Changing the title of 1000 pages slightly - is it an issue?

[MikeNoLastName]

Okay, another lame-brained situation which you've by now come to expect to hear from me. :)
A new site I started working on this week. It is a syndicated author's column. We noticed they weren't ranking highest for their own column title on their own site, but some of their syndicators WERE even though the main site has been around for 10+ years with lots of backlinks and their syndicators were only a couple years at most.
After a few minutes of looking around we realized there was a minor, practically unnoticeable, misspelling in the HTML title tag of every weekly column page going back 10 years! The column title is the first 20 chars of each title and the rest of the title describes the subject of that week's column. If you search with the misspelled version it comes up #1 everywhere, but only about 7th if you search on the correct spelling.
We can easily go in with a global search/replace and change all 1000 or so of them in seconds, but will G have an issue with it when it suddenly sees all these title keywords change?

[tedster]

I'd say you've got to do it. You might see a short term drop in traffic as Google chews its way through all the changes, but you're not adding new keywords or something like that.

I always fix typos, in the title element or anywhere else I spot them.

[AndyA]

You might try updating 100 or so pages per day, and spread that out over 10 days. I can't really say if it makes any difference, but I've always felt like making changes in smaller amounts is better than making site wide, sweeping changes all at once. I do agree with tedster, typos should be fixed when you come across them.

[LavingRunatic]

I'm no expert but I did something similar with my site. I changed a few words in every single one of 3000 pages on my site. I saw no negative effects and traffic has steadily increased since then. (I'm assuming that's because I changed to something more SEO friendly.)

[g1smd]

Even if you change all of the pages the same day, Google isn't going to respider all of those pages on the same day.

The main thing you need to be sure of is to not trigger some sort of indicator that the site has changed hands and is being repurposed.

[MikeNoLastName]

It's pretty much semantics, but G apparently makes a significant distinction between the two titles. It's equivalent to the difference between "John's Weekly Rants - What I Hate About Broccoli" (current) and "John's Weekly Rant - What I Hate About Broccoli" (correct) rather than an actual typo/misspelling that a spell checker would have found. If you search for "John's Weekly Rants", it comes up 1st, but if you search for the "Rant" version of the title it comes up 7th or worse. I just can't believe no one noticed it for over 10 years. I guess because they used a copied template and never looked at the first part of the title string.

Since there is a frequently auto-submitted sitemap on that site I think I will do as AndyA suggested, since otherwise G could SEE that all 1000 were changed all at once in the sitemap even if it doesn't spider them all right away. We can easily filter the files by year and work our way back through the years every couple days.

Thanks.

[lucy24]

If you search with the misspelled version it comes up #1 everywhere, but only about 7th if you search on the correct spelling.

That would seem to mean that g### recognizes the misspelling. (Idle but not completely irrelevant query: Do they distinguish anywhere between a misspelling and a variant spelling?) It would be another matter if you searched using the current (wrong) spelling and got results for a completely different word.

Meanwhile you've got to be losing a few potential human visitors-- first-timers or returns-- who look at the title and say "Can't even spell their own ### subject!"

[g1smd]

I just can't believe no one noticed it for over 10 years.

I can. I've seen several sites with similar long term spelling mistakes. One had an incorrect spelling of the company name on two thirds of their pages. Another couldn't spell the name of their main product.

[lucy24]

Now I'm impressed. Mike answered my question sixty seconds before I asked it ;)

[deadsea]

I routinely do title experiments where I change the titles on thousands of pages in mass to see if there is SEO improvement or not. Changing the titles to something that isn't going to rank well can hurt, but I've never seen a loss of traffic just because a change is made.

[menntarra 34]

I'm interested whether this title change really has bad affects on SERPS.
About 2 weeks ago, i saw that one of my sites had spammy titles/descriptions. So i decided to change all of the pages' titles/descriptions on this site, which means about 1.300.000 pages. The titles were like this:

##product## - Free Cheap ##product## - Buy Cheap ##product## Easily - etc

Now it is like:
##product## - Cheap ##product## - ##sitename##

What i see is that after google rescraping those pages, i see both rank drops and also some of my pages are still at 1st position in google. I haven't experienced any gain in SERPS.

What should i do? Wait a little longer, and if some keywords continue to drop in SERPS, is it wise to change back to the old description, title?

[deadsea]

Menntarra: I would have tried several different titles and changed only a portion of the titles. I would have set it up based on product id. If the product id is divisible by 5, it keeps the old title so you have something to compare to.
id mod 5 == 0: ##product## - Free Cheap ##product## - Buy Cheap ##product## Easily - etc
id mod 5 == 1: ##product## - Cheap ##product## - ##sitename##
id mod 5 == 2: ##product## - ##sitename##
id mod 5 == 3: ##product## best prices at ##sitename##
id mod 5 == 4: Cheap ##product## - ##sitename##

Then let that run for a month and see which group performs the best and use that one.

I've also done things like change the titles and see which ones change SERPS. If the SERP gets better or stays the same, keep the new title. If the SERP goes down, revert the title for that particular product.

[menntarra 34]

Yeah, first i only changed some, and wanted to test it, but then i'll decided to change all of them... Now in my situation what do you suggest? I already have a ton of my sites scraped with the new title. Maybe wait for a month and then try these titletesting or change back to the old and start the titletesting just now?

[deadsea]

I should also mention that experiments like that can be quite eye opening when you look at the outliers. In one case we were experimenting with adding the category to the title along with the product name. Our test templates were:
##product## - ##sitename##
and
##product## ##category## - ##sitename##
It turned out that having the category name helped a lot except in some cases. For illustation, I'm going to give a generic example of the category of "lawn mowers". One lawn mower was the "BrandX weed muncher". Another was the "BrandY lawn mower".
The title "BrandX Weed Muncher Lawn Mower" made a lot of sense, but the title "BrandY Lawn Mower Lawn Mower" didn't. We adjusted the algorithms to apply the category to the titles much more selectively based on whether the product name mentioned the category already. There were also some harder cases such as "BrandW mower". The best title for that might bet "BrandW Lawn Mower", but doing that programatically out of hundreds of thousands of products is a hard problem.

[deadsea]

If I were you, I'd look at the ones that did the best and the worst and see what you can learn. Look at click through rates as reported by google webmaster tools. See if they have gotten better or not. Is there are particular word in your title that users need to be able to see? You mention the product name twice. In the case of long product names, is the title name so long that your brand is being truncated? We have also had to implement shorter title templates for long product names. Set up an experiment at this point. It is never too late to experiment.

[menntarra 34]

Thanks you for all the help, i'll try experimenting with some title changes...
A lot of people here on webmasterworld suggested me not to make such titles, where the ##productname## is repeated, and they also told me that having a title like "Cheap reliable ##productname##" (so title with some adjectives) can be considered as spammy title. That's the reason i changed my titles.

An interesting thing i found in WMT is that ALL my single word pages (http://domain.tld/word), which i ranked very well, are now losing 30-50%!
While 3 word pages are gaining ranks. It is strange. And the bad thing is that my single word pages are the important ones for me of course.

[deadsea]

I've never had any success including adjectives with product names. Nor have I had any luck with duplication. Simple (but fully descriptive) titles with my brand have always best for me. The only thing that might be different now is that a really simple title with a matching simple url and the same keyword mentioned in the page may trigger an over-optimization penalty. Maybe when you tighten your titles for your one word urls, those pages suddenly look over-optimized.

[lucy24]

I've never had any success including adjectives with product names.

Does that apply both to "generic" adjectives and concrete adjectives? Say, "cheap widgets" vs. "portable widgets".

[deadsea]

Concrete adjectives can be good, but again hard to get right in a lot of titles at once.

[Planet13]

The only thing that might be different now is that a really simple title with a matching simple url and the same keyword mentioned in the page may trigger an over-optimization penalty.

Or not...

[webmasterworld.com...]

[matrix_jan]

You shouldn't have problems if you're just correcting the misspelling. I did a tittle - meta description - on page text swap few times for a website with close to million pages and hadn't had any issues.

You might have an issue with uniqueness of the titles. If the misspelling is on thousands of pages, G might take it as a "brand name" or a unique name associated with the website. In that case when you correct the word, the pages might not have the same weight before.

If you can, update few titles as mentioned above, and see what happens.