Your Page Exists. Google Just Won't Add It to the Index.
You publish a new service page. You check Search Console a week later and see the status that makes every small business owner's stomach drop: "Discovered - currently not indexed." Not an error. Not a penalty. Just... nothing. Google found the URL and decided, for now, not to bother crawling it — or crawled it and decided not to index it. Either way, the page can't rank for anything until that changes.
Most advice you'll find on this topic tells small business owners to worry about "crawl budget" — the same language Google uses in its official documentation, which it just updated on August 3, 2026 with new guidance on HTTP caching and how crawlers share server capacity (Search Engine Journal, Aug 2026). Here's the problem: that documentation is explicitly written for sites with a million-plus pages or sites publishing rapidly changing content at scale. Google says so directly. If you run a 40-page service business site, crawl budget in the enterprise sense almost certainly isn't your bottleneck.
So why is your page still stuck? The real answer lives one layer deeper, in how Google's indexing pipeline actually works — and it's the layer most small business SEO advice skips entirely.
The scale check: Google's own crawl budget documentation names its target audience explicitly: sites with 1 million+ pages updating weekly, sites with 10,000+ pages updating daily, or any site where a large share of URLs sit in "Discovered - currently not indexed." If none of those describe you, the August 2026 update isn't really about you — but the underlying indexing pipeline it describes still is. (Google for Developers, crawl budget guide)
How a Page Actually Gets From "Published" to "Ranking"
Every URL on the web moves through the same pipeline, and understanding each stage tells you exactly where your page can get stuck. A recent breakdown of Google's process lays it out as seven stages: discovery, crawl scheduling, fetching, rendering, indexing decisions, the index itself, and serving (IndexBolt, 2026).
- Discovery — Google finds your URL through links, your XML sitemap, or a direct submission in Search Console. No inbound links and no sitemap entry means Google may never even learn the page exists.
- Crawl scheduling — the URL waits in a queue. How long it waits depends on crawl demand (how much Google wants it, driven by links and site track record) versus crawl capacity (how much your server can sustain).
- Fetching — Googlebot downloads the page. As of Google's own August 2026 technical explainer, Googlebot fetches up to 2MB per URL (excluding PDFs, capped at 64MB), including HTTP headers. Anything past that cutoff is never fetched, rendered, or indexed (Search Engine Land, citing Gary Illyes, Mar 2026).
- Rendering — Google's Web Rendering Service executes your JavaScript on an evergreen version of Chromium to see the final page the way a browser would.
- Indexing decisions — Google clusters duplicate or near-duplicate pages, picks one canonical URL to represent them, and applies quality thresholds. Pages that don't clear the bar get discovered but never indexed.
- The index and serving — being indexed only means a page is eligible to rank. It says nothing about where.
For a small business site, the stall almost never happens at "fetching" because of raw crawl volume. It happens at rendering (Google can't see your content the way a visitor does) or at the indexing-decision stage (Google saw the page fine and decided it wasn't worth adding).
The August 2026 Crawl Budget Update: What Actually Changed
Google's update to its "Optimize Your Crawl Budget" documentation added two specific pieces of guidance worth knowing even if you're not running an enterprise site, because they reveal how Google's crawlers behave in general (Search Engine Roundtable, Jul 2026; SEJ, Aug 2026):
| What Changed | What It Means |
|---|---|
| "Every site starts with the same default, conservative crawl capacity limit" | New domains don't get extra crawl attention just for existing — capacity grows only as Google trusts your server's health and sees demand |
| Crawl capacity is now explicitly shared across all of Google's crawlers on your host | Heavy image/video crawling can quietly reduce the capacity left over for your main Googlebot crawls |
| New recommendation to serve HTTP 304 (Not Modified) responses | If a page hasn't changed since Google's last visit, telling it so directly saves your server's crawl "budget" for pages that did change |
None of this changes the fundamentals for a small site. But the 304 recommendation is a genuinely free technical win most CMS platforms and static hosts already support with minimal configuration — and it signals Google is paying closer attention to server-response efficiency across the board, not just for massive sites.
The Real Reason Small Business Pages Stall: It's Rendering or Quality, Not Volume
If your site has 50 or 500 pages, "Discovered - currently not indexed" almost always traces back to one of two causes that have nothing to do with crawl budget as Google defines it for enterprise sites.
Cause 1: JavaScript Rendering Delay
Google doesn't index your page in one pass. It crawls the raw HTML first, queues the page for rendering, then indexes what the rendered DOM actually shows. A large-scale study of over 100,000 real Googlebot fetches — primarily on nextjs.org — found that rendering happens with a median delay of 10 seconds after crawling, with the 25th percentile completing in under 4 seconds. But the tail is the real risk: the 99th percentile took around 18 hours, and the study found no correlation between how complex the JavaScript was and how long the delay ran (Vercel/MERJ study via SEOBRO, 2026).
The dangerous failure mode isn't content that JavaScript adds — Google typically renders and indexes that fine, just later. The dangerous failure mode is content JavaScript changes: a raw HTML canonical tag that points to your homepage, then gets rewritten per-page by client-side code after the page loads. A server-rendered noindex directive that JavaScript removes after hydration. A title tag that reads "Loading…" until a script finishes running. Each of these sends Google two conflicting signals about the same URL, and you don't control which one wins.
Diagnostic tell: If pages get indexed but rank for almost nothing — as if Google is seeing a thinner version of the page than a visitor does — or SERP snippets show wrong or homepage-default descriptions for inner pages, that's the signature of a raw-HTML-vs-rendered-DOM mismatch, not a crawl budget problem. Compare what your server sends versus what the rendered DOM shows using Search Console's URL Inspection tool. (SEOBRO, 2026)
Cause 2: The Page Doesn't Clear Google's Quality Threshold
This is the less comfortable explanation, and it's the one most business owners resist. Indexing is selective by design. Google evaluates content, clusters duplicates, and — critically — declines to index pages that don't clear a quality bar, independent of whether they were crawled successfully (IndexBolt, 2026). A thin location page that's 90% boilerplate copied from your other 12 location pages, an auto-generated tag or filter page with no unique value, or a blog post that says the same thing as three other posts on your own site — these get crawled fine and then never make the index, because Google looked at them and decided they added nothing new to what it already has.
This connects directly to a mistake we've flagged before on this blog: thin, templated location pages built purely to rank for "[service] in [city]" are exactly the pattern that triggers this outcome at scale — dozens of near-duplicate pages competing with each other and with none of them clearing the bar.
Diagnosing Your Own Site: A Practical Checklist
| Check | Where to Look | What It Tells You |
|---|---|---|
| Page Indexing report | Search Console → Indexing → Pages | Which specific URLs are "Discovered," "Crawled - not indexed," or "Indexed" — the exact status, not a guess |
| URL Inspection → "View Crawled Page" | Search Console, per-URL | Shows the actual rendered HTML Google saw — compare it to what you see in a browser |
| View source vs. rendered DOM | Browser dev tools, "View Page Source" vs. Elements panel | Reveals canonical, robots, or title mismatches that only appear after JavaScript runs |
| Server response headers | Dev tools Network tab, or curl -I | Confirms whether your server supports conditional requests (ETag / Last-Modified) for 304 responses |
| Duplicate content scan | Site crawler (Screaming Frog free tier covers up to 500 URLs) | Flags near-identical pages competing for the same index slot — common on location and category pages |
| Internal links to the stuck page | Search Console → Links → Internal links, or a site crawl | Pages with zero or near-zero internal links signal low demand to Google, slowing the queue |
The bottom line: For a small business site, "Discovered - currently not indexed" is rarely a crawl budget problem in the enterprise sense Google's documentation targets. It's almost always a rendering mismatch between what your server sends and what JavaScript changes afterward, or a quality/uniqueness problem where the page genuinely doesn't offer anything Google doesn't already have indexed elsewhere on your site or the web. Fix the diagnosis before you fix the wrong thing.
Five Fixes That Actually Move Pages Into the Index
- Serve critical SEO signals in raw server HTML, not just post-render. Canonical tags, title tags, meta descriptions, and robots directives should be correct in the initial server response — never dependent on client-side JavaScript to be correct.
- Support HTTP 304 responses. Most modern static hosts and CMS platforms support ETag or Last-Modified headers with minimal configuration. This is literally the free win Google just called out in its August 2026 update.
- Consolidate or de-duplicate templated pages. If you have location or service pages that share 80%+ of their copy, either differentiate them with genuinely unique local content (specific neighborhoods, local case studies, area-specific pricing context) or consolidate them into fewer, stronger pages.
- Strengthen internal linking to orphaned pages. A page with zero internal links signals low demand. Link to new or stuck pages from your homepage, blog posts, or navigation — not just your XML sitemap.
- Resubmit through Search Console once fixes are live. Use the URL Inspection tool's "Request Indexing" after making changes, then check back in a few days rather than assuming the fix worked instantly — Google's own guidance notes CrUX-style field data windows and reprocessing take time to reflect changes.
Once you've ruled out the rendering and quality issues above, it's also worth confirming your site isn't losing crawl efficiency to basic speed problems — slow Time to First Byte and inconsistent response times reduce the crawl capacity limit Google is willing to extend you, which is a separate but related lever covered in our Core Web Vitals breakdown.
Frequently Asked Questions
What does "Discovered - currently not indexed" actually mean?
Google found the URL — through a link, sitemap, or submission — but hasn't crawled and indexed it yet. It's not an error or penalty status; it simply means the page is in the queue or was evaluated and didn't clear Google's quality or uniqueness threshold. Check the Page Indexing report in Search Console for the precise status on each URL rather than assuming.
Does my small business site have a "crawl budget" problem?
Almost certainly not in the sense Google's official documentation addresses — that guidance is explicitly aimed at sites with 1 million+ pages or 10,000+ pages updating daily. If you're running a normal small business site with dozens or a few hundred pages, your indexing issues are far more likely to be rendering mismatches or content quality/duplication than raw crawl capacity.
How long should I wait before assuming a page won't get indexed?
Give it at least a few weeks after publishing and requesting indexing through Search Console before troubleshooting deeply — new domains and pages with few internal links naturally take longer because crawl demand builds gradually. If a page is still stuck a month later with strong internal links pointing to it, treat that as a real signal something needs fixing rather than more patience.
Can JavaScript-heavy pages still rank well in 2026?
Yes, but they carry more risk of a rendering delay or a signal mismatch between server HTML and the rendered DOM. Google's rendering queue adds a median 10-second delay after crawling, with a long tail that can stretch to 18 hours — and the real risk isn't delay itself, it's canonical tags, robots directives, or titles that differ between what your server sends and what your JavaScript changes after the page loads.
Will serving 304 responses actually help my rankings?
Indirectly. A 304 response doesn't improve your content or rankings by itself, but it reduces the server resources Google spends re-fetching unchanged pages, which Google's own August 2026 guidance frames as a way to preserve crawl capacity for pages that did change — including new content you actually want indexed quickly.
What's the single highest-priority fix if I only have time for one thing?
Check whether your canonical tags, title tags, and robots directives match between raw server HTML and the rendered DOM using Search Console's URL Inspection "View Crawled Page" feature. A mismatch here creates ambiguous signals that actively work against indexing — and it's a higher-priority fix than adding more content, because content additions get indexed eventually while signal conflicts can block indexing indefinitely.