offline revenue	Comparable metrics, independent budgets and goals	Needs separate oversight and offline value feedback
Geo-split local campaigns	Networks with uneven store density by region	Push budget where stores are genuinely nearby	Data fragmentation: small campaigns starve the bid strategy of conversions

Working rule: split the structure only down to the level where each campaign still collects enough conversions for the automated strategy. If a campaign ends up with a handful of conversions per week, you traded metric clarity for unstable bidding — a bad deal.

Why shopper behaviour changes, not just the numbers

A local offer is a different promise. The online listing says “buy and we’ll ship it”; the local one says “it’s nearby, collect today”. Three consequences show up in reports.

Higher CTR, lower site conversion rate. People click more often to check availability and complete more often offline. This is not “bad traffic” — it is traffic with a different completion path.
Different price sensitivity. In the local scenario, availability beats discount: “in stock now” outweighs “3% cheaper”.
Strong distance decay. Local performance falls sharply with distance from the store. It is the first thing to check when a local campaign misses target.

If you have not separated these scenarios before, it is worth refreshing the general approach in the piece on local and geo advertising.

A 30-minute diagnostic: eight checks

Is there a local data source in Merchant Center, and when was it last refreshed?
How many products are actually active in the local channel versus the online channel?
How many locations are verified, and do they all carry correct opening hours?
Do product identifiers match between the primary and local sources?
Are there disapprovals specific to local offers, and for what reasons?
How did clicks and conversion rate move in the 30 days before and after August 31?
Is an inventory filter applied anywhere, and does it match your intent?
Is offline value flowing back, and if so with what lag and coverage?

Answering those eight gives you the full picture. After that the only decision left is whether to split the structure or keep it blended on purpose.

Special case: franchises and partial store coverage

If the stores belong to franchisees but the ad account is shared, this change creates an awkward asymmetry: a common budget starts favouring the locations with better-maintained stock data. Technically fair, practically a tilt towards the tidier partners.

What teams do about it:

Split local campaigns by region or store cluster with separate budgets so spend cannot drift between franchisees.
Set minimum local feed quality requirements for participation: refresh frequency for stock, correct opening hours.
Tag products with custom labels per cluster and filter campaigns on them — far easier than maintaining dozens of feeds.
Measure each store group separately, otherwise the blended result hides both your best and your worst performers.

When the local campaign misses target

Check in this order, most common cause first:

Radius and geo settings. Too wide a radius is the usual culprit: impressions reach people half an hour from the store.
Stock freshness. Refreshing every few days means advertising last week’s inventory.
Local catalogue coverage. Often only 10–20% of the catalogue makes it into the local feed, so the campaign cannot compete on volume.
Offline value feedback. Without it the strategy undervalues the channel and lowers bids.
The ROAS target. A target carried over from online is almost always too aggressive for the local scenario.
Cannibalisation with PMax for retail. If local offers already served there, the two campaigns compete for the same impression.

FAQ: local inventory in Shopping campaigns

Can I turn local inventory off completely?

There is no campaign-level switch any more. You can still restrict a campaign to the online channel using the inventory filter in the UI or a listing-scope criterion with product_channel = ONLINE via the API — same outcome.

What happens if I have no local feed?

Nothing changes: with no availability data there are no local offers to serve. Just confirm that no local data source is connected in Merchant Center.

Why did clicks rise and site conversion rate fall after August 31?

Classic channel blending: local demand brings clicks from people checking nearby availability who then buy in store. Site conversion rate falls, actual sales do not. Read the data split by channel and upload offline sales.

What should I do about the enable_local field in the API?

Remove it from your code. It has no effect on Shopping campaigns, and on v25.1+ an explicit false throws OPERATION_NOT_PERMITTED_FOR_CONTEXT.

Do I need a separate campaign for local products?

If you have stores and bid to a ROAS target, yes. Separate campaigns give comparable metrics inside each and independent budgets; otherwise the algorithm splits between channels using incomplete value data.

How often should stock data be refreshed?

For retail, daily at minimum, and more often for fast-moving lines. Stale availability is worse than having no local channel: you pay for the click and disappoint the shopper.

Does this affect Performance Max for retail?

No. Local offers already served there, and enable_local continues to work as before for PMax and Demand Gen. The change is specific to Shopping campaigns.

Do I have to recalculate target ROAS?

Yes. A target set on purely online data, applied to a blended sample, will systematically over- or under-bid. Move it in 15–20% steps rather than all at once.

How do I prove the local channel is incremental?

With a geo experiment: enable the local channel in a subset of regions and compare total revenue — online plus offline — against holdout regions. A before/after calendar comparison mixes the effect with seasonality.

What if local offers keep getting disapproved?

Read the specific reasons in Merchant Center diagnostics. Most often it is a price, availability or identifier mismatch between the primary and local sources. Fix the data source; re-uploading the feed alone does not help.

Is a local feed partner (LFP) worth it?

If you have no availability integration and building one takes months, yes — it is a legitimate way to get local data flowing. Stock accuracy remains your responsibility either way.

Can I just leave everything as it is?

You can, if you have no stores. If you do have stores, “doing nothing” means you are optimising a campaign against a blended metric by default. That can be a valid choice — but it should be a deliberate one.

Local Inventory Ads Are On By Default: What Changed in Shopping Campaigns After August 31, 2026