We all want our work to have the perfect title. And we want to be able to market it. If I were to title a play Sibboleth, it might be the perfect title for the theme of the play, but if someone searches for it, Google will ask “Did you mean shibboleth?” although at least for now it does give the “sibboleth” results.
So when we give a blog post or Kickstarter page or some other page a title, it’s not surprising that, sometime after publication, we may find ourselves wanting to update it to make it snappier or catchier or more imaginative. And we may follow up that desire by re-titling the post or page.
Following that urge can have annoying consequences. Gory details follow.
The Playwrights’ Center of San Francisco (PCSF), for which I volunteer as webmaster, is in the process of producing a benefit evening of short plays called “Repro Rights.” The link in the previous sentence is active as of this blog post, but a previous link on the PCSF home page was reported to me as broken. That link is now fixed, but the story of its breaking serves as a cautionary tale for marketing folks who like to tweak their words to perfection, even after publication, because, on the web, they can.
Several months ago, PCSF decided to do a benefit for Planned Parenthood. We would ask playwrights to submit 10-minute plays on various topics related to pregnancy, then ask directors and actors to help us present them in a staged reading on Oct. 22.
We tweeted, we posted on Facebook (is Facebook a verb yet?) and we created a pitch page on Pitch Engine. We may have created a short URL of the Pitch Engine page so we could tweet the URL easily (I’m not sure about this one, as the tweets in question are a while ago).
Before I go any further I want to emphasize that I am not specifically picking on Pitch Engine. They are providing a useful service and the issue I am about to go into occurs across many websites and web services. This is about being aware as a marketer about the possible consequences of changing your mind; what you can do about it to mitigate any damage; and to be an assertive, knowledgeable consumer.
That said, let’s begin with what happened.
We created a page at Pitch Engine titled
“Repro Rights!” Evening of short play readings to benefit Planned Parenthood
We linked to it, most likely created a short URL from it, and tweeted it. We got the directors and actors as well as other support. Hooray!
At some point, a volunteer maintaining that page decided to change the title to
“Repro Rights!” a choice evening of theatre to benefit Planned Parenthood
Here’s where the issue came in. The change in title resulted in the old URL disappearing. The page got a new URL of http://pitchengine.com/reprorights/repro-rights-a-choice-evening-of-theatre-to-benefit-planned-parenthood.
And all links to the page broke. The link from the PCSF home page broke. The link from Google broke. The link from the PCSF Facebook page broke. If any of PCSF’s friends linked to that page, their links have broken.
I’ve fixed the home page link
In this case, “broke” means the old link goes to the Pitch Engine general feed, which lists selected Pitch Engine events, but not the PCSF event.
This blog post reflects the behavior of the Pitch Engine website as of the date of this blog post, Sept. 10, 2012. If Pitch Engine changes their server behavior, I’ll be happy to update this article to reflect that they’ve updated their software.
And this is not specifically calling out Pitch Engine for this behavior. I’ve seen it on other websites and I suspect its pretty common. It needs to be fixed everywhere on the web. If you use Pitch Engine and you’re otherwise happy with them, stick with them.
Nevertheless, I’d respectfully like to make some suggestions, both to theatre (and other) marketers and to websites that provide services to organizations (and others).
- Try to decide on a permanent title before creating a page. I’m not a marketing guru, but usability expert Jakob Neilsen suggests that you keep headlines short yet informative.
- If you really want to change the title of something that’s already been up for long enough that you’ve linked to it from elsewhere and gotten into search engines:
- Copy the old title and paste it into Wordpad or Textedit or somewhere else you can exactly restore it. Also copy the URL.
- Change the title.
- Save the change.
- View your page to see it with the new title.
- See whether the URL changed.
- Clear your browser history.
- See whether you can still get to it from some other website that you linked to it from, e.g. from your website to Pitch Engine, or from Facebook to your website, and that following the link shows the new title. If you still see the old title, press the F5 key to refresh your browser. If you still see the old title, you might need to back up to the previous website, clear your browser history and try again.
- If you now can no longer get from the previous site to the site where you changed the title and see the new title, you have a decisions to make:
- Do you suck it up and change the new title back to the old title?
Consequence: Your pride.
Advantages, provided time hasn’t passed since you changed the old title to the new title:
- It works immediately.
- No further effort is required.
- You can skip the rest of this blog post. Yay!
- Do you suck it up and go fix all the now-broken links and, if you also use short URLs, generating a new short URL? Consequences:
- Your time.
- Re-posting any Facebook posts having the link.
- Having old tweets out there in the Twitterverse with the old, now broken, short link; this one is probably not a big deal as Twitter is more an in-the-moment thing and you just need to tweet the new short URL.
- Having to wait for search engines to add the new link.
- Accepting that search engines may not necessarily remove the old, now broken link.
- Do you suck it up and change the new title back to the old title?
Yes, the old link, once indexed, might sit around on search engines forever. Or at least until after the event that you so painstakingly created and (re-)titled.
It all depends on how the website is set up.
When a web page gets a new URL, websites can react in at least three different ways when they receive a request for the old URL.
- (Best) The website automatically redirects from the old URL to the new URL. In the process, they send a secret signal to the browser or search engine robot (bot) that the page URL has been replaced. The educated bot stores the new URL and uses that for the link when providing future search results. Note that there may be a delay between when the bot discovers the change and when the search results change.
- (Not ideal) The website tells the browser or bot that the old URL is not found, the old 404 Not Found, and provide a useful fallback page. The educated bot removes the old URL from it’s database and never gives it as a search result again. You have to wait until the bot discovers the new URL in its travels across the interwebs.
- (Bad) The website the browser or bot that the old URL has been replaced by the useful fallback page. The educated bot stores the fallback page in place of the old URL and gives the fallback page as the search result.
Why is number 3 bad? Not only do you have to wait until the bot discovers the new URL in its travels across the interwebs, you also have the search engine giving the wrong link as a search result for your event.
It’s easy to tell number 1 versus number 2 or 3. You tried to get the old URL and you got the new URL instead.
I’m going to have to start getting mildly technical here, and for that I apologize. If your old URL is not redirecting automatically to the new URL and geekspeak makes you break out in hives, you might want to refer this article to a techie friend or staff member. But this isn’t going to require you to do anything too complicated.
It’s not so easy to tell number 2 or 3 apart. You see the fallback page, but you don’t know what the secret signals were to the browser or what the bots will get. To be able to see that secret signal, you’ll have to install a add-on or plug-in or extension for your browser that will let you view – here it comes – HTTP headers. HTTP headers come across the interwebs in front of the web content browsers and bots request so those browsers and bots know what to do with the content.
In the Pitch Engine case, my browser connects with pitchengine.com and sends a request for the old URL:
GET /reprorights/repro-rights-evening-of-short-play-readings-to-benefit-planned-parenthood HTTP/1.1
The Pitch Engine returns content, preceded by some headers. The very first header will distinguish between number 2 (old URL disappears from searches) and number 3 (useless [to us] fallback page lives forever as a search result for our content).
If it says 404 Not Found, we’re good. But it says:
HTTP/1.1 302 Found
Ack! That tells the search engine that the content that was at the old URL has been moved to the fallback page. We’re stuck with that search result until search engines decide its no longer interesting or we go back to the old title. (And if the search engines discover the new URL before we go back to the old URL, then we’ll have the same problem with the new URL. So time is of the essence.)
So, now I arrive at my suggestions for websites providing services:
- If feasible, when a title change results in an URL change, redirect the old URL to the new URL.
- If an old-to-new-URL redirect isn’t feasible, send a 404 Not Found status code when providing the fallback page, not a 301 Moved Temporarily or 302 Found. You don’t need to show a visible 404 Not Found page; just send the status code and make sure your content is long enough to avoid display of browser built-in fallback pages.
- See whether you can deliver a more useful fallback page.
In the case of Pitch Engine as of Sept. 10, 2012, the fallback page for an un-found old URL is http://pitchengine.com/Feed. This, alas, is only likely to be useful as a fallback page for recent pitches (or whatever criteria Pitch Engine uses to populate that page). But, if I trim the old URL http://pitchengine.com/reprorights/repro-rights-a-choice-evening-of-theatre-to-benefit-planned-parenthood to remove the last part, I get http://pitchengine.com/reprorights, which gives an extremely useful fallback page, and which might even be deserving of a 302 status code.
Bonus tip: If the database is down, it’s nice to send a status code of 503 Service Unavailable. I just implemented that one at work this year.
Hope this helps.
Oh, yeah, and I’m not going to change the title of this post.