Will I Lose My Rankings if I 301 Redirect a Page After Google Made the Page Rank Changes?

[Ryuzaki]

I made that title for the sake of anyone who searches a similar question in the future. I tried last night and couldn't find anything of the sort. It seems like nobody has spoken at length about using 301's since Google made the claim that 301's don't lose page rank any more.

The Story Thus Far

To fill everyone in:

• It used to be that 301 redirects, whether from an external site or from an internal page to a new URL, would lose exactly 15% of the page rank flowing through it.
• Recently when Google pushed for everyone to go HTTPS, they made the change that 301 redirects would lose no page rank if the change was from HTTP to HTTPS, to incentivize SEO's to make the leap since there were so many horror stories from before.
• They then later made the statement that all 3xx level redirects pass 100% page rank now with no dampening, no matter where they're from and where they're going.

My Past Experiences with 301 Redirects

I personally went HTTPS earlier this year and it worked 100%. Saw no traffic drop, didn't see any traffic increase either but I was rising regardless so it's hard to say.

I've bought sites and ported the content in and 301'd everything over, from homepage to homepage, contact to contact page, post to post. The posts themselves weren't ranking for anything big before but I did see a general rise in my traffic overall. It seemed to have gone just fine.

I've done simple 301's where I just change the slug or category and the new post was indexed, but I never bothered to make sure their traffic transferred over because they were inconsequential blog posts and I was just shortening URLs at the time.

My Incoming Tests

Of course I wouldn't just ask without contributing, so I'll be running a test too. But the question is, have you done this on a page that was ranking for terms that require links and juice, and did you see any negative effects or did it transfer over just fine?

If it goes well, and based on what other people say in the thread, I may or may not make the changes I'm thinking of making.

The reason I may not is because it's 100% aesthetic. It'll get 2 keywords out of a set of money URL's and look prettier, and set me up for some other changes I could make organization-wise. The problem is, there's really not much upside and a ton of downside. It's risky because 90% of the income hinges on these URLs. It's kind of dumb to do, but at the same time I'm a stickler for what I want for my own property and I think it's ridiculous for me to not make a change to my own site because of Google's unpredictable nature.

Anecdotal Evidence of Success

As I tried to search around, I found only one post that had anything to say about the topic, and I found it by surprise. The post was about 301's from before the change and confirming the 15% loss of page rank as stated in the patent. But at the top, there was an update to the post stating that since the changes were made by Google, that 301's are going down perfectly now with no loss of ranking.

This is that post by Wayfair.com on Moz: Accidental SEO Tests: How 301 Redirects Are Likely Impacting Your Brand

It's worth a read even thought it's barely relevant. They do 1000's of 301's constantly among other tests like screwing up their title tags accidentally. They're sharing some nice data.

Your Experiences?

I do want to ask everyone else on BuSo:

Have you done any internal 301's where you change the URL slug or category recently in the past 6 months?
​

Please fill us in here in the thread if you have anything insightful to add or better questions to ask! I'll come back with info about my tests as soon as I get them in place.

 

[Ryuzaki]

All of this thinking led me to check all of my old redirects. I didn't track any of them because none mattered, but I remembered moving my first post ever to another URL way after the fact and recalled having gotten some links to the page. So I pulled up the info on it in Ahrefs on the old versus new location. Here's the results:

​

The 301 occurred somewhere in November I think. The black line is a separation between the pictures, and the red arrow shows where the continuity is between them, and I circled September on both to try to highlight this.

This was before Google publicly announced that they quit dampening 301 redirects, although it appears they already had quit sucking out the 15% of page rank juice. The two images weren't at the same scale so I tried to stretch and match them the best I could. The highest peak on the right should be higher than the one on the left.

It doesn't show perfectly in this final result but I confirmed I lost no traffic and kept ranking for the same amount of terms I was before & after.

So that's one with some real data we can chalk up to a success.

 

[Encrypted]

I had a category page that ranked #1 for a 5.4k/month search term for about a year. Initially , i was cool with it but in the face of recent algorithm updates , i decided to de-index all my categories from search.

In mid August this year , I wrote a post on this keyword and 301ed the category page to the article. Everything was fine for the first 2weeks and the new post with a new url was still ranking #1.

Then all of a sudden , the post disappeared from position 1 to position N/A despite still indexed , and it stayed like that for weeks until late October when it suddenly reappeared in top 5.

It has crawled back to the top and is currently ranking position #1 and #0 (Snippet).

 

[Danxd]

Interesting.

I’d assume one of the main reasons for the lack of sufficient info on the topic is that the majority of similar tests/results are somewhat inconclusive and often impacted by a number of variables.

There’s a very thin line in measuring where natural progression & change attribution happens, even with 1:1 redirects where the amount of variables is limited to a minimum (updated external & internal links, speed, site structure, seasonal outliers, etc.).

With that said, I’ve seen the least fluctuations with HTTPs only switches (which is where the variables are limited the most as the only change is in the protocol + all relevant updates in internal linking etc.). I’m yet to see any conclusive improvements based on HTTPS alone, but overall the transitions happened without any notable drops in traffic or rankings on a page level, so I’m inclined to think 301 redirects generally transfer most of the value.

The below goes slightly above 6 months but it’s a page with a changed URL structure that was picked up rather quickly and value transferred over swiftly. Same content but slightly improved internal link structure across the site due to the URL parameter trimming, which is surely impacting this.

And then we have a site-wide CMS change where content on all pages had to be altered (quite improved as a matter of fact), yet this took a notable dip and will likely take months before everything is re-evaluated.

From what I've seen in terms of actual data, 301 redirects work as intended when other factors are as isolated as possible. When it comes to content changes, however, the efficiency of redirects is often irrelevant.

 

[Ryuzaki]

I wrote a post on this keyword and 301ed the category page to the article.

Thanks a ton of the input. My guess for the drop and delay was that your rank initially transferred over like it should, but then the algorithm realized there was a mismatch in content so it had to see if the new content was worthy of ranking for the term, and it was. It probably would have ranked without the 301. But the mis-matching of content might be what caused the bounce, if I had to guess.

With that said, I’ve seen the least fluctuations with HTTPs only switches (which is where the variables are limited the most as the only change is in the protocol + all relevant updates in internal linking etc.).

Same. I did the HTTPS switch earlier this year and it went over without a single issue. I didn't gain any traffic like some sites claim, but I think that boost is too small to matter when compared to link metrics. A smaller or newer site might notice the difference better.

From what I've seen in terms of actual data, 301 redirects work as intended when other factors are as isolated as possible.

Yeah, I get what you're saying. When you change the content at the same time as redirecting, the content 'values' have to be recalculated too, screwing with the test. As well as if internal linking is juggled around.

___________________

I started prepping for the first test last night. It'll be that same post that I 301'd without issue before. I wrote it as the first post on my site before I knew what my categories would really be, so it never really had a proper home.

But I know where it'll live now and can go ahead and create this new category. So I'm going to create a test project in SERPWoo to watch a handful of rankings. Once SERPWoo has them locked in, I'll make the change. That way I can compare positions in SERPWoo and traffic in Google Analytics, plus relative positions and traffic in Ahrefs as well.

If that goes over well, I'll do the same for one of the money posts. If that goes well, I'll do it again for 2 more. Then I'll do 3 more, as long as everything goes without a problem. Eventually it'll all be done.

I'm nervous since this is entirely a vanity thing for aesthetics, on the worst batch of URLs I could choose that shoulder all of the cash flow. We'll see... for the greater good of the SEO community! For NARNIA!!!!

 

[Ryuzaki]

Okay, so I gave SERPWoo a couple days to establish some rankings for about 10 keywords, and I took note of where Ahrefs pegs everything in terms of organic traffic and # of keywords ranking. I just did the 301. I'm also about to change all of the internal links pointing to the page. There will be nothing left lingering or pointing to the old URL. We'll see what happens and how fast it does it.

I'm not going to request a crawl and index of the new location or old. I want this to be totally organic so we can see how the behavior is. I'll try to check for when the page is crawled or not though to keep track.

 

[turbin3]

Over the past year, I've been doing TONS of redirects. Been trying to get one large site in order, that's been a #pageception hell on the duplicate/thin content side of things. That plus a few TLS migrations of some blog/magazine style sites.

From it, I saw some interesting behaviors:

Large Site Redirects

• A lot of these were long tail, no links required KWs. Even some of the short tail really didn't need any links, or very few.
• Goal for a lot of this was 301s and eliminating lots of longtail pages, aggregating to short tail variations to create fewer pages, stronger content.
• Saw positive ranking changes within ~2-4 weeks. A lot of the issue is crawl rate optimization, since it's such a large site.
• Also definitely saw a lot of the long tail rankings aggregating to the new short tail destination. Sometimes saw 3-7 day gaps in rankings, as the redirected long tail pages dropped off rankings, before the short tail came on, but again probably a crawl frequency issue.

Small Site TLS Migrations

• Did them earlier this year, so entirely on the "no 301 link juice loss" side of things
• For ~3-5 keywords, 8K+ vol/mth, definitely links required
  • The pages fluctuated a small amount in rank but recovered within a few days to a week roughly.
• For long tail variations of those KW
  • Awesome data from SERPWoo. As soon as I can get back into my account, I'll grab a few screenshots. I had a number of examples of a totally SEAMLESS rank transition from HTTP -> HTTPS.
  • In at least 1 or 2 cases, I saw a transition happen within the same day. At first I didn't understand why the charts had vertical lines until I realized it was the TLS version coming on tap within HOURS!
  • Most of these long tail did not have competition to demand links, and most ranked without them.

I monitor my traffic logs obsessively, to keep an eye on crawl behavior and make sure things aren't going off the rails. Considering the behavior I've seen on these ~5 sites over the past year, I'd definitely try to pin down the crawl frequency for those pages @Ryuzaki. As long as it's getting seen quick, I think you should see a quick result.

 

[Ryuzaki]

3 days after I did the 301, Google still hadn't migrated to the new URL. I looked in Webmaster Tools and saw that they encountered a 503 error on the original URL, so they must have crawled as I was setting up the 301. It didn't seem like they ever came back, so I requested a crawl and within 10 minutes the new URL was showing in the SERPs.

So the real question will be if the rankings get disrupted or not, or if there's a nice seamless transition. If it's seamless, I'll keep moving on those important posts, the money posts.

With that being said, I just checked out SERPWoo:

​

Zero drops. Perfect ranks moving right over to the new URL. I see this on 2 of the terms so far. I won't see the full results until tomorrow once SERPWoo runs the rest of the checks.

I want to ride it out and see what Ahrefs has to show too, since I'm using them to track total number of ranking terms and not specific terms' rankings.

Looking like Google has this running pretty flawless. Let me point out though that all I'm doing is changing the URL. I'm NOT:

• Merging small posts into larger ones
• Changing any content
• Moving to a different domain
• Changing from HTTP to HTTPS

This change is simply changing the URL behind the slug. So it's like... domain.com/folder1/the-slug being changed to... domain.com/folder2/the-slug

Like @turbin3 said, I also did a straight HTTP to HTTPS migration with no other changes and it was perfectly seamless.

 

[Ryuzaki]

So that initial test had a chance to marinate over the holidays.

I chose a low traffic choice that didn't matter. You see some variation with holiday traffic but I'm almost willing to say that not only did it go over with perfect results, but it may have benefitted from the 301. This benefit would only arise from the URL hierarchy behind the slug changing to include more relevant terms.

This gives me some hope that I might see a boost on these important posts I'm about to start with while conversions are down post-holidays. I'm going to start with 8 of the lower earning posts and see how it goes. If those go over well in the next few days, I'll do the rest in one shot, which is about 30 total. This will include:

• Keeping the slugs the exact same except for a couple I want to clean up a little
• Removing a folder and a word from another folder (removing 2 words total)
• Changing all internal links from around the site to point to the new URLs
• Adding the 301's in .htaccess
• Asking for each to be crawled and reindexed in the Google Search Console

I'm mapping out all of the links and work right now in a spread sheet so when it's time to fire I can make it all happen real fast. The first eight will get their 301's in place today.

I'll report back with how it goes, with a couple more graphs, but after that there'll be way too many to show proof so you'll have to take my word for it beyond that point. Hopefully this helps others who end up in this situation be a little more confident in the outcome. So far it's looking real good with the test above.

 

[Ryuzaki]

Quick update:

The 8 posts seem to have 301'd over fine. I set up the 301's internally, changed all my internal links, and then asked Google Search Console to recrawl the original URL. It notified that it saw a redirect, and I requested indexation.

Within hours, each one was indexed under the new URL if you searched the title or did a "site:" search with the URL. But if you searched for the ranking keywords, it was still showing the old URLs.

That was on the 5th, later at night. I see in SERPWoo that by the 8th Google had finally propagated the change to where it showed the new URL when you searched the queries they were ranking for.

The ranks didn't drop or jump. It was just a straight transition that took about 3 days.

Hopefully this helps you guys know what to expect if you do this, and also to feel comfortable going forward with it if you're unsure.

I'm going to 301 the other 22 or so posts ASAP. I'll come back with traffic graphs and all of that on the highest trafficked pages later so we have some proof on some posts with big numbers.

 

[Ryuzaki]

Final Results:

I put the final touches on all of the 301's on January the 12th. The three pages you see in the image below are the last ones I did later in the day.

​

What you're seeing is three pages, but for the first page I show a before & after as well. Instead of 30 and 40 pageviews per day, these are in the 200 to 400 per day and are money makers. These are representative of all of the 30 that I moved.

As you can see, there was zero interruption. On all of them you can see a slight lift in traffic, maybe some moreso than others. I don't think they'll be permanent but it was interesting to see.

The key takeaway was everything went perfect.

All traffic, it shouldn't be a surprise that the traffic stayed the same since the 301 would push everyone over to the new URL regardless, but the important part is that Google never dropped the rankings. They never fell out of the SERPs or even slipped a few places. Traffic stayed exactly like it would have if I hadn't done this at all.

I noticed that Google would show the new URL if you did a site: search for it, nearly immediately, if not within hours (I did ask them to recrawl each original URL so they'd see the redirect). The new URL started propagating across the actual search terms over the course of a few days. There are a few still where SERPwoo is seeing the old URL, but when I check manually I saw the new one. So it might still be spreading across the data centers, but with the 301 in place it doesn't matter.

Conclusion:

It's safe, in late 2017 and 2018 and beyond, to implement 301 redirects. Google seems to react quickly and doesn't react in any unpredictable way. It's a straight transition from old URL to new URL, if all you are doing is changing the URL and slug. No SERP bouncing, no loss of indexing, nothing. Smooth as butter.

I hope this can be found in the SERPs by all of the searchers, because I couldn't find anything, which is what led to me running this test. I also hope to never have to do this again, because it was stressful. If you choose to do it, feel more at ease, because it went fine. Zero problems.

Just map out all of your changes like I did in this thread, and create an order of operations you can roll through for each post, and there will be no surprises.

 

[Ryuzaki]

I wanted to add to this that TheWireCutter has finally absorbed TheSweetHome and shows the same as I saw above, a seamless transition, but they did it from one domain to another, where as I just moved URLs on the same domain:

​

Above is their Ahrefs data, and below is the same information from SEMRush.

​

You can view TheWireCutter 301 Redirect conversation here.

 

[luxer]

I read a recent post by Neil Patel within he claims that doing too may 301 redirects can be harmful.

"You don’t want to add too many redirects, though, since they can put a load on your servers. And that can slow down your site, which can negatively affect rankings. "

Is there any merit to what he was saying? Would this only be if there was 100s or 1000s of links pointing to said url?

I am currently syncing an Ecommerce website so doing 1000's of 301 redirects on old product links that have authority. I have been searching for more information on this.

 

[Ryuzaki]

@luxer, you only need to maintain 301's until Google has crawled them and re-indexed the new URL in the SERPs. The only other reason you would maintain them forever is if they have decent backlinks pointing to the old URL or if they have a lot of type-in/bookmark traffic to those pages.

A ton of inner page 301's that don't get any traffic besides Googlebot isn't going to hurt you that much, but for the sake of organization I'd drop them once everything was settled if they don't have links.

The other thing to consider is Neil might be talking about PHP redirects or ones coming out of a database, versus doing it right with the .htaccess, which has to be read and parsed every page load anyways. In that sense I totally disagree with him unless you make your .htaccess into some giant 10 MB text file.

 

[luxer]

Hey @Ryuzaki thanks for your insight and input. That is the thing, these urls do have links and bookmarks pointing to them already. Sometimes only a few sometimes hundreds.

Is there a quick tool you recommend to filter them by the amount of links/bookmarks pointing to them. I could clean it up in the future and leave the ones with lots of juice.