URL Enhancements

[AaronP]

Here's a question for IPS:

What sort of testing would I need to have done to at least convince you guys that the proposed URL structure is viable from an "efficiency" standpoint. >Would this be sufficient?

If that turned out to be true, would it even be considered or would I be wasting my time?

I'd hate to go through the trouble of hiring someone to do something like this to "prove" it, only to find that there'd be no interest in including this as an option in the first place. Can Matt or someone high up tell me whether this would be a waste of time or not?

Thanks

[kellenw]

Here's a question for IPS:

What sort of testing would I need to have done to at least convince you guys that the proposed URL structure is viable from an "efficiency" standpoint. >Would this be sufficient?

If that turned out to be true, would it even be considered or would I be wasting my time?

I'd hate to go through the trouble of hiring someone to do something like this to "prove" it, only to find that there'd be no interest in including this as an option in the first place. Can Matt or someone high up tell me whether this would be a waste of time or not?

Thanks

That sounds logical to me. I'd certainly be interested in seeing the results.

[Redmak]

You don't take one person's word for it (including mine) that the query result they copy/pasted is engraved in stone.

Not only that, you have to compare both the new query structure proposal to the old, under the same circumstances.

What I plan on doing:
Take a database with 5,000,000 topics and run a script that queries the database. One test using the current query structure and one test using the new query structure. Gradually increasing the queries from 1 per minute until a bottleneck is reached and we can compare the performance of the two.

This is more towards determining the viability of the structure than one person doing a query and saying "Ah ha! This one query took this long! See!"

I posted the query to show there are many different setups and therefore many different query times. Testing will only show you the power of your hardware imho

[bfarber]

To be frank, while efficiency could potentially be an issue, I don't think it is a showstopper. There seems to be a lot of focus on that as if proving the query could be made efficient (which I believe it can be for most setups) that is all that matters.

We debated on changing the URL structure fundamentally in 4.0 and decided against it. I don't believe we have any plans at all at this time to investigate removing IDs from the URL. We generally do not feel it will improve SEO or make the URLs clean enough to justify the additional code/development necessary and the changes existing clients would need to suffer through upon upgrading. It is fine to consider new sites, but we do have to consider sites that will be upgrading as well.

We also debated hierarchal URLs as we like to call them. We decided against them for now, but they are not completely and wholly off the table for all time or anything like that.

If you want my opinion - if you have years and years of experience with SEO and are convinced this is an important and logical change to make to the software, I would urge you to consider writing a plugin for the 4.0 suite that makes the changes you are describing and selling it in the marketplace. You can make some money while proving your proposals are indeed the best route to go, and of course we would be watching in the process.

Other than that, it is hard to tell you "if you do x we will do y". I can't promise any tests you run will make us change course necessarily, and I can't promise they won't either.

[AaronP]

To be frank, while efficiency could potentially be an issue, I don't think it is a showstopper. There seems to be a lot of focus on that as if proving the query could be made efficient (which I believe it can be for most setups) that is all that matters.

We debated on changing the URL structure fundamentally in 4.0 and decided against it. I don't believe we have any plans at all at this time to investigate removing IDs from the URL. We generally do not feel it will improve SEO or make the URLs clean enough to justify the additional code/development necessary and the changes existing clients would need to suffer through upon upgrading. It is fine to consider new sites, but we do have to consider sites that will be upgrading as well.

We also debated hierarchal URLs as we like to call them. We decided against them for now, but they are not completely and wholly off the table for all time or anything like that.

If you want my opinion - if you have years and years of experience with SEO and are convinced this is an important and logical change to make to the software, I would urge you to consider writing a plugin for the 4.0 suite that makes the changes you are describing and selling it in the marketplace. You can make some money while proving your proposals are indeed the best route to go, and of course we would be watching in the process.

Other than that, it is hard to tell you "if you do x we will do y". I can't promise any tests you run will make us change course necessarily, and I can't promise they won't either.

That's pretty much exactly what I expected -- so I can at least be thankful for the honesty.

I think you guys are way, way off in the thinking behind URLs and their hierarchy but hey, can't do much about that :sad:

Maybe I could pay you guys to develop this and then you could sell it as an extension and make some money! :P

[GreenLinks]

To be frank, while efficiency could potentially be an issue, I don't think it is a showstopper. There seems to be a lot of focus on that as if proving the query could be made efficient (which I believe it can be for most setups) that is all that matters.

We debated on changing the URL structure fundamentally in 4.0 and decided against it. I don't believe we have any plans at all at this time to investigate removing IDs from the URL. We generally do not feel it will improve SEO or make the URLs clean enough to justify the additional code/development necessary and the changes existing clients would need to suffer through upon upgrading. It is fine to consider new sites, but we do have to consider sites that will be upgrading as well.

We also debated hierarchal URLs as we like to call them. We decided against them for now, but they are not completely and wholly off the table for all time or anything like that.

If you want my opinion - if you have years and years of experience with SEO and are convinced this is an important and logical change to make to the software, I would urge you to consider writing a plugin for the 4.0 suite that makes the changes you are describing and selling it in the marketplace. You can make some money while proving your proposals are indeed the best route to go, and of course we would be watching in the process.

Other than that, it is hard to tell you "if you do x we will do y". I can't promise any tests you run will make us change course necessarily, and I can't promise they won't either.

Hey Brandon ,

If you are considering implementing Hierarchical URL structure in future for Forum , i can pass rights of GreenSEO to IPB without any hesitation.

[markopolo2002]

Quote

I think you guys are way, way off in the thinking behind URLs and their hierarchy but hey, can't do much about that

AaronP, did you ever have any experience of the CSEO addon which was developed and sold by a third party some years ago?. The reason I ask is that I started my forum in 2007 and began using that mod after about a year. Our site did particularly well, incredibly over the course of the next year or so after implementing the addon, but then IPB were releasing V3 (I think it was) which hinted the inclusion of built in SEO so the developers of CSEO threw in the towel and stopped working on their program. Almost immediately after I updated my site and lost the functionality of CSEO, our site all but fell from the search results and referrals from search engines suffered immensely. We were told this would be expected and it could take 6 months for things to get back on track, but things never did get back on track and our site has struggled ever since to even make hosting fees from ad revenue.

I'm no SEO expert, but I've said for years, and since the loss of the CSEO application that the current SEO of IPB isn't doing decent sites any favours. We literally lost three quarters of our traffic from search engines overnight and have never managed to recover that back. Strange coincidence or the fault of poor built in SEO?. Maybe you could answer that question better than me !!

[Neil2]

Quote

AaronP, did you ever have any experience of the CSEO addon which was developed and sold by a third party some years ago?. The reason I ask is that I started my forum in 2007 and began using that mod after about a year. Our site did particularly well, incredibly over the course of the next year or so after implementing the addon, but then IPB were releasing V3 (I think it was) which hinted the inclusion of built in SEO so the developers of CSEO threw in the towel and stopped working on their program. Almost immediately after I updated my site and lost the functionality of CSEO, our site all but fell from the search results and referrals from search engines suffered immensely. We were told this would be expected and it could take 6 months for things to get back on track, but things never did get back on track and our site has struggled ever since to even make hosting fees from ad revenue.

I'm no SEO expert, but I've said for years, and since the loss of the CSEO application that the current SEO of IPB isn't doing decent sites any favours. We literally lost three quarters of our traffic from search engines overnight and have never managed to recover that back. Strange coincidence or the fault of poor built in SEO?. Maybe you could answer that question better than me !!

I am also one of the unfortunate victims of that as well and I am still cleaning up that mess. I did not jump on the 3.0 upgrade and actually waited longer and have suffered the damage of waiting.

I am never installing anything on my site to change the url structure again and you supporters of it need to listen to the people who has experience in this, why someone wants to risk of damaging their site is beyond me and ridiculous to want to risk it.

[AaronP]

Quote

AaronP, did you ever have any experience of the CSEO addon which was developed and sold by a third party some years ago?. The reason I ask is that I started my forum in 2007 and began using that mod after about a year. Our site did particularly well, incredibly over the course of the next year or so after implementing the addon, but then IPB were releasing V3 (I think it was) which hinted the inclusion of built in SEO so the developers of CSEO threw in the towel and stopped working on their program. Almost immediately after I updated my site and lost the functionality of CSEO, our site all but fell from the search results and referrals from search engines suffered immensely. We were told this would be expected and it could take 6 months for things to get back on track, but things never did get back on track and our site has struggled ever since to even make hosting fees from ad revenue.

I'm no SEO expert, but I've said for years, and since the loss of the CSEO application that the current SEO of IPB isn't doing decent sites any favours. We literally lost three quarters of our traffic from search engines overnight and have never managed to recover that back. Strange coincidence or the fault of poor built in SEO?. Maybe you could answer that question better than me !!

I didn't get into forums until 2010 and have never started an IPB forum before so no, I haven't heard of CSEO and don't know what they did differently than IPB.

If you change your URL structure naturally you're going to lose traffic instantly. You have to have a way to redirect old links to new ones, probably using 301 redirects so you don't lose any link juice. In Google Webmaster Tools you'll have logs full of spider errors about broken or missing content, otherwise.

If you correctly forward visitors/spiders to the new content, your rankings won't be affected and if they are it would be a short term correction followed by an increase in rankings. This is a fact. I've done it before, and Google has instructions on doing it as well. They even have a tool in Webmaster Tools for changing your entire domain name so you can bring your rankings with you if you want/need to change your domain. It's not rocket science, you just have to do it right.

[markopolo2002]

There was a great deal more that CSEO did, wasn't just about URL formatting, but unfortunately I couldn't begin to detail what those other things were. My point was, when I finally realised something was very very wrong, I raised the fact that my (and others) sites had suffered immensely and were showing little signs of improving for some reason but we were dismissed for concentrating on SEO too much. Nothing gets us away from the fact that almost everyone who used CSEO suffered the same fate ... great results with it, sites that vanished from public view overnight without it. I also asked Invision to consider purchasing the rights to the CSEO modification with the intention of possibly incorporating it into the software but that never happened and was never even considered if I recall correctly.

The CSEO modification could have been ahead of it's time and who's to say it's day may have came anyway, but I guess we'll never know and I guess I'll always have to accept that it may just have been coincidence that my site suffered a fatal blow in search engine referrals almost overnight as CSEO was pulled, thing is, so did everyone else who found themselves without CSEO !!

[blair]

Sorry I'm late to the party. Stumbled across this topic searching for something else.

1) CSEO was no vBSEO, but it was ahead of its time, and most users saw very good results. It used a default FURL e.g. tld.com/community/this-is-a-topic-1234.html. They never had a hierarchical URL. They also had other a few URL options, stop words for keywords/URLs, strip session IDs, etc. Another option they had was transliteration, or custom FURL. For example, if file names are important to your site, you could replace "." with "-" in the URL, instead of ignoring them. It also reversed page titles (board title at end) and had relevant similar topics. Don't underestimate similar content. Of course, a lot has changed in the SEO world since then.

2) If you've ever migrated a forum, or wanted to migrate a forum to another software, topic ID is critical. Why migrate? Don't look any further than vB5. Removing topic ID doesn't offer enough benefit to justify the loss of migration options. If IPS wants to lock in users, get rid of topic IDs (please don't).

3) Don't underestimate speed. While server hardware is cheap, a faster site will win the SERP (everything else equal). Even if Google has oversold the performance/SERP benefit, performance will always be important.

4) Finally, no hierarchical URL option in IPB 4 would be a HUGE disappointment. URL changes should never be taken lightly, but recently I've found search engines more receptive of change with 301 redirects. I would jump at the opportunity to have hierarchical URLs, and some of my communities have 10+ years of URL history. Myself and others have made the case before, but it's just common sense. The current URL structure is broken. Don't force the change on users, but give them the option to redirect it with a graceful 301 redirect.

Even though this topic is a very bad example URL (too many subforums), it's still relevant. If you were looking for changes to IPS URLs, which would you be more likely to click in a search result?

http://community.invisionpower.com/invision-power-services/ips-client-services/ips-software-and-company-feedback/community-suite-feedback/ip-board/url-enhancements/2471724.html

Or,

In an A:B test, even in this bad example, the first result would get a better click through. Not to mention how keyword rich the URL is (IPS, URL), to increase SERP.

P.S. I didn't make $xx,xxx last month, but my communities do have about 500k registered users, and 4M posts.

[NewRockRabbit]

4) Finally, no hierarchical URL option in IPB 4 would be a HUGE disappointment. URL changes should never be taken lightly, but recently I've found search engines more receptive of change with 301 redirects. I would jump at the opportunity to have hierarchical URLs, and some of my communities have 10+ years of URL history. Myself and others have made the case before, but it's just common sense. The current URL structure is broken. Don't force the change on users, but give them the option to redirect it with a graceful 301 redirect.

I totally agree, but unfortunately IPB have indicated that there won't be a hierarchical option which I think is a big mistake. My forum is also 10+ years old but I would willingly go through the pain to get a logical URL structure, as suggested by Google. I appreciate that many would not want to do this, which is why it should be an ACP option.

[bfarber]

Even though this topic is a very bad example URL (too many subforums), it's still relevant. If you were looking for changes to IPS URLs, which would you be more likely to click in a search result?

http://community.invisionpower.com/invision-power-services/ips-client-services/ips-software-and-company-feedback/community-suite-feedback/ip-board/url-enhancements/2471724.html

Or,

In an A:B test, even in this bad example, the first result would get a better click through. Not to mention how keyword rich the URL is (IPS, URL), to increase SERP.

Actually, the second one I would be more likely to click on because it looks far less spammy.

That said, I never (*NEVER*) look at the URL in a search engine listing. I look at the title of the result and the snippet Google shows. I was looking at a topic earlier that suggested googling "install IP.Nexus" so I'll use that as an example.

Google does not present the URL to the end user (with our software, since we use microdata to provide navigational data), so I don't think an A:B test of what an end user might click on in a search result is really an interesting measurement. The search result would need to show you the URL for that to be a measurable factor in real world use, I'd think.

[teraßyte]

Totally agree there, url #1 is completely useless for search engines and in general too imo.

When I look at a link posted somewhere (let's forget search engines right now) I want to immediately find out the topic title if it's available, I don't want to spend time figuring out the url contains all the sub-forum names (which make it look like spam too as Brandon mentioned).

[GreenLinks]

Even though this topic is a very bad example URL (too many subforums), it's still relevant. If you were looking for changes to IPS URLs, which would you be more likely to click in a search result?

http://community.invisionpower.com/invision-power-services/ips-client-services/ips-software-and-company-feedback/community-suite-feedback/ip-board/url-enhancements/2471724.html

Or,

?do=embed' frameborder='0' data-embedContent>

In an A:B test, even in this bad example, the first result would get a better click through. Not to mention how keyword rich the URL is (IPS, URL), to increase SERP.

P.S. I didn't make $xx,xxx last month, but my communities do have about 500k registered users, and 4M posts.

I kind of agree with Brandon , all though i am supporting Correct Hierarchical structure i am not supporting including all forum names inside the URLs. It looks spammy and because of the mass amount of keywords , it is not easily readable.

I think the best solution is as following

http://community.invisionpower.com/ipboard/397771-url-enhancements/

or

http://community.invisionpower.com/ipboard/397771-url-enhancements.html

This also has a huge advantage when an url is shared within any other website. Because when an end user sees

?do=embed' frameborder='0' data-embedContent>

they don't easily understand which main category the url is related to. However on first instance , everyone who see's the link will easily recognise that Url Enhancement is about Ip.Board.

Making everything clear and bright to end users is also important. With combination to even minor SEO improvement , it is really important.

[blair]

I kind of agree with Brandon , all though i am supporting Correct Hierarchical structure i am not supporting including all forum names inside the URLs. It looks spammy and because of the mass amount of keywords , it is not easily readable.

I think the best solution is as following

http://community.invisionpower.com/ipboard/397771-url-enhancements/

or

http://community.invisionpower.com/ipboard/397771-url-enhancements.html

This also has a huge advantage when an url is shared within any other website. Because when an end user sees

they don't easily understand which main category the url is related to. However on first instance , everyone who see's the link will easily recognise that Url Enhancement is about Ip.Board.

Making everything clear and bright to end users is also important. With combination to even minor SEO improvement , it is really important.

Microsoft's Community does something like this:

Windows | Windows 7 | Repair and Recovery becomes: answers.microsoft.com/en-us/windows/forum/windows_7-system/

Windows | Windows 8.1 | Repair and Recovery becomes: answers.microsoft.com/en-us/windows/forum/windows8_1-system/

Windows | Windows 7 | Devices and Drivers becomes: answers.microsoft.com/en-us/windows/forum/windows_7-hardware/

Now say you're searching for an IRQL error. The post title won't likely include the Windows version, or whether it's hardware or system related. But the forum name AND URL help narrow the search. I wonder why they left "forum" in the URL?

I can't imagine there are many forums with a subforum as deep as the example, but having an option to define the forum slug would be an improvement over the generic "topic".

A Bing search result:

If I add "hardware" to the query, I get different search results because of the FURL (not forum name):

[teraßyte]

Microsoft's Community does something like this:

Windows | Windows 7 | Repair and Recovery becomes: answers.microsoft.com/en-us/windows/forum/windows_7-system/

Windows | Windows 8.1 | Repair and Recovery becomes: answers.microsoft.com/en-us/windows/forum/windows8_1-system/

Windows | Windows 7 | Devices and Drivers becomes: answers.microsoft.com/en-us/windows/forum/windows_7-hardware/

Now say you're searching for an IRQL error. The post title won't likely include the Windows version, or whether it's hardware or system related. But the forum name AND URL help narrow the search.

I can't imagine there are many forums with a subforum as deep as the example, but having an option to define the forum slug would be an improvement over the generic "topic".

You'd be surprised to see how many forums have even more subforums...

[NewRockRabbit]

This thread has url:

invisionpower.com/topic/397771-url-enhancements/

Every thread on this forum will appear at url level to Google (and user) as a thread within the generic term 'topic'. This is not search engine or user friendly, notwithstanding IPB's use of microdata.

Instead we should have the option to remove 'topic' from the thread and replace with, for example, either the top or bottom sub-forum (which may be one and the same). For example, this page would have url:

invisionpower.com/ips-software-and-company-feedback/397771-url-enhancements/ (using 'highest' sub-forum) or

invisionpower.com/ipboard/397771-url-enhancements/ (using 'lowest' sub-forum)

Either of these would be more seo and user friendly, would not look 'spammy', would get rid of the pointless 'topic', and allow Google to properly structure forum threads.

[stepy]

IPS 4 does have the ability to change the FURLs from within the ACP.

Matt: I am planning to use very simple URLs, e.g. www.mysite/topic/12345/ being "12345" single topic ID and www.mysite/blog/12345/ being "12345" single blog ID, etc.

Are you going to give us this option in ACP, I mean to use Post ID only, Forum ID only, gallery PIC-ID only, etc. in IP4 URLs ?

Thank you

[ExsysTech]

How about just plain easy to remember URLs for people to be able to type in.... everyone is arguing over SEO and I do understand the ID in topic because there can be way to many dplicates, but for forum categories and subcategories there is no reason you could not have an off on switch for hierarchal URLs without ids in the url.

domain.com/forums/category/subcategory

not

domain.com/forums/forum/id-category

 

This has the added benefit of keeping the url inline with where you actually are in the forum where categories have duplicated sub-category names.... we have 2 categories on our site which have the same sub-category names in each.

 

While I do feel this should be baked into IPS so it is an easy on off switch for all forums there is a mod currently out but requires people to enter each url in one by one.

calendars should also be cleaned up, I can't imagine a situation where a forum has so many calendars to look up that you must include id numbers in the URL for performance reasons.

domain.com/calendar/1-comminuty-caledar just seems excessive to me, why cant those button urls default to domain.com/calendar/month

 

Topics which for performance reasons should have the ID should also be placed under their sub categories and unfortunately the above mod would not work for:

 

domain.com/forums/category/subcategory/{id}-topic-title