Search and Indexing

Why Getting Indexed Is Its Own Job

A page Google has never seen cannot rank. That sounds obvious. Yet most new sites launch and then wait for search engines to wander by. Weeks go by. The owner thinks the site is not working. The truth is nobody has read it yet.

We treat getting found as a job with steps, not a hope. Before launch, the site is set up in Google Search Console and Bing Webmaster Tools. The sitemap is submitted to both. After launch, every change sends a ping. And every page is built so that an engine can tell in one read what it is, where it is and who it is for.

IndexNow Pings on Every Publish

IndexNow is a simple shared signal. You tell one address that a page changed. Bing and the other engines that share it hear about it fast. No waiting for a crawl.

Our build runs the ping right after a deploy, with the list of pages that changed. Here is a working version using only the Python standard library.

import json
import os
import urllib.request

ENDPOINT = os.environ["INDEXNOW_ENDPOINT"]  # the shared IndexNow API address
HOST = "wrinklehq.com"
KEY = os.environ["INDEXNOW_KEY"]


def ping(paths):
    body = {
        "host": HOST,
        "key": KEY,
        "keyLocation": f"https://{HOST}/{KEY}.txt",
        "urlList": [f"https://{HOST}{p}" for p in paths],
    }
    req = urllib.request.Request(
        ENDPOINT,
        data=json.dumps(body).encode("utf-8"),
        headers={"Content-Type": "application/json; charset=utf-8"},
        method="POST",
    )
    with urllib.request.urlopen(req, timeout=30) as resp:
        return resp.status

The key file has to sit at the root of the site, with the key as its only text. That is how the engine knows you own the domain.

Google does not use IndexNow. For Google, the sitemap and Search Console do the work. So we do both. One ping does not cover every engine.

Visible FAQs With Matching Schema

People ask the same questions before they call. How much. How long. Do you come to me. A short FAQ on the page answers them in the customer's own words. Those questions also match what people type into search and ask AI tools.

We add FAQPage schema so engines can read the questions and answers as data. The rule is strict. Every question in the schema must show on the page, word for word. Both come from the same list in the build, so they cannot drift.

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "Do you come to my house?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "Yes. We do most repairs in your driveway."
    }
  }]
}

You will read advice that FAQs are dead. We keep them. They answer real questions, they help AI tools quote you, and they cost nothing to keep. Read more in the FAQ schema guide.

Sitemaps and Clean Signals

A sitemap is a list of every page you want found. It sounds like a formality. Done wrong, it sends mixed signals that slow everything down.

Ours is built from the same origin setting as every canonical tag on the site. So the sitemap, the canonicals and the schema all name the same address for each page. Google never has to pick between a www version and a bare one, or a slash and no slash.

We leave out pages that should not rank, like a thank you page or a private demo. Those pages also carry a noindex tag, so the two agree. A page in the sitemap with a noindex tag is a contradiction, and Search Console will flag it.

Each page also gets a title and description the right length for Google to show in full. The checker in our static build fails the build if one runs long.

Photos That Say Where They Were Taken

A photo can carry GPS data inside the file. For a local business, that is one more quiet clue about where you work. So we tag site photos with the real location they belong to.

The catch is that most tools strip that data. Resize a photo and the tag is gone. Convert it to WebP and the tag is gone. So our build puts the tag back after every resize and every format change, then checks that it is there. A tag set once and lost on the next step is worth nothing.

What It Saves the Owner

You stop waiting weeks to learn if Google has read your site. You get a dashboard that says which pages are indexed and which are not, and why. New pages get found in days, not months.

You also stop paying for SEO tools that tell you what Search Console already says for free. We read the real data from Google and Bing. That is the base of our getting found work and the city pages we build, covered in city pages that rank.

What We Learned the Hard Way

Set up Search Console before launch. A new property starts empty and fills in over days. If you set it up the week you need data, you have no data.

Deploy the IndexNow key file with the site, every time. If a rebuild drops it, every ping gets refused. Nothing on the site looks wrong. We check the key file is live after each deploy.

Do not trust the ping's success reply as proof of indexing. It only means the engine got the message. Indexed pages show up in the Webmaster reports later, and that is where we look.

Watch for pages you did not mean to block. A leftover noindex tag on a template can hide a whole section. We grep the built output for it before any launch.

For the step by step setup, read Search Console and IndexNow. Next in the stack is analytics and reporting, where we prove the work paid off.