Site relaunch and impact on SEO

[MarkWill]

I have some tough decisions to make about a web site I run. The site has seen around for 20 years (September 1995, to be precise, is the date listed against the domain). Over the years, the effort I've expanded on the site has come and gone, but I am about to throw a lot of time and effort back into it.

The majority of the content on the site is pretty dated, isn't tremendously useful to the audience (since it's pretty old) and the site design and URL architecture isn't particularly SEO-friendly.

In addition, I have a database of thousands vendors (for the specific industry this site serves). I don't know if it's a factor any more but 100% of the links there have been populated by the vendors themselves specifically requesting inclusion (through a form we expose on the site). When the request is approved, the vendor link shows up on the appropriate pages for location (state) and segment of the industry. Though the links are all "opt-in" from vendors (we've never one added or imported any ourselves), I am sure this all looks like a terrible link farm to Google! And some vendors have asked us to remove their link for that reason

One final (very important) point. We have a relationship with a nationwide brand and have four very specific pages related to that brand on our site. Those pages are essential - they are by far the most visited pages and drive virtually all our revenue. The pages were put together with SEO in mind and the look and feel is very different to the rest of the site. The result is, effectively, a site-within-a-site. I need to carefully protect the performance of these pages.

To put some rough numbers on this, the site had 475,000 page views over the last year, with about 320,000 of those being to these four pages (by the way, for the rest of the content "something happened" around May 20th of last year - traffic almost doubled overnight - even though there were no changes to our site).

We have a Facebook presence and have put a little effort into that recently (increasing fans from about 10,000 last August to nearly 24,000 today, with a net gain of about 2,500 per month currently). I don't have any sense of whether that is a meaningful resource in the big picture.

So, that's the background. I want to totally revamp the broader site - much improved design, intentional SEO decisions, far better, current and active content, active social media presence and so on. I am also moving from one CMS to another (the target CMS / Blog platform being WordPress).

Part of me wants to do the following:

• Come up with a better plan for SEO and basically just throw out the old stuff and start again, with the exception of the four vendor pages I mentioned
• Implement redirection of the old URLs to new content (301s)
• Just stop exposing the vendor pages (on the basis that many of the links are old/broken and I'm really not getting any benefit from them)
• Leave the four important pages exactly as they are (URL and content-wise)

I am happy to rebuild the content afresh because I have a new plan around that for which I have some confidence. But I have some important questions.

• If I go with the approach above, is there any  value from the old content / URLs that is worth retaining?
• How sure can I be there is no indirect negative effect on the four important pages? I really need to protect those pages
• Is throwing away the vendor links simply all good - or could there be some hidden negative I need to know about (given many of the links are broken and go to crappy/small web sites, I'm hoping this is just a simple decision to make)

And one more uber-question. I want to take a performance baseline so that I can see where I started as I start making changes and measure performance over time. Beyond the obvious metrics like number of visitors, time per page, page views per visit, etc what metrics would be important to collect from the outset?

I am just at the start of this project and it is very important to me. Given the longevity of the site, I don't know if there is much worth retaining for that reason, even if the content changes radically. At a high level I'm trying to decide what questions I need to answer before I set off on this path.

Any suggestions would be very much appreciated.

Thanks.

[MarkWill]

The vendor links are an important factor for me (and potentially lost in this thread, which is broader in scope). For that reason I opened a separate thread specifically to drill down on that aspect:

http://moz.com/community/q/should-i-remove-all-vendor-links-link-farm-concerns

[MarkWill]

Monica,

Thank you so much for your response, which is greatly appreciated. As you might imagine, I have a few more questions.

Can I assume that when you indicate that the old URLs are worth retaining that use of 301 redirects falls into that? Or, instead, are you suggesting that I should actually retain the old URLs (no redirect)?

The content issue is a little more unclear to me. It is true that I get a small percentage (about 33%) from those other pages. But that still adds up to 150,000 page views. Not a ton of activity, admittedly, but also non-zero

My (tentative) conclusion at this point is to port the existing content as-is, but map to a better URL structure, redirecting the old URLs to the new ones (301). Then, as I start creating new, more valuable content (with completely new URLs) I will have pages that effectively supercede the older content, and I'd redirect the older URLs to the new ones. At that point I have new content and a new URL structure, so it's something of a phased approach but doesn't necessitate throwing away all my current content. Anyway, that approach is on my mind.

The SEO story around the vendor pages is not a good one Here's how it works. Each vendor can submit their details through a form and select a topic and a location (state). Once approved, that vendor would show on a topic page and a separate location page. We don't have pages where we don't have data (so, if Nebraska doesn't have any vendors we don't artificially create an empty page) - but with thousands of vendors we still have thousands of pages. So, numerically, we are probably well over 50% of these "trash" vendor pages. So, we are presumably in the "world of trouble" right now

I have some questions related to these pages:

• Is there a particular metric or metrics that might "prove" that these pages are doing more damage than good? Does Moz expose anything that readily expresses a page like this being assessed poorly by Google?
• Given that I am almost certain that the pages are "toxic" (but would like to know for sure - see above), would I be best served by simply deleting them (this is easy to do) and then presumably redirect them to the home page since there will be no obvious replacement content. I basically want them gone!
• If I do delete them, over time would the "penalty" be repaired, in terms of SEO

I have a lot of questions, don't I Seriously, this is really fascinating to me and I'm eager to understand this before I make any changes. In short, I need a plan.

Thanks again

[MonicaOConnor]

Hi there,

After reading your post i would just like to say, I feel your pain. I have been working the past few months at helping a company out a similar situation. Scrapping a website and starting fresh could be very instrumental in your SEO success, but you have to ask yourself is it the overall best thing for my customers?

The first things I would like to point out (just a little unsolicited advice) is that perhaps moving to a completely new site might be the easiest and most efficient way to do this. You will be able to keep your domain and all of its domain authority while bringing your site into the 21st century. You want to give site structure just as much attention as site content. Is your site mobile friendly? If not, you want to move to a responsive design asap. Now, to answer your specific questions.

• If I go with the approach above, is there any  value from the old content / URLs that is worth retaining? - The old URLs are worth retaining absolutely. There is a chance that all of the URLs drive traffic to the site, and you want to preserve that. As far as content goes, you seem to think only the content on your 4 "traffic driving" pages are actually driving traffic, so I would say scrap it. If your customers find no value in it, then there is no value to it. It is probably taking up good real estate on your site.
• How sure can I be there is no indirect negative effect on the four important pages? I really need to protect those pages - _There will be a little give and take when you redesign a site. You might see some ranking fluctuations, but if those pages are great performers now, chances are they will still be if you change nothing. I always believe when it comes to content,you have to keep what works, but also add freshness to it. That is why User Generated Content is so valuable! It lets you keep static, keyword targeting content while adding something uniquely valuable. While those pages are performing well now, chances are they will be outdated one day like the rest of your site's copy. _
• Is throwing away the vendor links simply all good - or could there be some hidden negative I need to know about (given many of the links are broken and go to crappy/small web sites, I'm hoping this is just a simple decision to make) - **I recommend you have someone do an indepth analysis of your link profile before you make any changes. Depending on how many links there are and whether or not you would have to disavow them, it could be extremely negative for your SEO. If these links make up 5% of your link profile, you shouldn't see a huge issue. If they make up 50% of your link profile you could end up in a world of trouble. The most important thing you can do for your link profile is build new links, and diversify your link profile. **

And one more uber-question. I want to take a performance baseline so that I can see where I started as I start making changes and measure performance over time. Beyond the obvious metrics like number of visitors, time per page, page views per visit, etc what metrics would be important to collect from the outset? - Heat mapping would be a great tool for you to use. The click trails will give you great in page metrics. GA offers this for free, and Moz offers HotSpot with a Pro membership. I always measure the following things:

• Organic CTR
• Search Impressions
• Percentage of increase/decrease in clicks
• Percentage of increase/decrease in New Sessions
• Bounce Rate
• Organic conversion rate
• Decrease in PPC spend
• _Increase/Decrease in Phone calls - I measure this because it is one of our Key Performance Indicators. _

I hope that you will get lots of good answers to your questions! These are my professional opinions based on experiences I have had with older websites. Good luck to you!