Bing SERP Tracker: What to Use After the API Retired
Microsoft retired the Search API, so a Bing position now comes from three places. One is free and first-party, and most articles do not mention it.
Copy this line to your agent to record a position with the parameters that produced it.
set up https://monid.ai/SKILL.md and use a search endpoint that returns a position, then store the parameters with it
Checking a Bing position used to be one API call. Microsoft retired that API, and the replacements are not equivalent to each other: one is free and first-party but only covers sites you own, one is a mirror of Bing's actual page, and several are independent indexes that are not Bing at all and will quietly give you the wrong number. This guide separates the three, then covers the part every tracker gets wrong regardless of source, running through Monid, the OpenRouter for agent tools.
How do you check a Bing position now?
Three routes, and the first question is whether you own the site.
Bing Webmaster Tools, for sites you own
Free, first-party, and it reports impressions, clicks and average position straight from Microsoft. For any property you can verify, this is better data than anybody can sell you, because it is measured rather than sampled: it is what Bing actually showed to actual people rather than one snapshot of one query from one location.
It is also the answer most articles in this cluster skip entirely, which is odd, because for a large share of readers it is the whole solution.
The limit is that it only covers verified properties. You cannot use it to watch a competitor.
A Bing mirror, for sites you do not own
A service that runs your query against Bing and returns the results page as structured data, positions included. This is what you need for competitor tracking or for any site you cannot verify, and it is what the retired API used to be.
An independent index, which is not Bing
Several search APIs that grew after the retirement run their own crawl. They are good products, they return relevant results, and a position from them is a position in their index rather than in Bing's. The next section but one is about why that distinction matters more than it sounds.
The three routes, side by side
| Aspect | Bing Webmaster Tools | A Bing mirror | An independent index |
|---|---|---|---|
| Covers competitors | No | Yes | Yes, but not Bing's ordering |
| Position is Bing's | Yes, averaged | Yes, sampled | No |
| Cost | Free | Per call | Per call or per result |
| Setup | Verify the domain | A key | A key |
| Good for | Your own sites | Anything you must measure on Bing | Retrieval, not ranking |
Row two is the one to read twice. Webmaster Tools gives you an average across real impressions; a mirror gives you one observation of one query. Those are different measurements and mixing them in one report is how a tracker starts disagreeing with itself.
📖 See also Bing Search API Retired: What Actually Replaces It
Why did tracking Bing get harder?
Because the official programmatic route closed and the replacements were not built to be replacements.
The API that everything was built on went away
Microsoft retired the Bing Search API, which is the reason this entire keyword cluster exists. Every rank tracker, every SEO tool and every internal script that checked a Bing position went through it. The full landscape of what actually replaced it, and what each replacement is and is not, is in our guide on the retirement.
The replacements answer a different question
Most of what appeared afterwards is built for retrieval: give an agent relevant documents to read. That is a real and useful product and it is not a ranking instrument. Retrieval products optimise for returning the best documents; a rank tracker needs the documents Bing chose, in the order Bing chose them, including the ones that are bad.
An index that quietly improves on Bing's ordering is worse for this job than one that reproduces it faithfully, which is an unusual requirement and the reason substitution goes wrong here.
Bing itself became a smaller share of a bigger question
Worth saying plainly because it changes the priority rather than the method. Bing's own results now sit alongside Copilot answers, and for many queries the thing a user sees is generated rather than ranked. A Bing position is still a real measurement and it explains less of the outcome than it did. That is an argument for tracking it cheaply rather than for tracking it hard.
How do you build a Bing tracker on a schedule?
Three steps. The third is the one that decides whether the series is worth anything in a year.
For agents
Grab an API key at app.monid.ai, then paste this to your agent and hand it the key:
set up https://monid.ai/SKILL.md
It learns the whole discover, inspect, run workflow itself. More in the agent quickstart.
For humans
npm install -g @monid-ai/cli
monid keys add -k <your-api-key> -l main
Step 1. Confirm the endpoint returns a position, not a relevance band
What it does. Separates a mirror from an index before you build anything on it, using the schema rather than the marketing.
The endpoints. api.strale.io/x402/google-search returns numbered results; context.dev/web/search returns a relevance band instead.
The call.
monid discover -q "search engine results with positions"
monid inspect -p api.strale.io -e /x402/google-search
What comes back. The schema. If it carries a position integer it is a mirror and it can answer a ranking question. If it carries a relevance label of high, medium or low, it is an index and it cannot, no matter how good the results look.
That one field is the entire taxonomy and it takes ten seconds to check.
What it costs. Nothing. Discovery and inspection never bill.
Step 2. Fetch with parameters you will repeat forever
What it does. Produces one comparable row of the series.
The call.
monid run -p api.strale.io -e /x402/google-search -w \
--query '{"query": "your keyword", "country": "us"}'
What comes back. From a live run on 2026-08-27: a results array of ten entries each carrying position, title, url, snippet, date and sitelinks, plus result_count, the knowledge_graph, answer_box and people_also_ask slots, a search_parameters object echoing your query back, and a _meta block naming the upstream and the fetch time.
What it costs. Cents per call, billed per call rather than per result, so a keyword-day costs the same whatever the page returned. Current figures at monid.ai/tools.
Step 3. Store the row, not the number
What it does. Makes a measurement from six months ago defensible.
The call. A schema decision:
row = {
"engine": "bing", "keyword": kw, "checked_at": now_utc(),
"position": my_position_or_none(res),
"result_count": res["result_count"],
"had_answer_box": res["answer_box"] is not None,
"params": res["search_parameters"],
"source": res["_meta"]["provenance"]["source"],
}
What comes back. A row that survives questions you have not thought of. Two details matter more than the position. Store source, because if a provider changes its upstream your series gets a discontinuity that looks exactly like a ranking change. And record a missing position as absent rather than as a hundred, because "not on the page" and "position 100" are different facts and averaging them produces nonsense.
The general form of this argument, with the arithmetic, is in Rank Tracking API: The Hard Part Is Not the Fetch.
📖 See also Automate SERP Tracking with a Structured Google Search API
Give this to your agent
Set up https://monid.ai/SKILL.md, and then use Monid to check where my domain sits for these 20 keywords, store the position plus the search parameters and the upstream source, and flag any run that returned fewer results than usual.Is an independent index the same as Bing?
No, and treating it as one is the most expensive mistake available in this cluster, because it produces confident numbers that measure the wrong thing.
What actually differs
An independent index crawls the web itself and ranks with its own algorithm. Its results are frequently good and sometimes better for a human reader. They are not Bing's results, they are not in Bing's order, and a page ranked third there may be absent from Bing entirely.
If your report says "Bing position" and the number came from somebody else's index, the report is wrong in a way nobody will catch, because the numbers look plausible and move in plausible ways.
How to tell which one you bought
Three checks, all free. Read the schema for a numbered position versus a relevance band. Read the provider's own description for whether they crawl or mirror. And run one query you know the answer to, then compare against what you see in a browser.
That last one takes two minutes and settles it definitively. It is worth doing before a purchase rather than after a quarterly report.
When an index is the right instrument anyway
If the question is "what exists about this topic" rather than "where do I rank", an independent index is often the better tool and the ranking question was never yours. We took the three kinds of search product apart, with the legal position of each, in What Is a SERP API, and Which Kind Do You Need?, and looked at one independent index in detail in Brave Search API: Check What You Are Allowed to Keep.
Which endpoint should I use for which job?
| Endpoint | What it does | Input | Output | Best for | Billing |
|---|---|---|---|---|---|
| Bing Webmaster Tools | Real impressions and average position | A verified domain | Impressions, clicks, average position | Sites you own | Free |
api.strale.io/x402/google-search | Numbered SERP with features | Query, country | position, feature slots, search_parameters | Sites you do not own | Per call |
context.dev/web/search | Index results with relevance | Query, freshness, filters | URL, title, description, relevance | Retrieval, not ranking | Per result |
context.dev/web/scrape/markdown | Reads a competitor page that moved | A URL | Markdown, headings, JSON-LD | Investigating a change | Per call |
api.strale.io/x402/keyword-suggest | Autocomplete expansion | A seed keyword | Questions, comparisons, long tail | Deciding what to track | Per call |
Every Monid row was verified with monid inspect on 2026-08-30. The table gives billing shape rather than figures; shape drives design and current numbers live on monid.ai/tools.
Worked through: fifty keywords checked daily on a per-call endpoint is fifty calls a day, and the monthly figure is a multiplication you can do before you start rather than a surprise. The saving nobody takes is the first row: if half your keyword list is on domains you own, that half is free and should never have been on a paid endpoint at all.
When should you not track Bing at all?
Three cases, and the first one covers a lot of people reading this.
Bing is a rounding error for you. Check your analytics before building anything. For many sites Bing is a low single-digit share of search traffic, and a tracker for it is real recurring cost against a number that will not change a decision. Track it quarterly by hand and spend the effort elsewhere.
You only care about sites you own. Then Webmaster Tools is complete, free and better than a sampled third-party number, and there is nothing to buy.
You want a dashboard rather than data. Ahrefs, Semrush and the dedicated rank trackers cover Bing, ship the charts, handle the localisation matrix and give you client reporting. If somebody needs to look at this on a Monday, buy the thing that looks at it. Everything in this guide is about what the tool has to do, not about who should build it.
There is also a case for the specialists over a general endpoint. Vendors whose whole product is rank data handle device splits, geographic granularity and SERP feature taxonomy in more depth than a general search endpoint exposes. If rank data is your product rather than your instrument, buy from somebody whose product it also is.
And the disclosure: this is Monid's blog, we sell per-call access to tools, and the argument we make best is the one where you already have a pipeline and need a position inside it. If you need a chart, buy a chart.
Conclusion
There is no single Bing SERP tracker because there are three different sources and they answer different questions. For sites you own, Bing Webmaster Tools is free, first-party and better than anything you can buy, and it is the answer most of this category forgets to mention. For sites you do not own, you need a mirror that returns Bing's actual ordering. An independent index is neither, however good its results are.
The part that survives whichever source you pick is the storage. A position measured under parameters you did not record, from an upstream you did not note, is a number rather than a measurement. Store the parameters and the provenance beside the integer, treat absent as absent rather than as a hundred, and the series will still be answerable next year.
Free next step: check whether the domains you want to track are ones you can verify in Bing Webmaster Tools. That half of the job is free, and knowing which half it is changes what you need to buy. Then run monid inspect on a search endpoint and look for a position field. Start at monid.ai.
FAQ
Is Bing Webmaster Tools enough for rank tracking?
For sites you can verify, it is better than enough: it reports impressions, clicks and average position from Microsoft's own measurement rather than from a sampled query, and it costs nothing. What it cannot do is show you a competitor, because verification is the gate. The practical split is to use it for everything you own and to buy a mirror only for the domains you do not, which is usually a much shorter list than the one people start with.
Do Bing rankings still matter in 2026?
They matter less than they did and more than nothing, and the honest answer depends on your own analytics rather than on any general claim. Bing feeds Copilot and several other surfaces, so a position there can reach people who never visit bing.com, which is an argument for measuring it. It is also a small share of search traffic for most sites, which is an argument for measuring it cheaply. Look at your own referrer data before deciding how much to spend.
Can you track Google, Yahoo and Bing in one tracker?
Yes, and the thing to get right is that they are three separate series rather than three columns of one. Each needs its own parameters pinned and its own provenance stored, and averaging across them produces a number that means nothing. Yahoo has long been served by Bing's index, so those two correlate heavily and tracking both often buys you very little; check that before paying for it twice.
What should a Bing SERP report actually contain?
Position, and then four things people leave out: the number of organic results on the page, whether an answer box or knowledge panel fired, the exact parameters the query ran under, and which upstream supplied the data. Without the last two a report cannot be defended a year later, and without the middle two a flat position line hides the month a feature appeared and took the clicks. That is a schema decision rather than a tooling one, and it is free on the day you make it.
Last updated August 2026.
