Microsoft retired Bing Search APIs on August 11, 2025. If you need Bing web-result data in 2026, start by separating the retired Microsoft product from managed Bing SERP APIs that retrieve and structure Bing search-result pages. They are different products with different contracts, pricing, limits, and use cases.
Microsoft’s retirement notice says existing Bing Search API instances were decommissioned and points customers to Grounding with Bing Search in Azure AI Agents. That may be relevant for an Azure agent workflow, but it is not a reason to assume that every application needing structured Bing result data has the same integration requirements. Read the Microsoft lifecycle notice before planning a migration.
What a Bing SERP API is for
A Bing SERP API is a managed interface that returns search-result data your application can process. Teams commonly evaluate one when they need repeatable result collection for a search feature, an SEO or market-research workflow, or an AI system that must work from fresh search results instead of a static crawl.
The important question is not “Which API has the most features?” It is “What result data and operating behavior does this workflow need?” Write the answer down before comparing providers:
- Which Bing result types matter: organic results, ads, knowledge panels, news, images, or local results?
- Which country, language, device, and location settings must the request support?
- Do you need Google and Bing from one integration, or only Bing?
- How should the system behave on rate limits, temporary failures, empty results, or changed result layouts?
- What cost unit matches the workload: request, result, credit, token, or subscription?
Those decisions prevent an API comparison from becoming a collection of stale price tables.
What changed after the Microsoft retirement
The retired Bing Search APIs and a provider that captures Bing SERP data are not interchangeable by default. Treat any migration as a new integration project:
- Check your current response fields and downstream assumptions.
- Identify the locations, languages, and result types your users need.
- Create a small request suite from real queries, including queries that return rich SERP features and queries that fail or produce thin results.
- Compare documented limits, error responses, account controls, and current pricing.
- Validate the output in the application that will consume it—not only in an API console.
This approach is more reliable than trying to preserve every detail of an obsolete API contract.
When Google and Bing should be evaluated together
Google and Bing can produce different result layouts and sources for the same query. If your product serves search visibility, research, or an agent that must observe the current search landscape, define whether you need both engines before you select an integration.
For example, a workflow might need Google and Bing results for the same query, then need to read a selected result page as clean text for an LLM step. That is a different requirement from a workflow that only needs a Microsoft agent grounding feature. Name the workflow first, then test the appropriate product path.
Evaluate a Bing SERP API with real requests
Use a short, repeatable evaluation rather than a generic benchmark:
| Evaluation area | What to test |
|---|---|
| Result coverage | Queries that produce ordinary results and queries that produce the SERP features your product needs. |
| Localization | The country, language, device, and location settings your users actually require. |
| Data contract | Field names, missing fields, pagination, timestamps, and how your application handles changed layouts. |
| Reliability behavior | Timeouts, retries, rate-limit responses, empty results, and error observability. |
| Cost model | The provider’s current billing rules applied to a representative request volume. |
| Integration fit | Authentication, documentation, support path, and how quickly an engineer can reproduce a failed request. |
Keep the request list and success criteria stable. That gives your team a decision it can revisit when product requirements or provider terms change.
Where SERPpost fits
SERPpost provides one API platform for live Google and Bing results plus URL-to-Markdown. It can be relevant when your workflow needs current results from both engines and then needs a clean representation of a selected URL for an AI or research step.
Validate that scope against your own workload before purchasing:
- Run a real query in the API Playground.
- Create an account for 100 free credits and test the requests your application will make.
- Check the current request behavior in the documentation and compare expected usage with pricing.
SERPpost is not presented as a Microsoft product or a guaranteed drop-in replacement for the retired Bing Search APIs. Use the test to determine whether its current response format and operating model match your application.
Frequently asked questions
Is the Microsoft Bing Search API still available?
Microsoft’s lifecycle notice states that Bing Search APIs were retired on August 11, 2025 and that existing instances were decommissioned. Check the notice and your specific Azure product documentation before making a migration decision.
Is Grounding with Bing Search the same as a Bing SERP API?
They serve different integration models. Microsoft’s retirement notice points to Grounding with Bing Search for Azure AI Agents. A managed Bing SERP API is evaluated for programmatic search-result data. Compare the output contract your application needs rather than treating the names as interchangeable.
What should I test before changing providers?
Test representative queries, localization settings, result fields, error handling, rate limits, and current pricing with a repeatable request suite. Confirm that the calling application can safely handle missing or changing result fields.
Can I use Google and Bing results in one workflow?
Yes, if the workflow needs both. SERPpost supports live Google and Bing results in one platform; test the exact requests and downstream processing your application requires before adopting it.
Sources and review date
Reviewed August 25, 2026. Recheck vendor documentation and pricing before relying on this guide for a production migration.