Site blew up - lost rankings only after it was fixed

[jo1ene]

For several years, I was ranking very well for a few keyword phrases which generated a small, but steady stream of referrals. Over the summer, my main site blew up. All of my clients had problems with their sites (major server issues...) and I had to fix their issues first. Mine was "more broken" than theirs because I could not retrieve the database.

In the fall, I rebuilt the website (new server - similar setup) and used archive.org to get my articles back. (No I didn't have a backup. I do now!) I came right back for my big phrases immediately thereafter, albeit a bit lower in the rankings. I used WMT to remove broken links, used htaccess and internal CMS features to redirect lost traffic, etc. I began writing articles again.

Sometime since then, I noticed I don't rank for anything anymore, just my business name. Not even close! Even with the site down, I still got distant rankings. The odd thing is that it took a few months and by then the problems were fixed.

Before - #1- 5
While broken = #20 - 35
After rebuild = #10 - 15
Now - #450 +

Ideas? Encouragement?

[jimbeetle]

Sometime since then..

It would be best if you could narrow down the date, then check the Google Update History [webmasterworld.com] thread to see if the cause was something besides the site problems, just to start ruling things out.

[Hoople]

Agree on checking Google updates. It may be a clue as to how much to expect recovery efforts to deliver.

I would check your backlinks thru several tools. It sounds like you may have left out some pages that were influential in your ranking.

Also check your logs for 404 errors - those are the backlinks driving traffic going nowhere.

[jo1ene]

FYI, I have tracked the problem down (quite by accident) to a website that created an exact duplicate of mine as of the same time my main sit tanked. I contacted their host to request it be taken down.

I found this out while trying to figure out why my publisher avatar was not showing up in the serps even though it shows as correct in the structured data testing tool. I did an exact title search (in quotes) for a unique article name and (!) up came the duplicate site.

I have been doing these types of searches to see how my pages were being listed, but had not run into this. The uniqueness of this particular title enabled me to find the dupe in such a short list.

[Robert Charlton]

jo1ene - Congratulations on tracking this down, and thanks very much for getting back to us on this. It's always helpful to the community to know how things get resolved.

I did an exact title search (in quotes) for a unique article name and (!) up came the duplicate site.

FWIW, exact title searches are not always the best way of retrieving a page of the native site. Anomalies have been reported, I believe, since the MayDay update. I think you're much better off using several words of unique text on a page, generally a clause of six or so words. If your site has been scraped, a longer string might not return the scraper, as a common scraper technique is to break sentences up by scrambling them with other content.

[gouri]

I think you're much better off using several words of unique text on a page, generally a clause of six or so words.

Should this search be performed in quotes?

[FranticFish]

Depends on how unique the character string is. Lately I've seen that quotes aren't always necessary. Without quotes will return more results, but not always what you're looking for in first place, whereas sometimes with quotes only returns one result.

[Marketing Guy]

You certain it's the copied site that is the problem? Could be a red herring.

What are the timings involved? How long was the site down for, how long after it was fixed did it take for rankings to recover, and when after that did they drop out?

What major changes were made to the site? Any content missing? Any changes to the structure? Could there be issues with the redirects?

I've had heavily scraped sites go down for weeks before and always been able to recover them pretty quickly.

I'm not saying the scraper site isn't the issue - just it might not be the main issue. Could be a combination of factors that are causing the problem.