Welcome to the Q&A Forum

Yes, the problem is that it is not flat HTML. I think that the site was originally a flat HTML site, with inline CSS (and JavaScript on page) that was added into a CMS (which isn't the best CMS for SEO, in my opinion). Personally, I think flat HTML (coded in modern markup) sites are better to work with on the small scale due to the amount of flexibility they offer. Actual HTML sites are not my real issue, my issue is the unclean code. It is way beyond current standards, there are bits of Java in there that don't do anything and the code is very long for the size of the page and the amount of content on there. My question basically is, would two sites with identical content (ignoring the issue of duplication for the purposes of this hypothetical), rank differently if one had new, clean and error free code and one was the original with the errors and the out of date code?

Sorry Doug, I should have said that there were two drops, one for Panda, and then one for Penguin, with one coming in Mid-Feb and one Mid-April. Panda was a large drop, then as the site was recovering slightly it got hit with the Penguin update. The competitors are mainly Amazon and eBay. Both sell the products, but only sporadically and don't have them in any number. The client has dropped on the 4 big keywords from 1st-2nd position, to 5th-7th. Not a massive drop by any means, but in a market so small it has had a huge effect. The rest of the site’s rankings did drop, but they have recovered since then. The backlinks are not overly impressive, but nothing too alarming. There have been no warnings in Webmaster Tools, nevertheless, I have been working on the link profile and trying to add variety. After originally thinking this was the cause I have begun to reconsider, and in my digging found several faults on-page. I fixed most of these when I first got to work on the site, such as the internal linking and general optimization. There was also a spammy element to the internal linking which I got rid of. Fixing the internal linking didn’t have the desired effect, so I have re-approached that and changed it further. I also found an issue with the CMS in which it was generating a series of duplicated page titles (that mirrored the index page due to an error in the CMS to the lower pages). This would fit with the Panda change and the aim to reduce duplicates. This month, I have eliminated these, cleaned up the internal linking further and looked to vary the link profile (I'm hoping this will fix the issue, I just need to wait until it gets indexed). Also, as an experiment, I made a series of HTML pages (not in the CMS) to test the site. These ranked well within a month, and have continued to grow since then. Another issue the site has is the site itself. The code is outdated and messy with inline CSS and Java (some of which seems redundant) that all make the code to content ration something to be desired. I’m wondering if this could be a cause. The CMS has a penchant for duplication and the ‘readable’ code is not very clean, dates, filled with redundant code and is old. Thanks