How to Instantly Double Your Online Sales and Profits.

37 replies
I remember coming here, to this Warriorforum Board back in 2003, then under the "shawnlebrunfitness" moniker.

Back then, I had reached some success online, after starting in 2001 I had gotten
my sales up to about $500 a day or so.

But I knew my sales letter for my first product could be better.

So, I went onto this forum and asked for feedback about my sales letter,
my opt in page, and traffic.

Being pretty new, I just wanted to get some feedback.

Well, that was the problem. I got so much conflicting advice, my head
was spinning and I didn't know what to do after.

I had some people say the sales copy sucked, I had others say that my
traffic source was junk.

Again, being rather new... I didn't really know who to listen to.

I'll never forget the PM that changed my business and my life.

It was from a pretty well known marketer, and he must have felt
bad for my rather "hey tell me what to do" approach.

He sent a PM that said "it's a mess to guess, so test".

Well, needless to say, that one message changed the course
of my business and life over time.

Because I immediately started using Google Adwords to split
test every single thing I could think of.

Using the same exact ads, but just rotating them to 2 different
landing pages... I kept testing, and testing, and testing.

If something won, I'd keep it and test it against another thing.

I kept doing this and over the course of the first year, my sales doubled.

And to this day, I don't guess about anything. I don't ask for opinions, none
of that.

Because DATA NEVER LIES.

By simply testing and tracking everything, you can get absolute
proof what works best, and it's coming straight from the market.

From the Facebook ad, to the opt in page, to the sales page,
to the emails... all of it can be tracked.

And what can be tracked, can also be measured. And what can be
measured can be improved.

Believe me, if you're focused on tracking every single thing you do,
you can SPOT where leaks are happening, where people are dropping
out of your funnel.

You can test new leads or new headlines... you can test new
opt in pages.

Where before, if you just asked other marketers, you're going to
get a rash of different thoughts and ideas.

Why bother?

Just set up a split test and ask the only person that cares...
your prospect.

Let them tell you the best solution... by letting them vote with
their wallet.

I know it sounds simplistic, but I guarantee that more than half of the
folks reading this aren't testing every single part of their funnel... from the
moment someone sees your ad, to the moment they're on your list
or buy from you.

When you track it all, you can SEE things that you otherwise
wouldn't be able to. Problems, leaks, or missed opportunities.

Again, it's a mess to guess... so just test and track every move
you make online.

It will pay off, big time. It doubled my business the first year.

Data never lies, so when it doubt, test it out... and build your
business on proof, not guesses.
#double #instantly #online #profits #sales
  • Profile picture of the author JohnMcCabe
    "Data never lies"...

    True enough, but sometimes it can speak in riddles.

    And if you don't speak data, it can send you on wild goose chases.

    One thing I've noticed is that, even among the people who do try to test, few understand when the results of a test mean something. Any test result could be the result of random chance, faulty test procedures, or something else.

    If you run a test to 500 results, and Option A gets 251 while Option B gets 249, can you really say that Option A is better? Outside of a local election, would anyone really be comfortable declaring A the winner?
    {{ DiscussionBoard.errors[11081893].message }}
    • Profile picture of the author shawnlebrun
      Originally Posted by JohnMcCabe View Post

      "Data never lies"...

      True enough, but sometimes it can speak in riddles.

      And if you don't speak data, it can send you on wild goose chases.

      One thing I've noticed is that, even among the people who do try to test, few understand when the results of a test mean something. Any test result could be the result of random chance, faulty test procedures, or something else.

      If you run a test to 500 results, and Option A gets 251 while Option B gets 249, can you really say that Option A is better? Outside of a local election, would anyone really be comfortable declaring A the winner?
      John, good point!

      It's why I'll typically wait until I get close to 1,000 clicks or so, total, before I start to make assumptions on the winner.

      I don't remember what the number was, but Gary Halbert used to declare a winner after 1,000 or 2,000 "actions"... I've also found that the more clicks you can get, the more likely that one will become a clear winner.

      Too few clicks and you're right... you often leave it to chance.
      {{ DiscussionBoard.errors[11081961].message }}
    • Profile picture of the author ryanbiddulph
      I am weird, but agree with the riddles take John. I go a bit beyond, even; when you get really clear on your direction, doing things less and less from fear and scarcity, the data just reflects your energy/intent.

      Sure I check stats and spot patterns but the more detached I get from the data as my prime marker, the less I fear and the higher those stats raise.
      Signature
      Ryan Biddulph helps you to be a successful blogger with his courses, manuals and blog at Blogging From Paradise
      {{ DiscussionBoard.errors[11082100].message }}
    • Profile picture of the author Jordan Mortimer
      When doing these sorts of tests, if your results aren't clear cut, simply run more numbers until you get a clear winner. When running tests, the more numbers you run, the closer you get to the actual true probability.
      {{ DiscussionBoard.errors[11082941].message }}
  • Profile picture of the author oppyeaunome
    Ever since I've started working online, I've found this to be the truth. Testing is the only way to find what works. When you test it out, and you get data, then you can make adjustments and try again.

    If you have an idea test it out. See what happens. Learn from the results.

    It's only after doing this kind of testing over time will you be able to find what works and when you do you can optimize and scale for more profit.
    {{ DiscussionBoard.errors[11081923].message }}
  • Profile picture of the author Tsnyder
    For every newbie posting threads looking for advice... stop... pay attention
    to this guy... it will be the last post you'll ever need to read.
    Signature
    If you knew what I know you'd be doing what I do...
    {{ DiscussionBoard.errors[11081952].message }}
  • Profile picture of the author PromoA
    Your should think buy yourself. Ask yourself,your friends and family what they want what others don't have. Try to offer it to your customers.
    {{ DiscussionBoard.errors[11082037].message }}
  • Profile picture of the author Mark Singletary
    Good thread!

    I always get a chuckle out of those that claim they tested how often to email and daily or multiple times daily worked the best every time without fail they say.

    When I ask them if they've tried weekly, bi-weekly, or monthly they don't answer because they haven't and won't test such a frequency just because some guru said the more the merrier.

    All the while folks like Terry Dean, Jimmy Brown, Paul Myers, Jeff Smith, Dave Ramsey, etc. all seem to be doing just fine with much less frequency.

    My point is that if you are going to test something, test it through a valid range and don't let our beliefs get in the way. Might as well not test if you are going to do what you want to do anyway!

    Mark
    {{ DiscussionBoard.errors[11082056].message }}
  • Profile picture of the author Mark Singletary
    Many years ago there was a "guru" (haven't seen him around in a while) that discovered by testing that putting a header or graphic or logo at the top of a sales letter caused conversions to go down.

    He announced his results and all the sudden that is the way sales letters were written by many.

    However, he was in the MMO area and it may have worked for them but single moms or overweight adults or dog lovers may have had a totally different reaction. Graphics may be better for other markets. Test things for your own market/niche!

    What do you think of those that publish their results so that people can get a baseline?

    For example, I've seen gurus say start with red/blue/whatever and test from there. Or they'll say start with daily emails, the baseline, and test from there?

    Mark
    {{ DiscussionBoard.errors[11082061].message }}
  • Profile picture of the author JohnMcCabe
    Originally Posted by shawnlebrun View Post

    John, good point!

    It's why I'll typically wait until I get close to 1,000 clicks or so, total, before I start to make assumptions on the winner.

    I don't remember what the number was, but Gary Halbert used to declare a winner after 1,000 or 2,000 "actions"... I've also found that the more clicks you can get, the more likely that one will become a clear winner.

    Too few clicks and you're right... you often leave it to chance.
    We still have to define "a clear winner". That's where the notion of statistical significance come in. Even if there appears to be a clear winner, how certain are you that the difference in actions is due to the variable being tested, and how much due to random chance.

    The law of large numbers (often mislabeled 'the law of averages') says that as the number of actions approaches infinity, the result will trend toward the statistical mean. In other words, if you flip a fair coin an infinite number of times, the result will tend to approach a 50:50 split. At less than an infinite number of actions, there can be wild swings in the ratio.

    There are a lot of free significance calculators that can tell you whether or not a given test's results are significant (and thus able to call a winner).

    Here's a warning, though. Statisticians will ask that you specify a "confidence level" for significance. The most rigorous is 99% confidence. While this level may be appropriate for pharmaceutical trials, it's gross overkill for marketers. I like to specify an 85% confidence level. It means that I can expect the same results ( the same clear winner) about 5 out of 6 times if I repeated the test.

    Some of the calculators will let you work backwards. You tell them what your confidence level is, and they tell you how many actions you need to establish significance.

    Your preference for ~1,000 actions should give you good results, but I still prefer to let one of the calculators tell me if my results are statistically repeatable. Unless you have a huge difference in results, it can be tough to 'eyeball' a test and tell if it means anything.
    {{ DiscussionBoard.errors[11082070].message }}
  • Profile picture of the author Regional Warrior
    seriously what great thread like the old days where you learnt something and did what you were told and it worked A/B always the way

    Jason
    {{ DiscussionBoard.errors[11082261].message }}
  • Profile picture of the author amuro
    Thank you for your sharing.

    Your story has been a great inspiration to all of us.
    {{ DiscussionBoard.errors[11082339].message }}
  • Profile picture of the author laurencewins
    Shawn, as always, your threads impart solid wisdom and I say a big thank you to you. You're a person who offers great advice.
    Thanks yet again.
    Signature

    Cheers, Laurence.
    Writer/Editor/Proofreader.

    {{ DiscussionBoard.errors[11082413].message }}
  • Profile picture of the author clairelynn23
    Great information. Thanks for posting, Shawn.
    {{ DiscussionBoard.errors[11082788].message }}
  • Profile picture of the author alexsilva
    I will certainly do my tests from the information you shared.

    When possible I will come back to share my experiences.
    {{ DiscussionBoard.errors[11083498].message }}
  • Profile picture of the author Steve B
    Just a reminder for those new to testing . . .

    Test only one variable at a time. It's the only way to know what variable is causing different results.

    As a practical example: test your new headline as the only change you make to your baseline standard. If your copy is also different and you decide to add a new call-to-action, you have no way of telling which of the changes is actually responsible for the new results.

    It seems logical to only make one change at a time, doesn't it? But you'd be surprised at how many people make wholesale changes and can only guess at what's causing the new and different results.

    The best to all of you,

    Steve
    Signature

    Steve Browne, online business strategies, tips, guidance, and resources
    SteveBrowneDirect

    {{ DiscussionBoard.errors[11083727].message }}
    • Profile picture of the author shawnlebrun
      Originally Posted by Steve B View Post

      Just a reminder for those new to testing . . .

      Test only one variable at a time. It's the only way to know what variable is causing different results.

      As a practical example: test your new headline as the only change you make to your baseline standard. If your copy is also different and you decide to add a new call-to-action, you have no way of telling which of the changes is actually responsible for the new results.

      It seems logical to only make one change at a time, doesn't it? But you'd be surprised at how many people make wholesale changes and can only guess at what's causing the new and different results.

      The best to all of you,

      Steve
      Good point Steve... you need to know what actually it is you're testing, and which version works best. So, unless you're an advanced marketer who has the tools to test multiple sections at once, I'd stick with one thing at a time.

      Test a headline versus another headline. A facebook ad image against another image.

      I just created a Facebook ad this morning and all I changed was the image. The new ad is getting a 77% CTR where the previous was 56%.

      One thing at a time, or A/B as it's called... always have a new challenger trying to knock off the champ.
      {{ DiscussionBoard.errors[11083792].message }}
  • Profile picture of the author Joe Ray
    Yes, you're absolutely right about data and testing.

    But now, in 2017 you can train Deep Neural Networks and let Artificial Intelligence (AI) - Deep Learning models do the "testing" for you.

    I develop Integrated AI Systems to look for patterns in data. I use Machine Learning engines to perform predictive analytics in everything i.e. recommendation systems, sentiment analysis, social media tracking, SEO, lead generation and CRM etc.

    Today, manual testing vs using Machine Learning technologies is like walking compared to driving a Lamborghini.
    {{ DiscussionBoard.errors[11083833].message }}
    • Profile picture of the author imjourney
      Originally Posted by Joe Ray View Post

      Yes, you're absolutely right about data and testing.

      But now, in 2017 you can train Deep Neural Networks and let Artificial Intelligence (AI) - Deep Learning models do the "testing" for you.

      I develop Integrated AI Systems to look for patterns in data. I use Machine Learning engines to perform predictive analytics in everything i.e. recommendation systems, sentiment analysis, social media tracking, SEO, lead generation and CRM etc.

      Today, manual testing vs using Machine Learning technologies is like walking compared to driving a Lamborghini.
      Your point is good but isn't practical or acceptable to a wide range of audience (marketers) because data science is too complex to understand and you can't properly outsource it unless you know how these things are implemented and have a degree in computer science, which most marketers don't have. I don't think so.

      In fact, I have been utilising AI (various open source frameworks are available for simple-to-deep calculations) for years, but it took me more than half a decade to understand how to implement those technology to solve real world problems. Can you find a marketer or two who is willing to invest that amount of time to learn things?
      {{ DiscussionBoard.errors[11083846].message }}
      • Profile picture of the author laurencewins
        Originally Posted by imjourney View Post

        Your point is good but isn't practical or acceptable to a wide range of audience (marketers) because data science is too complex to understand and you can't properly outsource it unless you know how these things are implemented and have a degree in computer science, which most marketers don't have. I don't think so.

        In fact, I have been utilising AI (various open source frameworks are available for simple-to-deep calculations) for years, but it took me more than half a decade to understand how to implement those technology to solve real world problems. Can you find a marketer or two who is willing to invest that amount of time to learn things?

        I find this a tad ironic. You're a new member here and one of your only posts is to come to a thread where golden seeds are being planted and you claim it's all nothing but BS.
        Signature

        Cheers, Laurence.
        Writer/Editor/Proofreader.

        {{ DiscussionBoard.errors[11083970].message }}
        • Profile picture of the author imjourney
          Originally Posted by laurencewins View Post

          I find this a tad ironic. You're a new member here and one of your only posts is to come to a thread where golden seeds are being planted and you claim it's all nothing but BS.
          Are you a data scientist? or a simple writer? It appears you have no clue what I tried to convey here. Its applying AI technologies to solve so-called split testing. Can you do that given that you're a senior member? I don't think so. My post isn't BS but you're. What's your qualifications? have any?
          {{ DiscussionBoard.errors[11084131].message }}
          • Profile picture of the author JohnMcCabe
            Originally Posted by imjourney View Post

            Are you a data scientist? or a simple writer? It appears you have no clue what I tried to convey here. Its applying AI technologies to solve so-called split testing. Can you do that given that you're a senior member? I don't think so. My post isn't BS but you're. What's your qualifications? have any?
            You may be missing the point. You ARE a data scientist, not a "simple writer". (Side note: I can almost see you looking down your nose at us - not a pretty picture)

            Most of the people here are "simple writers" or "simple marketers", not data scientists. So your comments (and the sig you wanted to show) will go right over their heads.

            Heck, I'm not a data scientist, but I do have a degree in engineering, and I still find a lot of the esoteric artificial intelligence stuff hard to wade through.

            If you want to get a feel for how this looks from our end, go to YouTube and look up some old episodes of The Big Bang Theory, particularly scenes where Sheldon (boy genius turned theoretical physicist) tries to explain things to Penny (wannabe actress and waitress).
            {{ DiscussionBoard.errors[11084542].message }}
            • Profile picture of the author imjourney
              Originally Posted by JohnMcCabe View Post

              You may be missing the point. You ARE a data scientist, not a "simple writer". (Side note: I can almost see you looking down your nose at us - not a pretty picture)

              Most of the people here are "simple writers" or "simple marketers", not data scientists. So your comments (and the sig you wanted to show) will go right over their heads.

              Heck, I'm not a data scientist, but I do have a degree in engineering, and I still find a lot of the esoteric artificial intelligence stuff hard to wade through.

              If you want to get a feel for how this looks from our end, go to YouTube and look up some old episodes of The Big Bang Theory, particularly scenes where Sheldon (boy genius turned theoretical physicist) tries to explain things to Penny (wannabe actress and waitress).
              Hello John, your suggestions are welcomed. Surely, we need to build tools (AI stuffs) so that most common men with non-engineering backgrounds can use them with ease. That should be the motto of any scientific breakthrough; no need to show/discuss the heavy-lifting parts unless its absolutely necessary.
              {{ DiscussionBoard.errors[11084554].message }}
        • Profile picture of the author quadagon
          Originally Posted by laurencewins View Post

          I find this a tad ironic. You're a new member here and one of your only posts is to come to a thread where golden seeds are being planted and you claim it's all nothing but BS.
          I could be wrong but wasn't the post just a rebuttal of Joe Ray and not the thread as a whole.
          Signature
          I've got 99 problems but a niche ain't one
          {{ DiscussionBoard.errors[11084591].message }}
      • Profile picture of the author Regional Warrior
        Originally Posted by imjourney View Post

        Your point is good but isn't practical or acceptable to a wide range of audience (marketers) because data science is too complex to understand and you can't properly outsource it unless you know how these things are implemented and have a degree in computer science, which most marketers don't have. I don't think so.

        In fact, I have been utilising AI (various open source frameworks are available for simple-to-deep calculations) for years, but it took me more than half a decade to understand how to implement those technology to solve real world problems. Can you find a marketer or two who is willing to invest that amount of time to learn things?
        Sorry but WT# really it is not that hard ! you have two landing pages or sales pages one has copy the other a video so you test to see which one converts better as this is A/B testing

        Nothing to hard about that , you may be confused by your over thinking and remember this is a marketing site not ergonomics or IT

        Jason
        {{ DiscussionBoard.errors[11084090].message }}
        • Profile picture of the author discrat
          Originally Posted by Regional Warrior View Post

          Sorry but WT# really it is not that hard ! you have two landing pages or sales pages one has copy the other a video so you test to see which one converts better as this is A/B testing

          Nothing to hard about that , you may be confused by your over thinking and remember this is a marketing site not ergonomics or IT

          Jason
          Remember Jase we reside in the land of the Push Button Marketers where anything outside of taking your thumb and pressing downward is too hard and to much work
          {{ DiscussionBoard.errors[11084549].message }}
  • Profile picture of the author Brent Stangel
    Today, manual testing vs using Machine Learning technologies is like walking compared to driving a Lamborghini.
    Really?

    So a machine has my human pattern recognition capabilities? Has it my personal experience and knowledge of my market and prospects?

    Some of us have high powered computers right between our ears.

    Like ole John Henry, I'll take on your machine anytime.
    Signature
    Get Off The Warrior Forum Now & Don't Come Back If You Want To Succeed!
    All The Real Marketers Are Gone. There's Nothing Left But Weak, Sniveling Wanna-Bees!
    {{ DiscussionBoard.errors[11084557].message }}
    • Profile picture of the author Joe Ray
      Really?

      So a machine has my human pattern recognition capabilities? Has it my personal experience and knowledge of my market and prospects?

      Some of us have high powered computers right between our ears.

      Like ole John Henry, I'll take on your machine anytime.
      Are you serious? You should carefully reconsider your comment.

      I just did a data mining job which would have taken me 16.5 years if performed manually.

      Come on, you can't compete with computers!

      You say: "I'll take on your machine learning anytime." Really? I don't think so.

      Example: Can you quickly read 1 million product revues and tell me how many of them are positive and how many are negative? My machine learning model will do that instantly. How long would this take you?

      You asked: "So, a machine has my human pattern recognition capabilities?"
      Yes, it does, except incomparable faster and more accurate in some cases.

      Example: Can you look at an image of person's face, then quickly look through 100 million images to determine if there is an other image of this same person in the database? I think this would take you a while. How long would it take you exactly? Well, if you look at all the images, it should take you about 67 years if you work 8 hours a day, 5 days a week. But, Artificial Neural Networks - Deep Learning will do this in seconds.

      I can go on and on with examples...but I think you must get my point by now.
      {{ DiscussionBoard.errors[11084709].message }}
      • Profile picture of the author imjourney
        Originally Posted by Joe Ray View Post

        Are you serious? You should carefully reconsider your comment.

        I just did I data mining job which would have taken me 16.5 years if performed manually.

        Come on, you can't compete with computers!

        You say: "I'll take on your machine learning anytime." Really? I don't think so.

        Example: Can you quickly read 1 million product revues and tell me how many of them are positive and how many are negative? My machine learning model will do that instantly. How long would this take you?

        You asked: "So, a machine has my human pattern recognition capabilities?"
        Yes, it does, except incomparable faster and more accurate in some cases.

        Example: Can you look at an image of person's face, then quickly look through 100 million images to determine if there is an other image of this same person in the database? I think this would take you a while. How long would it take you exactly? Well, if you look at all the images, it should take you about 67 years if you work 8 hours a day, 5 days a week. But, Artificial Neural Networks - Deep Learning will do this in seconds.

        I can go on and on with examples...but I think you must get my point by now.
        Hi Joe, this guy has hardly any clue what machine learning can do for us. If not, he couldn't comment in such a way. Google is valued more than 350 Billion because of what? Can you answer? How its search algorithms parse billions of webpages almost flawlessly? Its all AI and machine learning. Do you use computer processors? In fact, you do. Its AI system. This field has been around since 70s but no one thought it would be so popular and indispensable until Google's tremendous success.
        {{ DiscussionBoard.errors[11084745].message }}
  • Profile picture of the author hsahadath
    Thanks for sharing your story with us. It is really interesting and informative. Your story will inspire us.
    {{ DiscussionBoard.errors[11084561].message }}
  • Profile picture of the author laurencewins
    Seems there are a lot of contradictions in here and even a few dog fights.
    I am not going to reply to anything after this as I'm certainly not a "simple writer" and I have "qualifications" in lots of things. It's such a shame that Shawn's original thread seems to have been *******ised.
    Signature

    Cheers, Laurence.
    Writer/Editor/Proofreader.

    {{ DiscussionBoard.errors[11084715].message }}
  • Profile picture of the author Best Seller
    Originally Posted by shawnlebrun View Post

    I remember coming here, to this Warriorforum Board back in 2003, then under the "shawnlebrunfitness" moniker.

    Back then, I had reached some success online, after starting in 2001 I had gotten
    my sales up to about $500 a day or so.

    But I knew my sales letter for my first product could be better.

    So, I went onto this forum and asked for feedback about my sales letter,
    my opt in page, and traffic.

    Being pretty new, I just wanted to get some feedback.

    Well, that was the problem. I got so much conflicting advice, my head
    was spinning and I didn't know what to do after.

    I had some people say the sales copy sucked, I had others say that my
    traffic source was junk.

    Again, being rather new... I didn't really know who to listen to.

    I'll never forget the PM that changed my business and my life.

    It was from a pretty well known marketer, and he must have felt
    bad for my rather "hey tell me what to do" approach.

    He sent a PM that said "it's a mess to guess, so test".

    Well, needless to say, that one message changed the course
    of my business and life over time.

    Because I immediately started using Google Adwords to split
    test every single thing I could think of.

    Using the same exact ads, but just rotating them to 2 different
    landing pages... I kept testing, and testing, and testing.

    If something won, I'd keep it and test it against another thing.

    I kept doing this and over the course of the first year, my sales doubled.

    And to this day, I don't guess about anything. I don't ask for opinions, none
    of that.

    Because DATA NEVER LIES.

    By simply testing and tracking everything, you can get absolute
    proof what works best, and it's coming straight from the market.

    From the Facebook ad, to the opt in page, to the sales page,
    to the emails... all of it can be tracked.

    And what can be tracked, can also be measured. And what can be
    measured can be improved.

    Believe me, if you're focused on tracking every single thing you do,
    you can SPOT where leaks are happening, where people are dropping
    out of your funnel.

    You can test new leads or new headlines... you can test new
    opt in pages.

    Where before, if you just asked other marketers, you're going to
    get a rash of different thoughts and ideas.

    Why bother?

    Just set up a split test and ask the only person that cares...
    your prospect.

    Let them tell you the best solution... by letting them vote with
    their wallet.

    I know it sounds simplistic, but I guarantee that more than half of the
    folks reading this aren't testing every single part of their funnel... from the
    moment someone sees your ad, to the moment they're on your list
    or buy from you.

    When you track it all, you can SEE things that you otherwise
    wouldn't be able to. Problems, leaks, or missed opportunities.

    Again, it's a mess to guess... so just test and track every move
    you make online.

    It will pay off, big time. It doubled my business the first year.

    Data never lies, so when it doubt, test it out... and build your
    business on proof, not guesses.
    Thank you for this post.
    Signature
    Book Publishing Shortcuts for Online Marketers: the EXACT strategies today's top independent authors are using to sell thousands of books online every year.
    {{ DiscussionBoard.errors[11084729].message }}
  • Profile picture of the author shawnlebrun
    If there's one piece of advice I give to anyone these days, it's this:

    Keep things simple.

    There is power in simplicity because it means you can easily
    test and track things, to improve them.

    the more difficult you make something, the harder it is
    to test, to implement, to track, etc...

    So, instead of wondering, guessing or asking others...
    just run a simple A/B test and let your market tell you
    what works better.

    Again, there is power in simplicity, don't over complicate
    things because when you do, it becomes harder to keep
    taking action.
    {{ DiscussionBoard.errors[11085280].message }}
  • Profile picture of the author brookeharper08
    Insightful post, Shawn! Thanks for sharing your wisdom.

    In today's world, "data" is every business' best friend. Marketers to be exact depend on reliable data to figure out what works and what doesn't. This is also the reason why data should be approached with caution. You're not really sure when you're working with bad data or not, that goes the same for sales. So before getting your hands into data, make sure your current ones are not lying.
    {{ DiscussionBoard.errors[11087044].message }}
  • Profile picture of the author amuro
    Hi, thank you for sharing.

    The best advice to listen to, if I may be blunt, is from people who have already succeeded in what you want to accomplish and have proof to show for.

    As much as you want to ask around, do NOT get advice from those who have not done what you want to do but just give their opinion.

    Having done IM much earlier than me, I am sure you know what I am talking about and why I said that.
    {{ DiscussionBoard.errors[11087053].message }}
  • Profile picture of the author rozzski999
    Good advice, I have a site generating around 30,000 hits a month and no sales- so in the same boat- just experimenting now!
    Signature
    Latest Reviews and Free Training
    http://bobbydonlinemarketing.com
    {{ DiscussionBoard.errors[11104570].message }}

Trending Topics