I have seen it happen so often – a guy who knows “design” gets a new gig, spends the next three months coming up with a new look for a site, spends three months building it, launches it with huge fanfare and gets canned because the site loses 75% of its search traffic on day one.
If you are considering a website redesign then I strongly recommend you get advice from someone who understands SEO. Here are some questions to ask:
- What pages are we deleting?
- What keywords do the deleted pages get search engine referrals for and how many does each get?
- What pages do the deleted pages link to (what internal links are we losing)?
- What URLs are we redirecting the deleted page URLs to?
- Are these redirect pages targeting the same keywords that the deleted pages are targeting and how well do they do this v. the old pages? (this is a hard one to answer but it should definitely be considered)
- Do we have to delete all of these pages? Are there any we can keep?
- How is the site architecture changing?
- What pages will have their prominence within the site hierarchy change? (e.g. did the page go from being a main navigation tab to a sub navigation tab?)
- How are the internal links to each page changing and what is your plan to compensate for that change? In particular what links are we losing/gaining on the home page and the main navigation?
- The text of which pages (including page titles & meta descriptions) are going to change? What is your estimate of how this will change the search engine referrals for each page.
- Which graphics are changing? Do we currently get search engine referrals for any of the graphics that are changing and if so what is the plan to compensate for this?
There are a number of other issues, both technical and creative, that should be considered as part of website redesign, but chances are that if you have answered the above questions, you probably have a good grip on SEO and should be able to figure it out.
With any radical change to site architecture it is common to lose some traffic as the engines recrawl and figure out the new pages. But if you do your job right the traffic loss can be kept to a minimum and you might even start out with a traffic increase.
I also recommend that you compartmentalize the changes as much as possible. For example launch your new architecture and let that settle first before you start rewriting pages. This way you’ll be better able to isolate the cause of any problems.