We have car research product website. We have product page for each car model that has car photos, videos, price, features, specifications, alternatives, reviews and news.

I read that product pages with videos has higher conversion rate. Is it really true? Anyone has Done some similar experiment?

Currently we show image carousel in the first fold and there's video icon just below image carousel by which user can view videos too. This video play icon has Ctr of 3 to 4 percent.

Is putting video upfront on product landing worth an experiment?

I am bit apprehensive about this as we have majority traffic on mobile website and people need to have headphones to view videos, so, would placing video really help us improve engagement and conversion?

2 Answers 2


When I open a web page, if I'm faced with a video which starts playing itself, the first and only thing I do is to immediately close the website. When I'm faced with an image carousel, the first thing I do is to scroll past it in order to see the actual content.

But, you would ask, how is this relevant to the question? You couldn't care less about what I do...

This is relevant because some conversion rate statistics for a website may not be the same as the statistics for your website. Your audience may not be the audience of another website, and what is a huge success for them may not be a huge success for you

So, what should I do?

The question you should ask yourself is why users come to your website.

  • To read the reviews about the cars?
  • Or maybe look at the glamorous photos of the shiny new cars?
  • Or to compare prices between several models?
  • Something else?

If you have created the personas, you should be able to easily answer this question. Depending on this answer, the layout of the page could vary a lot. For instance, if you established that the users are mostly interested in comparing the prices and checking the spec, the element which would appear at the very top of the page would be the model selector followed by a “Compare” button, as well as the specs of the current model. Photos, videos and reviews would either require some page scrolling in order to appear, or may even require an explicit click on an element of a page in order to be displayed.

Example 1

As an example of terribly poor user experience, I can give an article on Medium. I came to this article because I had a problem with AngularJS. What would a developer be interested in?

  • The article itself, possibly split in parts by inserting pieces of code and/or illustrations, in order to accommodate persons who are unable to focus on long blocks of text.
  • Possibly a short summary, if technically possible, for people who don't have time to read more than three paragraphs.
  • The name and the credentials of the author, if the reader has doubts about the quality of the article or if he appreciated the article and wants to learn more about the person.

Opening the web page, what do I see instead?

  • Some stuff telling that I “must agree” (thanks, that's a very welcoming start) that they will share my personal data with advertisement companies. Given that I'm in EU, this prompt is illegal since GDPR anyway, so, irrelevant.

  • The photo and the name of the author and a few info about him. Nice. I'm interested.

  • A huge, ugly photo of a creepy guy smoking. Zero relevance.

  • Advertisement.

And that's all I see. Now, I have to scroll down to even see the title of the article in order to find whether it's about the subject of my study or not. And then scroll more for the article itself. You can hardly do worse.

Example 2

Take another example of Nikon website, and more specifically a page showing a camera.

There can be different personas:

  • A potential customer who wants to buy a camera. He would usually be looking for the price, the photo of the camera itself, some promotional content highlighting the key features of the camera and demo photos made with this camera.

  • An informed customer who is choosing a new camera. He doesn't care about promotional content. He would be interested by the price, the photo of the camera (to check the ergonomics), and the detailed spec.

  • Someone who already bought the camera and is looking for either the specs or the support.

The third persona is the most unfortunate; however, even he gets mostly relevant information: “Tech Specs” tab, which appears when scrolling down a bit, and “Support” tab which, unfortunately, doesn't contain much useful info (it would be, for instance, more appropriate to put the link to the PDF manual directly within the page).

On the other hand, look at how relevant are the page elements for two other personas?! The price is here, and remains at the top when you scroll. The carousel (which, by the way, is not intrusive; you have to click to switch images) shows the product. The tabs make it easy to switch between promotional content, specs and the list of accessories. The only annoying thing—annoying because irrelevant—is the large “I am full frame freedom” zone; everything else is either completely or at least mostly relevant.


  1. Don't rely on anyone's statistics. Their audience is not yours.
  2. Define your personas.
  3. Based on your personas, define which elements are (1) essential, (2) important, (3) less important, (4) unimportant.
  4. Move (1) to the top. Ensure (2) can be visible once the user scrolls down. (3) should either appear without wasting too much space, or be moved to other pages. Get rid of (4).
  5. In doubt for a given persona, ask your users or do A/B testing or do both.

¹ Sometimes, I get a reply: “What if the statistics come from our competitor's website?” Good point. Even if the audience is the same, it doesn't mean you should use the statistics, for two reasons. First, what makes you think that the interest of your competitor is to give you for free something which could make you more powerful? What if the statistics were specifically designed for them to steal traffic from you? Second, even if we assume that the competitor published the statistics by pure kindness, this doesn't mean they were collected correctly. I've seen numerous IT persons who gather statistics to make a point, but who have a deep misunderstanding about the way statistics should be gathered, analyzed and presented. Therefore, you should only trust statistics from trustworthy scientific papers, but as previously stated, those statistics won't be relevant for your particular domain anyway.

  • For what it's worth, my reactions to the cases in your first paragraph would be identical. I suspect there's some bias towards "shiny flashy" because the people that just want to get to the key facts probably don't take part in surveys much...
    – TripeHound
    Commented Sep 26, 2018 at 13:04

For mobile, it can be a problem if you have narration or voice-overs in your video which might eventually lead to increasing bounce rate on your video.

If ever I would have been in your case, I would have gone with A/B testing as only experimentation can lead us to some conclusions.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.