Anyone ever do VSL in addition to written copy?

16 replies
I'm developing a product right now that has a strong emotional element to it, so I think a VSL is optimal to better communicate emotion visually and orally. However, I need to put a lot of details into the copy so I was thinking of putting a VSL to convey the emotional message on the top of the page and then have written copy below that hammers home the practical features and benefits. I may then put another short VSL at the bottom with a recap of the emotional sales points as well as a strong call to action.

The cool thing is that I could put a call to action to read the written copy in the first VSL.

Have any of you tried this sort of thing? Have you seen anyone else do it?
#addition #copy #vsl #written
  • Profile picture of the author Cam Connor
    Originally Posted by Nathan Lewis View Post

    I'm developing a product right now that has a strong emotional element to it, so I think a VSL is optimal to better communicate emotion visually and orally. However, I need to put a lot of details into the copy so I was thinking of putting a VSL to convey the emotional message on the top of the page and then have written copy below that hammers home the practical features and benefits. I may then put another short VSL at the bottom with a recap of the emotional sales points as well as a strong call to action.

    The cool thing is that I could put a call to action to read the written copy in the first VSL.

    Have any of you tried this sort of thing? Have you seen anyone else do it?
    Yea, it's been done countless times. If you think it'll work for your market, then test it.
    {{ DiscussionBoard.errors[9748755].message }}
  • Profile picture of the author BrianMcLeod
    Based on what you've shared, I don't see any real benefit to this strategy and several ways it might work against you...

    A great VSL communicates both the emotional and logical case for buying. Not one or the other. They work in tandem... yin and yang. Hot and cold cognition (look it up).

    You can exit pop to a written version of the copy (or link to it beneath the video).
    {{ DiscussionBoard.errors[9748899].message }}
    • Profile picture of the author MikeHumphreys
      Brian is spot on.

      I'd test VSL vs. non-VSL and figure out which version converted best. If both are converting well then I'd test both of them on the same page vs. VSL with exit pop to non-VSL.
      {{ DiscussionBoard.errors[9748962].message }}
      • Profile picture of the author Cam Connor
        Originally Posted by MikeHumphreys View Post

        Brian is spot on.

        I'd test VSL vs. non-VSL and figure out which version converted best. If both are converting well then I'd test both of them on the same page vs. VSL with exit pop to non-VSL.
        I agree with Brian's comments as well, but you might as well split-test all three. No real reason not to. Once you get ~35 clicks or so from the same traffic source (assuming you're doing clicks not views) then you've got reasonably statistically significant data. So, really, there's no reason not to run all 3 at once, unless you've got a super-low budget.
        {{ DiscussionBoard.errors[9748969].message }}
        • Profile picture of the author MikeHumphreys
          Originally Posted by Cam Connor View Post

          I agree with Brian's comments as well, but you might as well split-test all three. No real reason not to. Once you get ~35 clicks or so from the same traffic source (assuming you're doing clicks not views) then you've got reasonably statistically significant data. So, really, there's no reason not to run all 3 at once, unless you've got a super-low budget.
          It depends on how well the OP is able to drive traffic. The more things you're testing at once, the more targeted traffic you'll need to get to 95% statistical confidence. I usually find that takes at least 100 actions for the a basic A/B split-test (buys for a sales page, opt-ins for a squeeze page). It's been my experience that adding multiple variations can bump up the number of actions needed significantly... which could add days, weeks, or months to the entire testing process.
          {{ DiscussionBoard.errors[9748994].message }}
        • Profile picture of the author Complex
          [DELETED]
          {{ DiscussionBoard.errors[9749109].message }}
          • Profile picture of the author Cam Connor
            Originally Posted by Complex View Post

            35 clicks?



            You need a lot more than 35 clicks.
            Nope.

            10char
            {{ DiscussionBoard.errors[9749131].message }}
            • Profile picture of the author Complex
              [DELETED]
              {{ DiscussionBoard.errors[9749301].message }}
              • Profile picture of the author Cam Connor
                Originally Posted by Complex View Post

                Thank you for reminding me why it's best not to come here anymore.

                Don't forget to thank me for saving you a ton of adspend on split tests.
                {{ DiscussionBoard.errors[9749424].message }}
        • Profile picture of the author Alex Cohen
          Originally Posted by Cam Connor View Post

          I agree with Brian's comments as well, but you might as well split-test all three. No real reason not to. Once you get ~35 clicks or so from the same traffic source (assuming you're doing clicks not views) then you've got reasonably statistically significant data. So, really, there's no reason not to run all 3 at once, unless you've got a super-low budget.
          What's "reasonable"... 60% confidence? 75%? 90%?

          Here's the formula for computing statistical significance...

          A/B Split Testing Minimum Sample Size: How Many is Enough?

          Where did 35 come from... Dreamland?

          Alex
          {{ DiscussionBoard.errors[9750038].message }}
          • Profile picture of the author Cam Connor
            Originally Posted by Alex Cohen View Post

            What's "reasonable"... 60% confidence? 75%? 90%?

            Here's the formula for computing statistical significance...

            A/B Split Testing Minimum Sample Size: How Many is Enough?

            Where did 35 come from... Dreamland?

            Alex
            Experience with PPC dear Alex. Many authorities on the subject quote around there as well. It's not an exact figure, but you're not going to get too much more statistically significant data with another 100-1000 clicks once you pass the 40 mark, if you know what it is you're doing.

            Imaginationland...

            {{ DiscussionBoard.errors[9751238].message }}
            • Profile picture of the author MikeHumphreys
              Originally Posted by Cam Connor View Post

              Experience with PPC dear Alex. Many authorities on the subject quote around there as well. It's not an exact figure, but you're not going to get too much more statistically significant data with another 100-1000 clicks once you pass the 40 mark, if you know what it is you're doing.
              You're comparing a free snack to a paid meal.

              PPC: The prospect is risking zero money of their own to click on the ad. They're curious about what the ad is offering or mentioning. It's a free snack for them. If they don't like it, it's no big deal because it cost them nothing but a little bit of their time.

              Salesletter: The prospect is risking their own money to click on the order button. Will they like the product or not? It's a paid meal for them. If they don't like it, they're out time and money. THEIR time and money. Sure, they can get their money back on the product, but that's additional time and hassle for them to do it.

              There's a huge difference in the amount of risk for them. So bottom line, there's a huge difference in the sample size needed to get accurate conversion rates between a PPC ad and a salesletter.
              {{ DiscussionBoard.errors[9751944].message }}
            • Profile picture of the author Alex Cohen
              Originally Posted by Cam Connor View Post

              Experience with PPC dear Alex. Many authorities on the subject quote around there as well. It's not an exact figure, but you're not going to get too much more statistically significant data with another 100-1000 clicks once you pass the 40 mark, if you know what it is you're doing.
              Those "many" authorities are settling for a very low confidence percentage. Which in reality is no confidence at all. Might as well just flip a coin.

              "35 clicks or so" is laughable and not statistically significant.

              You'd be wise to stop simply parroting what you've read and think for yourself.

              Alex
              {{ DiscussionBoard.errors[9752806].message }}
          • Profile picture of the author stevet563
            Originally Posted by Alex Cohen View Post

            What's "reasonable"... 60% confidence? 75%? 90%?

            Here's the formula for computing statistical significance...

            A/B Split Testing Minimum Sample Size: How Many is Enough?

            Where did 35 come from... Dreamland?

            Alex
            Hi Alex,

            I checked out the link you sent and it was good information. Thanks for sharing.

            Steve
            Signature

            Highly skilled, professional, passionate and experienced web designer For Hire.

            {{ DiscussionBoard.errors[9771070].message }}
    • Profile picture of the author Nathan Lewis
      Originally Posted by BrianMcLeod View Post

      Based on what you've shared, I don't see any real benefit to this strategy and several ways it might work against you...

      A great VSL communicates both the emotional and logical case for buying. Not one or the other. They work in tandem... yin and yang. Hot and cold cognition (look it up).

      You can exit pop to a written version of the copy (or link to it beneath the video).
      So are you of the mind that it's usually a bad idea to mix the two? Or have you seen instances when it's been done well?
      {{ DiscussionBoard.errors[9749515].message }}
  • Profile picture of the author Oziboomer
    This is definitely one of those areas that does need testing.

    There are plenty of variables to test.

    I hesitate to say it doesn't matter whether it is short or long copy, VSL or VSL with exit pop to copy.

    Some markets respond better to VSL and others to copy and I've made the mistake more than once of second guessing the market and thinking one thing will work without testing.

    Either manually test different pages or invest in split-test systems like Visual Website Optimizer or Optimizley and let them run until they have sampled a number that they deem to be significant.

    You will see various results with different levels of traffic but you need to keep going and get volumes that deliver enough valuable statistical data. I'd say 40 is too small...but it depends on the audience.

    Keep testing and modifying relentlessly ...I was going to say until you get the desired result...but really testing never ends..
    {{ DiscussionBoard.errors[9754027].message }}
  • Profile picture of the author Don Grace
    Back about 5 years ago I did a lot of "hybrids" and they did work. But these days I prefer exactly what Brian said. With a VSL you control the whole flow, but some folks would rather read than watch (myself included) so the exit pop is great (look at the end of america vsl from stansburry). I do however have a hybrid up right now that is selling, but that's only because I can't find a damn wordpress plugin that functions the way I want it to for the exit pop.

    When you test though, there's a lot of variables to consider such as is the traffic cold, or have they opted in to a list and been warmed up first? Have they been pre-sold and how targeted is it? My traffic converts way better from affiliates who do a pre-sell vs cold.
    {{ DiscussionBoard.errors[9754251].message }}
  • Profile picture of the author Coach Comeback
    not to overly generalize here... but typically I have found that VSL's convert better than anything. Especially for higher priced items.

    Sure, you canvas warriors and they will tell you that they hate them and just skip it and just want to scroll the bullets and check the price.

    But the numbers don't lie. If you do a good job of grabbing their attention and the offer is spot on, they will sit through you vsl. I have some over an hour long that convert better than any text sales letter I have ever written.

    There is no way to really pull of the emotions in text like you can in video. Plus you are hitting them on all modalities: Visual, audio and emotional/feeling with a video.

    The only thing is, a video embeded on the forum will still just have people fastforwarding through it. I typically put a headline and a strong call to action and host my vsl on my own site so I can control the play and delay content.

    WORKS!

    This is not speaking in theory. I have been selling for over a year with the same VSL's and text sales letters. This is base on my actual numbers.

    Hope that helps
    {{ DiscussionBoard.errors[9757482].message }}

Trending Topics