Welcome to the Q&A Forum

Hi Paul,

Will do this in reverse order to make life a bit easier - and I suspect there will be other posts suggesting other ways to do this too - google has made several changes about how to do this - none of which are very clear!

Point 3. To link your google+ profile (personal profile - not company page!) go to your profile page, hit edit, scroll down to "contributor to" and add your homepage url(s) there and save. If you write for several different sites, you can add those too. It's basically googles method of verifying the author against the publisher and vice versa.

Point 2. This as far as I am aware is only needed if you have an author page with multiple authors on it - but yes, that would be the implementation method.

Point 1. Not quite sure I understand what you mean there - so here is a few things about that.

You can add your author name in each post, and rel=author that back directly to your g+ profile

If you have multiple authors and want to link it to your author page on site - you could do that, but I am not sure that google will still pick each article up and accredit you as the author - my suggestion (unless anyone else can clarify the correct way) would be to do as above.

There is another point if you have a g+ company page - Go to https://developers.google.com/+/plugins/badge/#connect to add a rel=publisher tag to the head of your homepage only and that will qualify your for direct connect.

Hope that helps!

Ok, first, does the url with /errors/error_404 (without the query string) exist?

Also - are you using something to create your 404 pages for tracking purposes?

The query string on the url would appear to be stating the referring url anyway - hence ?q=-i3 basically equates to the page /i3

At a guess - it could be there is a script running to create logs on specific 404's, or create a new log each time one occurs - if this folder is visable to the crawlers, it would get spidered and subsequently the problem would arise.

I would agree with the points EGOL has made - there are a few other things to take into account too.

Is it older static content they are wanting to use? or, is it an article repository that you constantly add more articles to?

If it is static content - then as per EGOL and consider licensing it to them.

If it is on site article content then you could consider creating an rss output of it, and have them rework it to include a cross domain canon as you suggest yourself, if the content you write has links embedded into it that link to other relevant pages on your site, that will help too as far as "riding their traffic" is concerned.

If you do go down this route - I would suggest a delay of 6 hours or so on updating your rss feed to them - then at least you stand a good chance of google seeing your content on your site before they outrank you for it. (It will still happen occasionally though, but correct the following day)

In addition to Gianluca's response there, the pages that you tag with "noindex,follow" (i.e. the duplicates) add a canonical tag pointing at the original page.

It does, when you get the report, filter column D marked "4XX (Client Error)" and you will see your 404's there - further along in the report you will also be able to see referring url which will show you which page is linking to it.

Custom crawl right here on SEOmoz, or you could use Xenu

Thanks for your input EGOL, I was hoping you would spot this and chip in having seen your comments on Cre8asite.

Do you use g+ profiles in this fashion on any sites your run/manage? (or anyone else reading this for that matter?)

For those that have got it implemented - have you seen a rise in traffic from plus.google as a result? and/or a rise in ranking?

It might of course be a little early to tell yet - and googles promise of company profiles on g+ might have a workaround built in.

A bit of a mammoth question for discussion here:

With the launch of Google+ and profiles, coupled with the ability to link/verify authorship using rel=me to google+ profile - A few questions with respect to the long term use and impact.

As an individual - I can have a Google+ Profile, and add links to author pages where I am featured. If rel=me is used back to my G+ profile - google can recognise me as the writer - no problem with that.

However - if I write for a variety of different sites, and produce a variety of different content - site owners could arguably become reluctant to link back or accredit me with the rel=me tag on the account I might be writing for a competitor for example, or other content in a totally different vertical that is irrelevant.

Additionally - if i write for a company as an employee, and the rel=me tag is linked to my G+ profile - my profile (I would assume) is gaining strength from the fact that my work is cited through the link (even if no link juice is passed - my profile link is going to appear in the search results on a query that matches something I have written, and hence possibly drain some "company traffic" to my profile). If I were to then leave the employment of that company - and begin writing for a direct competitor - is my profile still benefiting from the old company content I have written?

Given that google is not allowing pseudonyms or ghost writer profiles - where do we stand with respect to outsourced content? For example: The company has news written for them by a news supplier - (each writer has a name obviously) - but they don't have or don't want to create a G+ profile for me to link to. Is it a case of wait for google to come up with the company profiles? or, use a ghost name and run the gauntlet on G+?

Lastly, and I suppose the bottom line - as a website owner/company director/SEO;

Is adding rel=me links to all your writers profiles (given that some might only write 1 or 2 articles, and staff will inevitably come and go) an overall positive for SEO? or, a SERP nightmare if a writer moves on to another company? In essence are site owners just improving the writers profile rather than gaining very much?

I would agree with Stephen on point 1, but would throw the following thoughts in on point 2:

For example:

page A is mysite.com/seo-india

page B is mysite.com/seo-services

I would guess that there would be a minor difference in ranking impact on a couple of counts with respect to what phrase is ranked for in each case - despite having the same content.

Things to consider would be the url itself, the page title, the H1 and the metadescription.

I can't see that using a canon from one page to another will migrate additional ranking terms to the destination page in its own right - if you have links to page B with anchors that reflect the terms it might - but how long you would continue to rank for the additional term would be questionable.

If it was that simple to get the same page to rank for multiple terms just by duplicating content, linking to it, and then redirecting with the canon tag - I am sure google would get wise to that pretty quick, and there would be no end of "top secret ranking" mails doing the rounds from "SEO gurus"

Would be interesting to test though!

How unique is your content? I took a couple of posts at random and searched the text in quotes and a lot of other sites with the exact same content appear;

http://www.google.com/search?q=%22The+term+%E2%80%9Cmullet%E2%80%9D+is+rarely+used+in+a+positive+light.+Generally+it+conjures+up+images+of+bad+haircuts,+but+this+fashion+week+the+mullet+has+a+new+life%E2%80%94and+a+chic+one+at+that.+The+%E2%80%9Cmullet%E2%80%9D+skirt+has+been+seen+on+runways+from+BCBG+to+Diane+von+Furstenberg+to+Jason+Wu.%22&hl=en&safe=off&prmd=imvns&ei=f_h5Tr2lA9O48gP83ZEh&start=0&sa=N&biw=1280&bih=911

http://www.google.com/search?hl=en&source=hp&biw=1280&bih=911&q=%22When+it+comes+to+jeans+for+plus+sizes+we+could+go+as+far+as+saying+that+this+is+an+area+for+plus+sizes+that+is+not+lacking.+There+are+a+number+of+plus+size+denim+brands+%28including+straight+brands+like+Paige+Premium%2C+James%2C+and+Citizens+for+Humanity+which+come+in+plus+sizes%29%22&oq=%22When+it+comes+to+jeans+for+plus+sizes+we+could+go+as+far+as+saying+that+this+is+an+area+for+plus+sizes+that+is+not+lacking.+There+are+a+number+of+plus+size+denim+brands+%28including+straight+brands+like+Paige+Premium%2C+James%2C+and+Citizens+for+Humanity+which+come+in+plus+sizes%29%22&aq=f&aqi=&aql=&gs_sm=e&gs_upl=1040l2372l0l2536l3l3l0l0l0l0l188l373l0.2l2l0

Although your site appears in the search engines for these long exact match phrases - a rather large number of other sites appear too.

If you are only copying snippets out from other sites and publishing it as-is with little or no unique content - your pages generally won't rank for very long.

I would look into why your site is generating multiple urls for the same page first - so far it would seem the duplication is occurring because of the query string being appended to the url.

If you can prevent that - the next thing is to use the canon tag to point the duplicate pages at the original, and on the dupe pages add a "follow noindex" tag - this means you will be telling google that you know this page exists and you are aware it is the same as another page, and tell them which page it is you actually mean.

The canon tag should be in the head, and have the full url of the original page that you want to rank.

The duplication of the page is the issue here - changing your titles will only mask it for a few days.

Duplicate content on site is not good - google basically sees the pages and thinks you are creating pages for the sake of it to occupy more of the results and gain more traffic.

What to do about it.

Although not ideal - you could just remove the pages and wait for the recovery - but this would likely mean a large loss of urls.

Best bet is to either sort your site structure out so you don't have the sub folder pages in the first place - and canonical tag the spare pages back to the originals so you are telling google which page is the one you want to rank for. On the pages that you add canon tag - add "follow noindex" to your meta tag so that google knows your intent.

Redo your xml sitemap and resubmit through GWT and you should see it sort itself out.

Another option would be to 301 the duplicate pages - but I would be inclined to go with the canon rather than build up a lot of 301's

I would agree with Manischa - also, look at the age of the page itself, if it is new it will naturally have a low score that should pick up in time anyway. Relevancy though as always is the key.

The answer is right in your question - "runs a number of whitelabel sites". As mentioned, it is largely due to the original publisher publishing the content first and getting indexed - from there, anytime the google bot stumbles across the same content - it will figure out that it has seen the content before, and attribute the ranking to the original. Something that google themselves covered last year here (although more specifically for news at the time).

Duplicate content unfortunately isn't just "not shown" by the search engines (imagine how "clean" the SERPS would be if that were the case!) it's just ranked lower than the original publisher that google is aware of. Occasionally you will get the odd page that will rank from a different domain - but that is usually due to being fresh content, I have seen this myself with my own content being aggregated by a large news site - they might outrank me on occasion for a day on one or two pieces - but my original url comes out on top in the end.

A very handy guide here on regex specifically for this kind of thing, with a bit of tweaking you can also filter right down to number of words in the phrases too if you want to (handy if you are looking for more specific conversion phrases)

I have seen the same thing and posted a private Q along the same lines, having seen substantial changes with a couple of site profiles, both of which have had link building done very carefully with high value sources. Majestic and Google WMT still show the links - just a spanner in the works at the mozplex by the look of things until the next crawl.

Response as follows:

| Kenny Martin | Aug 31, 2011 |
| Customer Service Expert at SEOmoz |

Sorry that you're seeing such drastic changes with the new Linkscape update. We've been adjusting the way that we do our crawls over the past couple of months, and the updates made in July were a bit off the mark. Our new crawler was built to be fresher, but it was going deeper into domains, and, unfortunately, not visiting as many domains. Domains with a high MozRank were getting crawled deeper, but domains with middle to lower MozRanks were not getting crawled. With our newer Index, we tried to get the domain diversity back up to or better than that of our last update (as was originally designed) so we didn't crawl as deeply into some of these higher ranking domains as we had for our July index.

Also, keep in mind that it can take some time for links to be included in our Linkscape index because it takes a few weeks to crawl the web and a couple weeks to process all the data that is acquired.

I hope that helps clear things up and we are anticipating on improving our crawler and index each month. My apologies for any inconveniences

As has been mentioned - links pointing to you need to be relevant and related. The place to start is look in depth at your competitors. Ascertain who they are - and see who links to them. If you can gain links from the same places that are relevant and regarded as authorities, and leave out their spammy links it will give you a good start.

Once you have found your competitors - read this post on using googles custom search engine for potential link mining here on SEOmoz, and delve into the world of search parameters to get your search for potential links on target.

When you are seeking links - avoid specifically going for just links to your home page - link diversity is key. If you are approaching an authority site about a specific product and you have a guide on how to use it or positive reviews on it, ask for a link to the deep page to ensure the relevancy is there.

In general I would agree with Ryans comment - paid links are a no-no as far as google is concerned.

Yahoo directory and business.com appear to be the few exceptions to the rule though (many companies list in both without any detriment), there may well be a specific reason for this, but one I won't claim to know!

With association sites, most will word their t&cs to state that you are paying to be a member, and by being a member it entitles you to a link, thus escaping googles TOS. As to how viable associate site links are will be dependent on the site on which the link is, it's traffic, rankings and so on.

In general - tools like this are rarely worth the time or hassle at all, as has been said - however, thats not to say that all directory submission is a waste of time.

If a directory stands up to scrutiny - then submission can be worth it, for example:

It has a good DA and the subject section has a good P/A

It has a relevant section to your niche

It will actually give you a link and not strip it out after you post it.

The page on which you link is displayed is not buried deep in the directory structure (like dmoz)

The directory and its pages are not stuffed with adsense or advertising

The directory is not full of dupe content.

The chances of any bulk submission software or tool finding a significant number of directories that meet the above conditions are virtually non-existent, and the handful that will apply to your business and be worth getting a link on will be reasonably easy to find yourself.

In addition to EGOL on the news front, your news needs to be unique. Aggregate or scraped news won't last long in google.

With respect to the site, are you dealing with duplicate content? After taking a large phrase from a new news item and wrapping it in quotes, the SERP page shows a large number of pages with different titles/urls.

Without knowing how your site is structured it's a bit difficult to give specific advice, but run the SEOmoz custom crawler over your site and see how much duplicate content you have or are creating.

Unless you have a very very broad site subject, chances are if you submitted to a large number of directories, very few (if any depending on your niche) would be relevant, and only a handful of those would pass any real link juice. If you find a few quality directories in your niche they could help, just have a good look at their status first before submitting.

The time and cash would be better spent (as already stated) on gaining links from credible sources.

This is a topic I feel needs extensive testing. As you have said, although google has not commented on the subject to any great degree other than best practice (with respect to relevant linking) it would appear that there is methodology out there that would suggest the comparative quality of inbound links does have an effect - Eric Ward/Adgooroo are avid promoters of "domain profile" when it comes to inbound links.

The penalty side of things has only come into play recently with the PANDA update - many sites that rely only on syndicated content to exists have been penalised - as a result, perfectly legitimate sites that have syndicated content out with links in have suffered .

Of course the efficiency factor comes in to play too - both from the time and effort needed to gain a high quality link, over the "loads of links in 10 minutes" through directories/blog comments/syndication. We all know that 10 high quality, relevant .edu links will earn way more brownie points than 100 "easy" links from non-relevant sources.

I suspect if you took a large site with a high number of back links and reduced the number of spammy links - you might, over time, see an improvement in ranking. This is of course an assumption though, would be interesting to see if anyone has experimented.

You're right, Google doesn't appear to have an opinion on length that is has voiced - more pointing out the quality of the content itself - my statement was more from what I had seen here and there on various publications, and on also as a result of generally having seen pages with more content on performing better organically over time that pages with thinner content (this is not to say that the 500 number was a max at all - far from it, I too have pages 2 or 3 times that size). I would rather the task of optimising performance of 500 word pages than say 250 word pages.

Apologies to all for not being a bit clearer with my original answer

Completely agree with everything mentioned, what would be interesting to hear though - and possibly amusing - would be why your client thinks this might be a good idea?

Quality first, and enough of it to avoid looking "thin". Post Panda, the general consensus of opinion appears to be around the 400 - 500 word mark.

As much as it is more work, that aim from the SE angle is to show results that are going to answer a searchers query.

The way in which SE's can quantify this is through bounce rate, time on site, and page views per visit. Bounce rate is fairly well known - high bounce rate means a good chance your content is not relevant and/or not ranking correctly for the terms it is targeting. Time on site, although not openly taken into consideration by google publicly - if you think about it, the more time a prospect spends on your site, the more chance of them performing your chosen call to action. Pages per visit will come naturally as a result of mastering the previous 2 points, and again increase your chance of a call to action happening.

I would agree with Alan on this one - user experience and relevance are paramount. I get approached by guest bloggers regularly that are outside my niche and have to turn them down, and would also be expect to be turned down if I did the same.

In a simplistic form, imagine going to a local bicycle shop that only sells bicycles, to find they had started to sell deep sea diving helmets - a cyclist would be confused, and a deep sea diver would be highly unlikely to even consider coming into the shop anyway. Regular customers would likely think the reason for the new products would simply be down to some kind of deal, rather than expanding on a related/natural niche.

Further to what Aaron has already said - take a good look at the directory you are going to list in. Is it tidy and clean, or full of adsense? Does it rank ok, and do the section and region pages rank?

Also - see how deep your link will be in their structure. Dmoz used to be one of the best directories to be listed in until it got too big. If you list in the correct area now in Dmoz, most of the time your listing will end up being 5 or more directories deep, in which case the SE's are unlikely to spider it any time soon.

Well selected directory submission = great

Paid "we'll submit your site to a 500,000 directories" = waste of time and cash

If its the improvement of user experience - then the addition of relevant links is worth doing definitely, with respect to changing the content itself, I would be inclined to only make very minor edits and not disturb any of the long tail phrases you are ranking for.

Blog posts tend to be timely and topical and can date with time, (whereas articles (IMO) should be as timeless as possible and state fact). With blogging, situations change over time and so something written 6 months ago could be seen and viewed in a very different light - if that's the case, time for a new post, citing the changes and differences - resulting in both new (more well written) post, and a shiny new piece of content with links to the old post.

I find I might cover the exact same subject 3 or 4 times a year, writing a new post and citing the older ones shows consistency and awareness on your part for users, increases internal linking and subsequently increases the stickiness of your site.

Obviously if you have any glaring SEO errors or holes in your older posts it wouldn't hurt to patch them up - but time and progression are always excellent reasons for new posts and new content.

As has been said - natural? Something to consider when link building is the content you have that can be linked to. Then base your prospect hunt in niches with relevance specifically to your content - almost along the lines of treating every page you have as a landing page target. The more relevant your page, the more likely the visitor will stick to your site.

Initially i would be inclined to suggest anything that can be termed as "hidden" would be advised as a big no-no from the google point of view as everyone has said so far.

Having said that - in the source of the page you have put there - the content that is hidden is fully visible - which would suggest it is not as hidden as it might be if it were being called in from an external source. Would be interesting to here a definitive answer on this particular case though.

If you run the seomoz term target tool on the page it scores very well across the board.

Looking at the title, meta description and page content it appears to be a well optimized page for the term. For it to have no links to it at all would suggest it is a newish page, so it might drop a few pages over the next few days - it's not uncommon for a new, well written page to rank well for a week or so and then drop a little.

If you follow this guide on SEOmoz and use specific queries, you can limit the results to just show you known easier targets as far as blog commenting is concerned - and resource sites in your niche.

Although I have only used adgooroo on and off over the past 12 months, it has come along way in that time in my opinion, which would suggest it has likely progressed a great deal since your trail i would think. They have added a lot of filters to it, including a follow/no-follow filter - and the ability to segregate trust/social/geo and spam - my input as a tester was largely on the user side of things as well.

2 years ago, I would have agreed that no-follow links were a waste of time - but as we have all seen recently the SE's are allegedly giving them some credit now.

Raven tools I have not used at all - so can't comment on that I'm afraid.

As for whether adgooroo can/does find links that SEOmoz cant - to be honest - I haven't done any direct testing with one tool against the other so it would be a little unfair of me to say either "wins" in that respect.

The links that come through from adgooroo can be a little puzzling initially when you look at them to start with - but then their methodology to link building does differ slightly to that of SEOmoz - I have found very high quality links using both tools - but SEOmoz does more as a general all round tool.

Adgooroo as a tool (in my opinion) is very specific and targetted - and well worth swapping out your competitor urls every couple of months to bring in fresh data - as long as you are able to find new competitors - it holds a lot of value - I found I exhausted my first set of competitor urls very quickly and when changed to a broader set of competitors - the volume of potential links came back to life.

If I were to start at a new company tomorrow with a tool budget - I would use both tools again. Adgooroo in my mind is only really limited by your own market sector rather than by what it can find. It is also handy to have a tool that will just go and crawl your competitors and look for their links while you get on with other stuff. If the budget was limited to one or the other - SEOmoz would win because of the other tools it provides.

You could do a single month, but I suspect you would be looking at limited data, simply because of the size of job the tool is trying to do. (When I first used it we had 3 months in mind).

The mindset within adgooroo as you may have seen is along the lines of domain profile - looking at data across all of the links that point at you and your competitors - and if that mindset is maintained beyond just the simple acquisition of links - then yes, getting rid of links to your site from "bad" neighborhoods would make sense.

It will take the tool a good 10-14 days to show you usable data, from there - if you can afford to dedicate at least 2 or 3 hours a day to using it and doing the ground work to pursue the links it suggests, you will get value from it, our link profile has improved as a result.

The tool itself is based on the ideas/methods of Eric Ward (a serious link guy thats been around almost since the dawn of the web) his methodology is a little different to most (things like "anchor text is not important" and "no real difference between follow and no-follow").

Other tools - see this post here on seomoz a bit intense to start with, but not hard to set up - check the links/domains that you are targetting with the mozbar running and it will give you a good idea of the site value to you from an seo perspective.

Hope that helps!

I have and do use adgooroo link insight - both paid for and as a beta tester pre "Qbot2".

I have found it useful - and sourced links I think I would have been unlikely to find as a result, so certainly brownie points for them there. Depending on your sector or niche - you can end up running out of new link targets quite quickly - which then calls for some "outside the box" thinking and switching your competitor urls to a broader sector, and changing keywords in the same fashion.

Its certainly not a 5 minute tool by any stretch of the imagination, it will will work better for some niches than others. (I found some areas very "noisy" because sites had simply press-released their way to the top as it were)

The tool does have the odd hiccup with a few false positives - but then in fairness I have yet to see a tool in this field that is 100% - the support from them as a whole is good too.

Price wise - I personally think it is a little on the high side - and you do really need to leave it to run for the first 2-3 weeks just to gather enough data (although again this will be dependant on sector), if you are in a highly competitive market then there is value to be gained from it provided you can commit the time to using it.

There is a step by step piece here from google - it does involve building a dedicated feed for them (aside from a standard rss or xml) - there is also a couple of tags you might want to consider with respect to whether you are the original source or not.

If your press is unique and original, you should be ok. Google won't take syndicated/aggregated content though.

MSN and Yahoo I haven't specifically looked at so can't help you there, would be interested also to hear from anyone who has.

Hope that helps!

Look at the sources that your no follows are coming from using open site explorer and that will tell you firstly where they are coming from. Although not knowing your business or site, I would hazard a guess they are either from blog comments, or syndicated content that is getting picked up from an rss feed.

Then undertake a link building campaign to gain more followed links - as you select sites to acquire links from, take a look at the site first and see if it generally follows or no follows. Obviously if they are follows, great, if not, not so great.

No follows are not all bad - if they are from a high traffic site and are getting clicks, they are serving their purpose. It is reasonably well acknowledged now that no follows are taken into account by the SE's now as social media has grown.

Of course followed links are preferable - but as the SE's change to limit page sculpting, they are now giving more weight to no follows than they used to.

Whilst we don't have syndicated content on site, we do syndicate out with success.

In November last year, google released some new tags for syndicated content although as usual did not go into a great deal of depth as to how it would all work as such.When syndicating out - there are occasions where other sites will outrank us for the same piece, but usually only for about 24 hours or so - and then they tend to pretty rapidly fall by the wayside.

Although this is speculation - I would suggest if you have syndicated content on your site, if you cite source and leave the content intact, and include the tags mentioned - you shouldn't see any problems. If you whole site is nothing but syndicated content, you might rank from time to time for a day or so at best for the odd popular piece - but as google crawls it will inevitably detect who published first and credit them accordingly, leaving syndicated content to drop in ranking.

From what i have seen its a little unclear, and would largely be dependant on how you create the pdf. Provided your pdf has been created using a text editor for the text (and not mad up of a bunch of images) then if pdf's are crawled - at least you stand a chance of the text and so on ranking in the first place. (You can google search by file type, including pdf, so one would assume they should rank in their own right if tagged and as text).

Whether a pdf will pass rank or not? I would suggest it should - provided it actually ranks itself in the first place (sounds obvious I know).

If you have a very large site, check google webmaster tools, and run a custom SEOmoz crawl to look for a rise in site errors. Larger sites appear to have been heavily crawled pre-panda and subsequently shown large numbers of errors on some sites as panda was rolled out.

If your content is all unique, i would check your site structure/performance and see if there are any nasties there first.

Meta-descriptions are an opportunity for you to instruct the SE's what your page is about specifically. If you leave it blank - they will extract an excerpt that they think applies to the page. Whilst this is ok, it might not be exactly what you want to display on the SERPS.

Use the meta descriptions to make the most of the SERPS page, if you page is about blue widget investment - then your description should have text that is relevant to that term, and contain text that includes or is at least similar to the term - if it is a large page and you leave the field blank - the SE's can pull in a snippet of an entirely different part of the page that isn't relevant at all.

As for titles - If you have old pages that have long titles it can be a bit of a drag redoing them - but if you are continually producing content I would suggest restricting to sub 70 characters from here on from an "adhering to best practices" point of view.

SEO advantage I would suggest is negligible, if any in this case - more a case of aesthetics as has been suggested, as well as character count saving.

Section pages I would go with "Section | Company Name"

Product pages I would go with "Product - Section | Company Name"

I have used this in the past http://www.auditmypc.com/free-sitemap-generator.asp - (Click on the image in the top right of the instructions) a free tool for site map generation that will show broken internal links in the process. I don't think it has any limits to it, although I have not tried it on a site as large as you are suggesting. Just ensure you are not logged into your site when you run it. Although Google webmaster tools is ok, you can't verify changes made very quickly.

I would be inclined to use option 2 - provided the new replacement products are a close match, and the pages are well written/optimised. Option 3 is fine to start with, but as you acquire more archive pages over time, you then end up having to deal with the headache that comes with archiving in it's own right.

Another option might be is to take the original content from the old page and append it below the fold on your replacement new product page, and label it as previous products - at least them you will still have brand names and item codes on-page giving you the chance to still rank for older terms - and offer the equivalent replacement. (still 301'ing the old page to the new).