Key Takeaways
- Average position on a query with few impressions is noise. Read it next to the impression count before you act on it.
- Impressions from the wrong country are not demand. One of my exports showed 2,096 US impressions and 1 click.
- One search, one page. Several pages aimed at the same search split the result between them.
- Duplicate URLs and unredirected changes make Google guess which page you mean. Redirect or canonicalize them.
- Check indexing page by page. A published page is not a ranked page, and four of mine still show no impressions.
What is an SEO mistake?
An SEO mistake is a decision that stops Google from finding, understanding or trusting your pages, or that makes you read your own results wrongly. The second kind is the more expensive one in my experience, because a wrong reading of the data hides the first kind. Several of the mistakes below are about measurement, not about code.
All the data below comes from the Search Console exports of rahulranjan.com.np. The July, August and September exports each cover about three months. The November export covers 28 days (17 October to 13 November 2026). I name the window every time, because comparing a three-month window with a 28-day window can mislead you, and that is mistake number one in disguise.
The eight mistakes at a glance
| Mistake | Where you see it | The fix |
|---|---|---|
| 1. Trusting average position | Performance, Queries, with an Impressions column | Ignore rows with under about 10 impressions |
| 2. Wrong-country impressions | Performance, filter by country | Judge success on clicks from your market |
| 3. Pages competing | Filter by query, then the Pages tab | One owner page per search |
| 4. Duplicate URLs | Page indexing report | Redirect, canonical, clean sitemap |
| 5. No redirect on changed URLs | 404s and lost impressions | 301 from old URL to new |
| 6. Bought links, mass pages | Provider's proposal | Ask for sources, refuse volume promises |
| 7. Judging too early | Week-two panic | Set three time horizons |
| 8. Assuming a page is indexed | Pages with zero impressions | Inspect each URL |
Mistake 1: Reading average position as a verdict
Search Console reports an average position for each query. Google's help page explains that for each query it counts the topmost position your link held, then averages across queries. That is a useful summary when many impressions sit behind it. It is close to meaningless when one impression does.
Here is one query from my own exports, "digital marketing in nepal":
| Export | Impressions | Average position |
|---|---|---|
| 24 July (about 3 months) | 1 | 2.0 |
| 23 August (about 3 months) | 3 | 27.0 |
| 27 September (about 3 months) | 26 | 42.9 |
| 16 November (28 days) | 57 | 62.3 |
My data: In July that query looked like a win, position 2.0. It was a single impression. Nothing collapsed afterwards. As impressions for that query grew, the average moved toward where my page really ranks for most of them, which is around page seven. The July number was never information.
If I had celebrated in July, I would have ignored a page that actually sits on page seven. If I had panicked in September over a fall from 2.0 to 42.9, I would have hunted for a change that never happened. Both reactions come from the same misreading.
The fix: add the Impressions column to the Queries table, sort by it, and make decisions only on rows with roughly 10 or more impressions. That threshold is my working rule, not Google's. Compare positions across exports only when the windows have the same length.
Mistake 2: Treating impressions from the wrong country as demand
A rising impressions graph feels like progress. It only counts if the people seeing your result can become your customers. If you sell to customers in Nepal, that means checking the country filter.
My data: In the 27 September export, the United States produced 2,096 impressions and 1 click. Nepal produced 2,978 impressions and 30 clicks. In the 16 November export (28 days), the United States produced 677 impressions and 0 clicks, while Nepal produced 1,678 impressions and 16 clicks.
My guess is that the technical guide on AI crawlers and robots.txt, which sits near position 9 for queries with no connection to Nepal, drives most of the US impressions. I have not yet confirmed that by filtering page by country. Whatever the cause, those impressions inflated the chart and produced one click in September and none in November.
The fix: set the country filter to Nepal (or the market you serve) and judge clicks there. Then decide on purpose whether the other audience matters to you. For a Nepal business selling to Nepal customers, it usually does not.
Mistake 3: Aiming several pages at the same search
When two pages of one site compete for the same search, Google may show one, the other, or alternate between them, so impressions and clicks get split. The pages weaken each other. This is easy to do by accident when you keep adding blog posts around one service.
I did it. In August I found five of my own pages aimed at the same "hire an SEO in Nepal" search. I merged one into the main guide with a permanent redirect, moved the pricing page onto its own search, and retargeted the freelancer guide to the "SEO freelancer" wording.
My data: It is not solved. Two queries with enough impressions in both exports moved the wrong way between the July and August exports: "freelance seo expert in nepal" went from position 10.6 to 27.8, and "technical seo expert nepal" from 9.8 to 17.5. The main guide's head term, "seo expert in nepal", averaged 12.2 in September and 19.7 in November. I do not yet know how much overlap explains that, so I am still testing it.
Overlap is only one possible cause of that drop, so I am not claiming the cleanup worked or failed. The lesson is the process: find the overlap, pick an owner, measure again. For the main guide itself, see hiring an SEO expert in Nepal.
The fix: in Search Console, filter Performance by one query, then open the Pages tab. If two URLs share the impressions, they compete. Keep a simple keyword map with one row per search and one owner URL. Rewrite or redirect the others.
Mistake 4: Leaving duplicate URLs live
Duplicate URLs are the same page reachable at more than one address. Two caused me trouble. First, I added a ?src= tracking parameter to internal links to see which page sent visitors, and Google was crawling the parameter versions as separate URLs. Second, /index.html returned the home page alongside /, and the Page indexing report listed it as "Alternate page with proper canonical tag".
Google mostly handled both, so nothing broke. But it was spending crawling on addresses I should never have created. Google's duplicate URL guidance ranks the ways to fix this: redirects and rel=canonical are strong signals, and listing a URL in the sitemap is only a weak one. It also says to keep your sitemap and canonicals consistent.
One honest caveat. Google's crawl budget guide is aimed at big sites, roughly 1 million or more unique pages, or 10,000 or more with daily changes. It also lists a large share of URLs stuck at "Discovered - currently not indexed" as one sign a site belongs in that group. My site has about 30 pages, so I treat crawl waste as my best explanation, not a proven one.
My data: Six sitemap pages showed zero impressions in the 27 September export. In the 16 November export, two of them appear: the freelancer guide with 143 impressions and the technical SEO checklist with 8. The GBP, web development, React and SaaS service pages still show none in the 28-day window.
The fix: 301 /index.html to /, remove tracking parameters from internal links (read the source from the referrer instead), give every page a self-referencing canonical, and list only canonical URLs in the sitemap. The full list of technical checks is in the 25-point technical SEO checklist for Nepal sites.
Mistake 5: Changing a URL without a redirect
New designer, new CMS, a renamed file: the address changes and the old one starts returning a 404. Every ranking signal and inbound link the old URL had loses its destination. Google recommends a permanent server-side redirect whenever a URL changes in search results, and names 301 and 308 as the permanent types. Temporary types such as 302 keep the old URL in the results.
My data: An old CV PDF on my site was indexed and had 23 impressions at an average position of 8.1 in the September export. When I renamed the file, I added a 301 from the old address to the new one instead of letting it 404. I did the same for old image paths after renaming images for image search.
The fix: before any redesign or migration, list every old URL and its new address, and 301 each one. On Cloudflare Pages that is a _redirects file. On WordPress it is a redirect plugin or the server config. Then inspect a few old URLs to confirm they land on the right page.
Mistake 6: Buying links or mass-producing pages
Some SEO offers sell volume: hundreds of links, or dozens of articles a month. Google's spam policies describe link spam as links made mainly to manipulate rankings. The list includes buying or selling links for ranking, excessive link exchanges, and automated link creation. Paid links are allowed when marked with rel="nofollow" or rel="sponsored". The same policies cover scaled content abuse, which means many pages produced mainly to manipulate rankings, however they are made.
I cannot tell you which Nepal providers do this, and I will not guess. What I can give you is the test. A proposal that promises large link counts or large article counts in the first month sits close to what the policies list, which is reason enough to ask three questions: which sites will the links come from, can I see them before I pay, and who writes each page and what does it add that is not already online.
The fix: choose a provider who commits to specific pages and specific fixes, and who shows you the work. The guide to hiring an SEO freelancer in Nepal has a vetting checklist, and the main guide lists the red flags.
Mistake 7: Judging SEO results too early
Google's SEO Starter Guide says some changes take effect in a few hours and others take several months, and it suggests waiting a few weeks before judging. Judging in week two gets it wrong.
My data: Specific local searches moved first. In the 28 days to 13 November, "seo expert in biratnagar" averaged position 1.0 on this site. The national search for the same topic, "seo expert in nepal", averaged 19.7. A technical guide I published on 23 August had 2,960 impressions at position 8.7 by the 27 September export, five weeks later, but only 6 clicks.
So the timeline depends on the search. Narrow local searches move first, broad national ones take longest, and impressions arrive well before clicks. These are my working horizons, not Google's rules: check indexing and titles at two to four weeks, expect movement on narrow local searches within two to three months, and judge broad national searches over six months or more. My SEO expert in Biratnagar page is the local example.
The fix: write down the date of every change and review at three points, not weekly. If a page is not indexed, that is a different problem, and it is the next mistake.
Mistake 8: Assuming a published page is indexed
Publishing a page does not put it in Google. Google's Page indexing report notes that a new site or page can take "a week or so" to be crawled and indexed, and that not indexed is not necessarily bad. But a page that is not indexed cannot rank, and you will not see it unless you look.
My data: Four service pages on my site (Google Business Profile management, web development, React development and SaaS development) show no impressions in the 28-day window, and they showed none in the September export either. A 28-day window is shorter than the three-month one, so I am inspecting each URL in Search Console instead of assuming they are missing.
The fix: open the Page indexing report, read the reason listed for each page that is not indexed, then use URL Inspection on any page with zero impressions. Check that the page is in your sitemap and linked from other pages. Fix the reason Google lists, then request indexing.
A 10-minute Search Console review
You can find most of these mistakes in one sitting. Open Search Console and do this in order:
- Performance, Queries. Set three months, add Impressions, sort by it. Ignore rows under about 10 impressions.
- Add a country filter. Compare clicks from your market with the total.
- Pick your three most important searches. Filter by each and open the Pages tab. Note any search where two URLs share impressions.
- Page indexing report. Read the reasons under Not indexed, and look for parameter or index.html duplicates.
- URL Inspection. Check every important page that has zero impressions.
- Write the keyword map. One row per search, one owner URL.
When to get help
You can do all of the above yourself, and the free tools are enough. Help is worth it when the problem is in the code, such as redirects, canonicals and speed, or when you do not have the time. I audit the site, fix what is wrong in the code, and report from your own Search Console account. See the SEO case studies for results with their date ranges, or request a free website audit. To work with me, contact me for a quote.
Want these checked on your site?
Send me your website address
I will review it for indexing, duplicate URLs, competing pages and speed, then reply with what I would fix first and a written quote.
Frequently Asked Questions
What are the most common SEO mistakes in Nepal?
The ones that cost this site the most time were reading average position on tiny samples, counting impressions from other countries as demand, aiming several pages at one search, leaving duplicate URLs live, changing URLs without redirects, and never checking that pages were indexed. Buying links is another one to avoid. Each is shown above with data.
Why is my Nepal website not ranking on Google?
Check three things in order. First, whether Google has indexed the page by inspecting the URL in Search Console. Second, whether another page of yours targets the same search. Third, whether you are reading a reliable number, because positions on queries with few impressions are noise. Google says a new page can take about a week to be crawled and indexed.
How long does SEO take to work?
Google's SEO Starter Guide says some changes take effect in hours and others take several months, and suggests waiting a few weeks before judging. In my own data, one specific local search reached an average position of 1.0 while the broader national search for the same topic is still around position 20.
Do two pages on the same topic hurt my SEO?
They can. When two URLs target the same search, impressions and clicks split between them and neither ranks as well as one stronger page would. Check it in Search Console: filter by the query, open the Pages tab, and see whether two URLs share the impressions. Then pick one owner and rewrite or redirect the other.
Are paid backlinks safe?
Google's spam policies list buying or selling links to influence ranking as link spam. Paid links are allowed when they carry a rel=nofollow or rel=sponsored attribute. If a provider promises large numbers of links, ask which sites they come from and whether you can see them before you pay.
Should I add tracking parameters to internal links?
Avoid it. A parameter such as ?src= makes a second URL for the same page, which gives Google duplicates to sort out. Read the visitor's source from the referrer in your analytics instead, and keep parameters for links you place on other sites.
Can an SEO expert fix these problems for me?
Yes, most of them are quick to find with Search Console access. I audit the site, fix what is wrong in the code, and report from your own Search Console account. Send me your website address through the contact form for a free review and a written quote.
Sources
- Search Console Performance report: position, clicks and impressions (Google Search Console Help)
- Page indexing report (Google Search Console Help)
- How to specify a canonical URL and consolidate duplicate URLs (Google Search Central)
- Redirects and Google Search (Google Search Central)
- Spam policies for Google web search (Google Search Central)
- Crawl budget management for large sites (Google Search Central)
- SEO Starter Guide (Google Search Central)
Google documentation was read on 16 November 2026. Google updates these pages, so check the wording if you cite them later. Search Console figures come from the exports of rahulranjan.com.np dated 24 July, 23 August and 27 September 2026 (about three months each) and 16 November 2026 (28 days).