Content API for Shopping stopped being available on 18 August 2026, and from 1 September the calls that kept arriving started returning progressive errors. If nobody on your side confirmed the migration to the Merchant API in writing, the honest position is that you do not know whether your product data is still updating — and this is a bad quarter to find out late.
Google shut down Content API for Shopping on 18 August 2026 and remaining requests began erroring progressively from 1 September. Only API integrations are affected: manual uploads, scheduled fetches and Google Sheets feeds still work. The failure is silent — products stop updating, then stop serving, over weeks. Check your feed's last update timestamp, active item count and product impressions today, and get written confirmation from whoever owns the integration.
What happened, and on what dates
| Date | What changed | What it means for a merchant |
|---|---|---|
| Through 2025–2026 | Merchant API published as the replacement; Content API marked deprecated | The migration window most integrations were meant to use |
| 18 August 2026 | Content API for Shopping access ends | The formal shutdown date |
| 1 September 2026 | Remaining Content API requests begin returning progressive errors | Unmigrated integrations start failing, gradually rather than all at once |
| September 2026 onward | Error rates escalate; Google offers an extension request form for teams needing more time | The window for an orderly fix is closing, and Q4 traffic is arriving |
"Progressive errors" is the phrase doing the quiet damage. A hard cut-off would have generated an obvious incident in the last week of August. A ramp generates something much easier to misread: some writes succeed, some fail, the catalogue drifts out of date, and nothing in the interface announces a structural failure.
Who is actually affected
This is the first question to settle, because a large share of merchants reading about the shutdown are not affected at all.
| How your products reach Merchant Center | Affected? | What to check |
|---|---|---|
| Manual file upload | No | Nothing — unaffected by the shutdown |
| Scheduled fetch from a URL | No | Nothing, though feed freshness still matters |
| Google Sheets feed | No | Nothing |
| Platform app or plugin (store platform, marketplace connector) | Yes, via the vendor | Vendor's migration status, and your app version |
| Feed management platform | Yes, via the vendor | Written confirmation that your account, not just the platform, is migrated |
| Agency or in-house custom integration | Yes | Whether the code calls Merchant API endpoints, and the date of the last successful write |
| Mixed setup (main feed by API, supplemental by Sheet) | Partially | The API side only — and mixed setups are the ones that look healthiest while broken |
The mixed case deserves the warning. If a supplemental Sheet is still updating prices or availability while the API-driven primary feed has stalled, Merchant Center looks active and recent. The rot shows up in the products that stopped being added, not in the ones still being edited.
Why this fails without telling you
An integration failing this way does not announce itself. Campaigns do not pause with a banner. Products simply stop updating, then stop serving, over a period of weeks that looks from the outside like ordinary listing churn. Here is how the symptoms map to the cause, so you can tell this apart from the ordinary problems that produce the same-looking chart.
| Symptom | Could be this shutdown | Could also be | What separates them |
|---|---|---|---|
| Shopping impressions drifting down over 2–4 weeks | Yes | Seasonality, bid or budget change, competitor pressure | Check whether active item count fell in step; a bid problem does not reduce item counts |
| Rising count of expired products | Strongly | Genuinely discontinued lines | Expiry clustered by date rather than by category points at a sync failure |
| New products never appearing | Strongly | Disapprovals, missing attributes | Disapprovals are listed with reasons; a silent absence is not a disapproval |
| Prices or stock stale on Google but correct on site | Yes | Cache or crawl delay | A delay resolves; a failed integration keeps diverging |
| Everything fine in Merchant Center, revenue down | Possibly | Site, checkout or tracking problem | Compare against your own store analytics before blaming the feed |
The ten-minute audit
Anyone with Merchant Center and Google Ads access can do this. No developer required.
- Open your primary feed and read the last update timestamp. Compare it against how often it is supposed to sync. Older than the interval is the first red flag; older than 18 August is close to conclusive.
- Note the active product count, then compare it with four weeks ago. A steady catalogue with a falling active count is not a merchandising story.
- Check expired and disapproved items. A rising expired count with no corresponding disapproval reasons is the signature of a feed that has stopped being refreshed.
- Pick three products you changed recently — a price edit, a stock change, a new SKU — and confirm each is reflected in Merchant Center. This is the fastest single test.
- Look at product-level impressions in Google Ads over six weeks. A gradual slide with no bid, budget or seasonal explanation is consistent with products quietly falling out of eligibility.
- Ask the four questions below, in writing.
Steps 1 and 4 together take about five minutes and settle most cases. If both look healthy, you are almost certainly fine and can stop.
Do step 4 first if you are short of time. A price change you made yesterday, visible in Merchant Center today, proves the write path works — which is the single fact everything else in this article is trying to establish.
The four questions to send your feed provider
Most teams managing feeds through a platform, an app or an agency assume that layer handled the migration. Usually it did. Occasionally it did not, and the two are indistinguishable from the outside until listings start disappearing. Ask these in writing, because a verbal reassurance and a migrated account are different things.
- Is our specific Merchant Center account migrated to the Merchant API? Not "does the platform support it" — ours.
- What was the date and time of the last successful write to our account? A concrete timestamp, not "it's syncing".
- Are any Content API calls still being attempted on our behalf? Legacy code paths often survive behind a migrated main path, particularly for supplemental data, promotions and inventory updates.
- Has an extension been requested, and if so until when? If the answer is yes, you now have a deadline you did not know about.
If the answers are slow or vague, treat that as the answer. This is exactly the kind of dependency that should be checked before peak season rather than during it — the same reasoning behind the checkout tracking audit we recommended after Shopify's extensibility deadline.
If you are not migrated, do this in order
The instinct is to start the development work immediately. That is the second step, not the first.
1. Keep the listings alive. Scheduled fetches and Google Sheets feeds are unaffected by the shutdown. Exporting your catalogue to a hosted feed file or a Sheet and pointing Merchant Center at it keeps products serving while the real integration is rebuilt. It is not elegant, and it converts an outage into a delay — in late September that trade is worth making.
2. Establish what was actually lost. Products that expired during the gap need to come back, and anything that changed price or availability while the feed was stale may have been serving wrong data. Wrong prices in Shopping are worse than absent ones, both for conversion and for your Merchant Center standing.
3. Request an extension only if you genuinely need one. Google provides a Content API for Shopping extension request form. An extension buys time; it does not change the destination, and a team that needed an extension in September will need a plan in November.
4. Then rebuild on Merchant API. Scope it as an integration project rather than a config change, and have whoever does it document which sub-APIs they use, so the next deprecation is a known quantity.
What changed in the API, in merchant terms
You do not need this to fix the problem, but you do need it to sanity-check an estimate. The Merchant API is not a version bump. It is split into sub-APIs with a new URL structure, supports both REST and gRPC, identifies resources by name rather than by id, addresses child resources through a parent field, and expresses prices as a micro-denominated amount with a separate currency code rather than the old price format.
Two practical consequences follow. A quote of a couple of hours is unlikely to be a real migration; a quote of several weeks for a standard catalogue is probably including work that is not this. And if your integration also handles promotions, local inventory, or returns and review data, each of those is its own surface with its own migration — which is where "we migrated" and "we migrated everything" stop meaning the same thing.
Google has also published a Merchant API MCP service in alpha, giving AI coding assistants authoritative API documentation and read-focused access to Merchant Center data. For an in-house team doing the migration this is worth knowing about; for a merchant it changes nothing about the deadline.
The broader lesson for anyone running Shopping
Feed integrations are the least-watched critical dependency in most e-commerce stacks. They are installed once, they work invisibly for years, and they are usually owned by nobody in particular: the platform vendor assumes the merchant is watching, the merchant assumes the agency is, and the agency assumes the app handles it. A deprecation with a published date and a two-year runway still caught a meaningful number of stores, which tells you the notice period was never the binding constraint — ownership was.
The cheap fix is a quarterly integration review: list every system that writes to Merchant Center, Google Ads, your analytics and your email platform; name an owner for each; and record the date of the last verified successful write. Fifteen minutes a quarter. It is the same discipline that catches a broken conversion tag before a quarter of reporting is lost, and it belongs in the same review as your ongoing site maintenance. If feed health and Shopping performance are genuinely unowned in your business, that is a gap worth closing before Q4 rather than after — it is part of what we take on in performance marketing engagements.
Frequently asked questions
When did Content API for Shopping stop working?
Google provided access to Content API for Shopping until 18 August 2026, and from 1 September 2026 remaining requests began returning progressive errors. It is replaced by the Merchant API. Because the errors ramp rather than switch off at once, an unmigrated integration can appear to keep working for a while before products stop updating and then stop serving.
Is my Google Merchant Center feed affected by the Content API shutdown?
Only API-based integrations are affected. Manual file uploads, scheduled fetches from a URL and Google Sheets feeds continue to work normally. If your products reach Merchant Center through a plugin, an app, a feed management platform, an agency's connector or custom code, that path uses the API and needs to have been migrated to the Merchant API.
How do I tell whether my Google Shopping feed has stopped updating?
Open Merchant Center and check the last update timestamp on your primary feed against your expected sync schedule; a timestamp older than your sync interval is the first signal. Then compare active product counts week over week, look for a rising count of expired items, and check whether product-level impressions in Google Ads have drifted down over recent weeks without a bid or budget change. Three of those four pointing the same way is enough to escalate.
What should I ask my feed provider about the Merchant API migration?
Ask four questions in writing: is our specific account migrated to Merchant API, on what date did the last successful write occur, are any Content API calls still being attempted on our behalf, and has an extension been requested. A general statement that the platform supports Merchant API is not an answer, because support at the platform level and migration of your account are different facts.
What can I do if my integration is not migrated yet?
Keep the listings alive first, then fix the integration. A scheduled fetch or Google Sheets feed is unaffected by the shutdown and can carry your catalogue while development work happens, which turns an outage into a delay. Google also provides a Content API for Shopping extension request form for teams that need more time, but an extension is a grace period, not a plan.
What changed in the Merchant API compared with Content API?
The Merchant API is split into sub-APIs with a new URL structure and supports both REST and gRPC. For developers the practical changes are that resources are identified by name rather than id, child resources are addressed through a parent field, and prices use amountMicros with a currencyCode instead of the previous price format. It is a genuine rewrite of the integration layer rather than a version bump, which is why quotes for the work vary so much.
The takeaway
The shutdown itself is unremarkable: an API was deprecated with years of notice and then switched off. What makes it worth ten minutes of your afternoon is the shape of the failure. Progressive errors, a catalogue that decays instead of breaking, a Merchant Center that looks broadly normal, and a slide in impressions that every advertiser has a more familiar explanation for. Open your feed, read the last update timestamp, check that a change you made this week is showing, and send four questions to whoever owns the integration. If everything is current, you have lost ten minutes. If it is not, you have found it in September rather than in the third week of November.