Major overhaul of old site, how long before search notices?

[Sgt_Kickaxe]

I had the pleasure of working with a 500 page site that had stellar content painstakingly typed into html pages one by one about 9 years ago and left largely untouched since. Before doing anything tracking was added about 3 months ago to get a baseline and the site enjoys 650 visitors from search daily. Begin the mass overhaul 4 weeks ago.

Virtually nothing but the content itself(including titles and urls) remains the same. The site is now on a popular CMS, has proper descriptions, loads quickly and is optimized for mobile and is now html5 compliant. The owner couldn't be happier.

Tracking shows a marked increase in number of pages per visit, a marked decrease in bounce rate and the beginnings of getting socially recognized.

After 4 weeks however search engines, including Google, have yet to change anything for better or worse. All keyword rankings seem to be remaining the same and the total number of visitors, on a per page basis, remains completely unchanged. As recently as 2012 such a change would have caused some sort of bump in number of visits from search, positive or negative, but it's gone completely unnoticed thus far.

I'm surprised that 4 weeks has gone by with no change as this seems an eternity in serps time. The pages have all been crawled repeatedly now so how much longer would you guess it will take before any movement in search is noticed? Has search already reached a point in which nothing but the actual content/title matter on a page? It sure feels that way.

[FranticFish]

Just to play devil's advocate, another way of looking at it could be that
(a) search has reached a point where the means of presenting content are no longer that important, and
(b) PPV, BR and speed are not used or hardly used in the algorithm - or at least are not considered important for this niche, or have not improved so much that re-ranking is a result.

Perhaps the site was already achieving what it was capable of based on the content, and if that hasn't changed, there are no ranking improvements to be gained from re-wrapping it - even if the result is to make it more user-friendly?

Either way, it's a very interesting case study.

[engine]

I assume you've used the site operator to see if the pages have actually been added to the index, even if they, as you say, have been crawled. If they have been added, I would have thought you'd have seen some changes in four to six weeks. If the pages have yet to be indexed, that would explain the lack of movement, and it'd require some other research to establish why.

What did you do with all those old pages that were in the index?

[jimbeetle]

If content stayed the same and URLs stayed the same (so any incoming links stayed the same) exactly what changes in the SERPs were you expecting?

[bumpski]

Virtually nothing but the content itself(including titles and urls) remains the same

I've come to believe that based solely on the unchanging content of a site's pages alone, Google will not re-evaluate much of anything about a site until completely new content is added by some percentage, (who knows what percentage though).

So given what you've said above, I actually believe Google won't change much of anything about the ranking of the pages on the site. You must add new content on new pages, to trigger substantial re-evaluation of the site by Google. Even adding 50% new original content to an existing page's content (regardless of URL) will not trigger a re-evaluation. Maybe 90% added, but I can't truly say.
This may all only be true if the site was already "compromised" by Panda.

I do know that Google will still honor the basics like noindex and probably robots.txt. So you can remove pages to change "reading level" statistics.

[bumpski]

If content stayed the same and URLs stayed the same (so any incoming links stayed the same) exactly what changes in the SERPs were you expecting?

Actually JimB I think he said the Titles and URL's did change, only the content remained the same. What was done with 301's wasn't mentioned.

[aakk9999]

Virtually nothing but the content itself(including titles and urls) remains the same.

The way I understood this is that the content, URLs and page titles remained the same.

So what I get from the OP is that what changed is HTML, design, addition of meta descriptions.

It is not clear if the internal navigation structure remained the same or not.

If the internal navigation and interlinking also remained the same, I would not expect any change in ranking at least for a while (until perhaps some usage signals are folded in, but this may take the time).

In fact, you can congratulate yourself that your redesign and move to CMS has been executed successfully because it did not adversely affect the site ranking :)

[FranticFish]

Google will not re-evaluate much of anything about a site until completely new content is added by some percentage

There's an interesting (blackhat) article I read recently that made a point of Google's over-fondness for fresh content and sites that provide fresh content on a regular basis.

If the site is nine years old, and largely untouched since, has the subject matter really not changed since then?

Maybe pick some pages that could be due a refresh, make a note of traffic/rankings, then carry out an update and let this be noted throughout the site (i.e. cite the updates from other relevant pages), then see what that does to the linked pages and the linking pages alike?

[phranque]

nothing but the content itself(including titles and urls) remains the same

If content stayed the same and URLs stayed the same (so any incoming links stayed the same) exactly what changes in the SERPs were you expecting?

Actually JimB I think he said the Titles and URL's did change, only the content remained the same. What was done with 301's wasn't mentioned.

perhaps jimbeetle could clarify but i understood this to mean the urls and titles did not change.

[jimbeetle]

It's a bit of a confusing sentence so Sgt_Kickaxe will have to be doing the clarifying.

[phranque]

sorry, jimbeetle - i pasted the wrong nick in there.
of course i meant to address the OP, Sgt_Kickaxe.

[nomis5]

I did a large layout change to one of my sites which only affected 100 or so pages but that represented about 90% of the page views.

Almost no change in contents but a very different design. One year later and the effect is zilch.

In fact I considered the whole exercise a waste of time! The only plus side I can see is that pages were changed to be mobile friendly but, at this point in time, it hasn't made a shred of difference to rankings.

One day however, G will get their heads around providing a completely independent set of results for mobile users - at the moment it seems they only do that for some local results.

[Robert Charlton]

If the structure and internal navigation changed, it's possible that Google takes a while to evaluate those factors. Ditto for evaluating user engagement on the site.

Hard to know where to go with this line of speculation, though, until we know whether internal navigation changed...

...and we of course don't know how Google might measure all this, but that's another question..

[Sgt_Kickaxe]

Actually JimB I think he said the Titles and URL's did change

Sorry about that, the urls did not change. Titles, actual content and urls have not changed at all. Everything else, including internal link structure, has greatly. The site is better organized as the current bounce rate and number of pages per visit suggests.

[JD_Toims]

Well, this and some of the info presented in the update thread, along with what I've seen wrt launching new sites seems to fall in line with MC's statement about making "churn and burn" tougher moving forward -- One way of doing that is to "delay trusting changes" until enough time has passed and data has been collected to determine if the change is "permanent" rather than "manipulative", much the way they've implemented a "delay" in fully trusting [passing value through] a 301 redirect.

BTW: Mobile only affects mobile searches currently, so if you're hoping a "responsive mobile design" will effect "desktop search ranking", sorry, but it's not going to happen today and Bounce Rate, Pages Per Visit, Speed aren't the "huge be-all, end-all" ranking factors they're sold as by many -- I really don't notice them as factors, since I have sites/pages with ultra-high bounce rates [which also means very low page views per visitor] that rank well for their chosen terms, and they're fast, but could be faster if I wanted them to.

[edited by: JD_Toims at 1:38 am (utc) on Nov 13, 2013]

[Dymero]

I'd probably be happy for the decrease in bounce rate (and presumably the increase in time on site). How about the bottom line metrics, if you have access to that information? If the site sells anything, are those metrics increasing? What about newsletter or forum sign-ups? If anything like that is doing better, I'd say part one of the mission is accomplished. Changed design and navigation with no hit but increased engagement or conversion is always great.

Now update that content and see what happens. I'd also produce some fresh content, since as FranticFish notes, Google loves them some fresh content.

[minnapple]

Has search already reached a point in which nothing but the actual content/title matter on a page?

I had a site all original content, banned 8+ years ago, [sorry, the actual date escapes me] due to a linking technique

I moved it to a new domain name, [no redirects] the new domain never gained traction the old one had.

My conclusion was the content foot print was banned.

The interesting thing I found about bans, is that if you do not take any action they expire after 1000 days. Based upon 200+ occurrences

[jimbeetle]

Sorry about that, the urls did not change. Titles, actual content and urls have not changed at all. Everything else, including internal link structure, has greatly. The site is better organized as the current bounce rate and number of pages per visit suggests.

So, quite possibly (as in maybe, perhaps, for all one knows, etc.), what parts of the site that might affect rankings that actually did change, such as linking structure, did not change enough overall to influence rankings. And that also quite possibly (again, as in maybe, perhaps, for all one knows, etc.), other things that did change just don't have any impact on the rankings.

And actually, if page titles, content and URLs did not change I would expect just what you're seeing: better engagement with no or only incremental changes reflected in the SEs.