Organic Traffic Drop and Rank Increase After Video Thumbnail added
-
So my company has created a large amount of videos and I took a couple of them and created a test video sitemap to see what effect adding a video thumbnail/rich snippet to the SERPs would be.
It worked on one page and gained 2 spots (position 6 to 4) for the highest keyword, but traffic didn't increase too much. Then a week later I tested it with the page that we get the most organic traffic for, which is ranking for a very big keyword. It worked and gained a bit of ranking, but traffic decreased 50% ever since according to Google Analytics.
It seems to me that the traffic from users clicking on the video thumbnail is not be tracked as google / organic even though it lands them on the intended page/doesn't redirect anywhere else. I've looked to see if traffic to this page increased overall to see if it was being tracked via a referral or as something else, but couldn't find any traffic discrepancy. The only thing I did find is that impressions under SEO > Landing Pages > Video Property increased, but this could be from the page ranking in the Video SERPs now.
Has anyone experienced a similar situation? Do you think having a video snippet could be that big of a turn off for customers that people just aren't clicking like they used to? I don't think so, that's why I'm leaning towards a tracking discrepancy in Google Analytics.
-
Agreed. It took about a week to get the thumbs in the SERP, so hopefully I'll have some more information within a week.
-
So, I think it would be fair to say that those thumbnails are not great.
If you're a consumer looking to buy a desk, you probably want to at least see the deck you're buying, so i'd advise testing some thumbs with detailed high quality close ups of the product itself.
See if that helps. If not, It could be that for this vertical, video snippets are not particularly valuable.
-
Hi Phil,
Here are the URLs in question:
- http://www.nationalbusinessfurniture.com/Reception-Desks-Type.aspx
- The traffic didn't change too much for this page, but the rank briefly increased from 6 to 3 for "reception desk" - currently at 5
- http://www.nationalbusinessfurniture.com/L-Desks-Type.aspx
- This page experienced the traffic drop and briefly increased rank from 8 to 6 for "l shaped desk" - currently at 8
I just submitted an updated sitemap to try to change the video_loc to a product page to get the thumbnail removed from the SERPs for the L-Desks page and change out the thumbnail for the Reception-Desks page to one that reads "Shop Reception Desks" in big letters with branding, etc. I also added the autoplay directive on the updated version, which I've seen doesn't always work.
These videos are technically relevant, however, they are also product specific. We have multiple videos on our product pages so I figured using one in the SERPs for a category page would attract people to our site where they could find many more product videos.
Any insight would be appreciated!
-
Exactly this. In my original reply I mentioned "testing a different still image" and by that I meant testing the thumbnail. Make sure the thumbnail gets your attention as well as showcases what you'd want to see for the particular search queries you are targeting.
-
Hi Joanna,
as others have said, it's plausible that you have seen a decrease in traffic due to the thumbnails. I've seen this happen very frequently, but without exception this has happened when video thumbs were low quality or unrelated to the page in question.
Video thumbs can increase traffic significantly, but it's not a guarantee - one the most important factors here is what the thumbnail actually is. We could as easily say "meta descriptions can increase click through rate" - which is absolutely true, but it really matters what the description is.
I'd recommend changing the thumbs on your videos and testing a few pages to see if the thumbnail style might be a factor here. If you've increased rankings as well as having gained a snippet, and you're not competing against queries with more relevant snippets (i.e. stars for very transactional queries) then my first port of call would be testing thumbnail variations.
Would you be willing to share the URLs of some of the pages in question here?
-
Hi Joanna,
That's really interesting. I do not know the answer to whether rankings with thumbnails are treated differently upon a user clicking, but I agree with Anthony that you might be turning visitors off for some queries. It's certainly worth testing if we can confirm that the traffic is not being misappropriated in analytics.
I oversaw a campaign a year or so ago that placed thumbnails on rather transactional pages for a retailer and we did not see traffic decrease dramatically, but the thumbnails only showed up intermittently. I would certainly agree with attempting to keep thumbnails for pages where a visual guide like a video makes sense, given the results you've seen.
-
Thanks for your response. I was so excited to realize I could control which videos Google puts in the SERPs, having this traffic drop was a major bummer.
I looked in the SEO > Landing Pages section of Google Analytics (which pulls from Google Webmaster Tools - so not exact numbers) and it does show that both pages have had a dramatic decrease in CTR (from 4% to less than 1%) . However, it also shows that they've had a decrease in Impressions. Both of these pages are ranking in the top 10 for their highest keyword.
My guess would be that Google may not be serving up the page as often as they were to certain searchers and/or long tail queries because of a video being associated. I'm going to change the thumbnail on one of them and try to get the thumbnail taken off of the other by changing the video content_loc on the video sitemap to a less prominent page.
Interesting note: Since we first started embedding videos on our pages, Google had been intermittently including video thumbnails to the SERPS based on when/if Google picked up on them by crawling our pages naturally. However, when I added a more extensive XML sitemap, Google removed all the videos from the SERPs and we saw a dramatic increase in traffic with no visible dramatic change in ranks (aka the ranks we are watching didn't change).
So I'm hoping the video sitemap will ensure Google only uses the videos on the pages we tell them to. If not, I'll have to talk to my developer on how to ensure Google doesn't pick up on the video links on the page.
At least now I know how not to use videos! Interesting stuff.
-
Adding a video snippet to your search result can definitely decrease the CT rate of your organic listing for certain queries. If it's a highly transactional query, like the name of a product, and the searcher is planning on making a purchase, they may mistake your result for a page that only features a video. They will choose other results because they want to purchase.
Adam Audette has been recently talking about this fact and included a slide in a recent SEO for Ecommerce presentation he did.
You are going about this the right way by testing and monitoring your click-through rates. Perhaps you can test a different still image for the video next. Ultimately, you may end up removing the video sitemap. Be sure to test how it effects the actual conversion on the landing page as well.
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Whats (Other) traffic in Google Analytics?
When I look through all our clients, a few are receiving the majority of their traffic from (other). [ Acquisition > Channels > (Other) ]. The only option in (other) is "website" or "offline", whatever that may be. And even weirder, the avg session duration is 0:00. Any idea what this may be?
Reporting & Analytics | | W2GITeam0 -
Spam Direct Traffic
Hello, Lately, I have been receiving a big amount of unexpected direct traffic from Boston. After analyzing with Analytivs, this is what I get (please, check attachment). Normally I would be blocking this traffic source straight away from my Google Analytics account, and also blocking this traffic from accesing my servers, but check out the analytic metrics: this traffic represents 12% of my total traffic right now!!! av. session duration is 4:53 !! bounce rate is 72% !!!! pages/session 1.44 !! Service provider is "Microsoft Corporation" who looks like one of the typical spammy service providers. My question is, is this a bot?? what do you think ? Thanks, Luis zUlVHIi
Reporting & Analytics | | Yeeply.com1 -
How to filter out traffic from subdomain using regex?
Hi There, We have a test server that lives on a subdomain which we'd like to filter out in Google Analytics. We're using the hostname filter but are having trouble understanding the regex needed, we're using this as the filter pattern: .domain.com. Will that work? The profile is new so we can't test the pattern. Any pointers in the right direction would be much appreciated. Cheers Ben
Reporting & Analytics | | cmscss0 -
Domain Authority Drop
We just witnessed a domain authority drop. We went from 24 to 21.Not sure why. We've been creating fresh new content for our site over the last few days and haven't done anything that would cause it.
Reporting & Analytics | | StickyLife
Any advice would be greatly appreciated.0 -
Traffic has dramatically fallen. Why?
Hello there, On the 16th of April 2013 our website traffic dropped dramatically. See attached a screenshot of our Google Analytics. Our traffic had been growing steadily for the last 10 years. Our website http://www.wattsandco.com/ Does anyone know what could have caused this? Thanks wBeewWL
Reporting & Analytics | | roberthseo0 -
Does anyone know of a way to do a profile level filter to exclude all traffic if it enters the site via certain landing pages?
Does anyone know of a way to do a profile level filter to exclude all traffic if it enters the site via certain landing pages? The problem I have is that we have several pages that are served to visitors of numerous other domains but are also served to visitors of our site. We end up with inflated Google Analytics numbers because people are viewing these pages from our partners' domains but never actually entering our site. I've made an advanced segment that serves the purpose but I'd really like to filter it at the profile level so the numbers across the board are more accurate without having to apply an advanced segment to every report. The advanced segment excludes visits that hit these pages as landing pages but includes visits where people have come from other pages on our domain. I know that you can do profile filters to exclude visits to pages or directories entirely but is there a way to filter them only if they are a landing pages? Any other creative thoughts? Thanks in advance!
Reporting & Analytics | | ATIseo0 -
Huge Traffic Drop?
I am seeing a huge 30-40% traffic drop which started around 8-9 PM EST last night. I am compare the stats of today with last Saturday and the drop is significant. I am checking it in Google Analytics and Clicky and both report the same. The Adsense impressions drop also show the same. The rankings of my top keywords seem to be fine. I re-checked them using ranktracker for Google US, UK and others. I don't understand what has happened here. Anyone else seeing the drop? Is there a Google Panda update underway?
Reporting & Analytics | | abhim120 -
Posting on blog comments with anchor text on high ranked pages effective?
So i've identified some blogs which have a fairly high ranking and lots of traffic. They also allow anchor text in the name field. Does it make sense for me to comment on these blogs, or does google treat these with less authority that true page links? Any advice is greatly appreciated! TIA
Reporting & Analytics | | symbolphoto0