Web Issues Costing you Traffic and Sales – a Checklist
The first thing I do when working with a new company on their Web Marketing is to review their site to see what we can quickly fix. In this post, I discuss many of the issues that can cost you Web traffic and sales:
Unreliable Site Hosting
Problem If your site is down, customers and Google can’t access it.
To see if your site is unreliable – Try a service like: Siteuptime or Basicstate (these are not free). Services can check your site every 15 minutes or so, and give a complete picture of your site’s up time. If you don’t want to use the service, you can manually visit your site each day–not as effective as a service, but better than nothing.
To address the site failures – you may want to change hosting providers, or at least upgrade to a higher level of service on your existing host.
Slow Site Performance
There is some talk of Google using “page load” performance as one of the many measures of quality and trust. Even if this were not the case, the attention span of the average user grows ever shorter. If your site loads too slowly (a somewhat subjective concept), you can use the Firefox plug in “yslow” to help detect why the page loads slowly.
To fix the issue – You can upgrade your host provider or service level. In addition content distribution networks are useful in many cases.
Invisible or Missing Site Content
Problem If Google has no content to index, you won’t rank. Often caused by using FLASH or Graphics instead of text.
To determine if you have a content issue – Look at how Google “sees” the text on your page to make sure there is enough text for Google to understand what the page is about. You need to write for user experience, and to be sure there is text for Google to “see”.
To check this:
1) Type the command: cache:http://www.name-of-your-page.com into the Google search box. This returns the screen below:
2) Click on the hyperlink labeled “Text-only version”
3) Google will display the web page with the text as Google “sees” it. If you don’t see any text, or just a few words, you need to check and see why.
To Address the Issue – If you use FLASH or graphics instead of text on your site, you may find there is no text shown at all. You need to change the page (by adding text) so the keywords that you are optimizing for are visible to Google.
Link Naming Problems (Canonical Naming Issues)
Problem – When you link to a page in your site, you need to be consistent with the link structure or you will reduce your ability to rank well for important terms. It turns out that: “mysite.com”, “www.mysite.com” and “www.mysite.com/index.html” are all different addresses. Most standard hosting setups will resolve the addresses and show the right content, but when you link to a page, you need to be consistent and only use one address. This can be an issue if you are not careful with you internal site links and will often be an issue when other sites link to yours
How to see if you have a canonical naming issue – To see if you have this issue try this:
In Yahoo Site Explorer, (http://search.yahoo.com) in the search area:
Type: “links:www.your-site-name.com” ….. see how many links are returned
Type “links:your-site-name.com” ….. see how many links are returned
Type “links:www.your-site-name.com/index.html” ….. see how many links are returned
Type “links:www.your-site-name.com/index.php” ….. see how many links are returned
If the number of links is different, you have the issue. (Note, not all addresses may be valid for your site.)
To address the issue -There are a couple of ways to address this issue:
1) Use an .htaccess file on your Linux server (or httpd.ini for MS server), or PHP code on the page itself to redirect all permutations of a page to the one true (canonical) name. With this change any address will be redirected to the one true address and all links will by extension point to the same (correct) page. You must be sure to use an 301 (permanent) redirect to adhere to Google’s guidelines.
2)The second method is Google specific. It requires that you change the preferred URL in Google webmaster tools for your site to you preferred address. This method will ensure that Google give you “credit” for all links, but will not redirect bad addresses (that don’t resolve ) to the proper page.
Here is a post describing the Google method from their site: naming issue
–There are many articles on redirects, try searching for: “301 redirects” in Google.
Site Virus or Malicious Code
Problem – This is one of the worst issues, you may have a problem and not even know it. This often happens when you leave the access to files on your server open to the public. It can also happen because of malicious hacks in plug-ins for platforms like WordPress.
How to detect it – Turn on safe browsing in Firefox (options->security) or Internet explorer (options->security) these settings should flag malicious pages on your site.
Use Google webmaster tools for your site. Google will tell you if any pages are suspect and allow you to fix the problem and get the page reindexed. Google’s search results will also show a warning screen when you go to any page that has malicious code on it. If you see this screen on a page from your site, Google has flagged it bad. This can affect the traffic and rankings for your entire site.
How to address the issue – Remove the bad source code from the page and all files on your site. It is beyond the scope of this post to cover the details. Please refer to the technical documents for your platform for help.
Incorrect Site Changes
Problem Creating “orphan” pages that are no longer linked to your site. If you make changes that leave old pages out of your new site structure, you lose built up ranking and trust. You also create a bad experience for users who find the old pages in a search. Here are the typical things that cause this issue:
- Creating a new version of an existing site with new pages names
- Creating a new site on the same URL and deleting the old one
- Moving a site to a new domain
- Changes in technology or implementation (pages ending in .php vs .html)
- A company is purchased and their site is folded into the site of the buyer.
The core problem is that your existing site has pages in Google’s index that rank for terms that you care about. Your pages also have links from external sites and your own site (internal links). When you make any of the changes above, you are creating new pages and often new content. To compound matters, if the implementation of the site changes (say from a static HTML to a dynamic PHP based site) all the new page names will be different.
If you just blow away the existing pages on your site, you lose all those indexed pages (and credit for the links to them). Worst of all, when a user finds the old pages in the search results, they get an error when they try to visit.
How to tell if you have a problem – Go to the Google search window, type: “site:www.your-site.com”
Google will return all pages from your site in it’s index. Each of these pages needs to be redirected to the new pages in your site so the user sees the expected result. You can also use Google’s Webmaster tools to determine any visits to your sites that result in errors.
How to fix the problem – Use 301 redirects to point the old pages to the new pages on your site. For larger sites, you may want to move the pages over in groups and see how long it takes Google to index the new pages. Over time you can migrate all your pages.
Useless Pages Indexed in Google
Problem indexed source code, script files, webshop pages, and other pages of this ilk may be pushing valuable pages out. I see this often with technology companies. Their web site has example scripts and code that may be of use to users, but not much use for ranking for keywords. Worse yet, many of the script files contain common text (headers, common parameters, etc.), which also causes Google to flag the pages as duplicate content.
Google decides how important your site is using proprietary ranking factors. Based on this decision, there are a finite number of pages that Google will index for your site. The only way to get Google to index more of your site’s pages is to keep adding quality links. The problem with indexing useless pages is that these pages take some of the slots allotted by Google that could be used to index more desirable content.
How to tell you have the problem – From the Google search prompt, type the command:
Google will return all the pages that they have indexed for your site. Us the page navigation to move through each of the pages looking for files and directories that you do not expect to be indexed. These pages and directories will become your target list for removal.
How to fix the Issue – There are several ways to get content out of Google’s index:
- Use your robots.txt file to keep Google from indexing files and directories
- Use the robots meta tag (noindex, nofollow) on any page that you do not want in the index
- Use java script to block access
- Use .htaccess to password protect pages
- Tell Google’s Webmaster’s tools to drop the page
Content is Blocked
Problem If Google can’t see your content, it can’t rank it. This happens when tags or robots.txt are used to stop Google from accessing pages that should be indexed.
How to Tell if You have the Problem – As we did in the example above, use the site command or use Webmaster tools to verify that all pages that should be in the index are.
What to do about the issue – To get the pages back in the index, fix the technical problem. If you don’t see the page back in the index in a few weeks, try adding some internal and external links to the page.
Title and Tag issues
Problem Simple, yet often overlooked, title and tag issues are crucial to ranking in Google. The title and description tags control what the user sees when your pages come up in the search results. At minimum you need to:
- Have a unique title for every page of your site with the key words that you want to rank for in it.
- Have a unique description for every page that tells the user a bit of detail about the page and spurs them to click on the result.
Ugly Page Names with session IDs
Problem Auto-generated page names look like ugly to users. These are those funny page names that have questions marks and numbers (due to session IDs) in them.Â Ugly page names are bad for your users and can cause duplicate content issues.
What to do about it – Use “mod_rewrite” ( Apache Server parlance) or any other scheme supported by your CMS or platform to change the title to “human readable” form.
If you have questions, please leave a comment below.