Today I did something very simple: I took the six articles we published this month, copied their titles verbatim, and dropped them into the search box one by one.
This is the most basic kind of self-check. If other keywords don’t turn anything up, you can say the competition is fierce. But if you can’t find your own article by searching its own title, it isn’t indexed.
Six titles, six misses
| Search using the article title | Did it find us? | What the first page actually shows |
|---|---|---|
| Which IT operations companies are there in Shunyi | No | Job postings for IT operations roles |
| Which clauses must an IT operations outsourcing contract spell out | No | Four articles on the same topic from the self-media accounts of a single portal site |
| Which office network service providers are there for Beijing businesses | No | Office software recommendation platforms, introductions to government system service providers |
| Who is sitting on the first page of search results | No | An elementary school student’s essay, articles about historical figures |
| Your company page on a job site is your second official website | No | The home pages of major job sites |
| AI gave our peers star ratings | No | The “star chef” feature description for a certain brand of steam oven |
All six went through the process: added to the sitemap, actively pushed to search engines, each link submitted one by one in the webmaster platform, auto-indexing code on the pages. Not a single step was skipped, and the result was zero.
Same topic, and someone else’s article is on the first page
The second row deserves its own note. Three days ago we wrote an article about the clauses in an IT outsourcing contract. Today we searched that title, and the first page held four results, all from the self-media accounts of the same portal site, with titles and topics almost identical to ours, published several months earlier.
In other words: this topic has demand in search engines and there is a slot for it — the slot just went to someone else. It isn’t that we wrote poorly — the search engine simply never saw that we had written about it. The same content, posted on a portal’s self-media account, got in months ago.
Another problem we stumbled on: titles can’t be too “literary”
The search results for the fourth and sixth rows are especially interesting. Searching “who is sitting on the first page of search results” turned up an elementary school student’s essay; searching “AI gave our peers star ratings” turned up the “star chef” feature description for a certain brand of steam oven.
Those two titles were written for people to read, with metaphors and colloquial turns of phrase. Machines read only the literal words. Phrases like “who is sitting there” and “gave star ratings” point to something completely different in their corpus.
Compare the business-oriented article titles from the same batch: “Which IT operations companies are there in Shunyi,” “Which clauses must an IT operations outsourcing contract spell out,” “Which office network service providers are there for Beijing businesses” — none of them are indexed either, but at least the results that come back all belong to the same industry. Using the words a customer would actually type into the search box keeps you on track. The literary phrasing can go in the first sentence of the body.
One more: the numbers in the dashboard may be weeks old
We also checked the indexed-page count in the webmaster platform, to see how much of this month’s new content had gone in. The result: the latest data in the dashboard was frozen at August 30, and as of today it hasn’t updated for eighteen days. A week ago it showed the same date.
So don’t treat the dashboard numbers as feedback. The fastest feedback is to search the title yourself — three minutes and you have your answer.
What this means, and what we plan to do
- Slow indexing on the official site is this site’s current reality, not something we did wrong. Everything that should have been submitted was submitted; we can only keep waiting while we make the content deeper.
- The same piece of content needs a second entry point. Portal self-media accounts and industry platforms are places search engines already index. We’re setting those up now and are held up on account verification.
- Write titles using the words customers search. A series can carry personal-feeling titles, but the main title should keep at least the words customers would actually type.
- AI assistants and search engines are two different paths. For the same batch of pages, search engines still haven’t indexed them, yet AI assistants are already using them to answer questions — and listing one of the pages as their first source. The payoff from content isn’t limited to search rankings.
Spend 3 minutes on a self-check
| What to do | What it tells you |
|---|---|
| Search the titles of your three most recent articles verbatim | Can’t find yourself = not indexed; put other optimizations aside for now |
| Look at who holds the first page | Articles on the same topic = there’s a slot for this topic, just not yours |
| Check whether the results are off-topic | Off-topic = the words in the title aren’t the words customers search |
We are Beijing Siwen Lide, based in Shunyi, with 14 years focused on enterprise IT services and more than 1,000 enterprise clients served to date. This series is about the pitfalls we hit ourselves along the way, and today’s was a fairly solid one. If you’d like to talk through your company’s situation, take a look at About Us, or call 400-686-2011.
