(This Changes Everything, Part 2 of 4)
Dear Friend,
In my last post, I told you about a new marketing system I’ve been developing.
For most internet marketers, if you put the new system to work, you will get more dollars from each visitor than you’re currently getting, and probably more dollars per visitor than any competitor in your niche selling similar dollar value items.
And that’s just on the front end. You’ll also naturally develop your back end as you go.
And it’s relatively easy to use — for what it does.
And the software that facilitates the system is much more platform-independent than the Split Test Accelerator.
I’m not quite ready to take beta testers yet, but I’m getting close.
In this post, I want to share a little bit about how the new system came to be.
Actually, . . .
I kind of stumbled upon it by accident.
I started out developing an interactive chat agent. But it evolved into something else.
Have you seen Mysteria (the “Happy Medium”)? She’s my interactive opt-in form companion? She sits in a box beside the opt-form, and interacts with visitors to help them understand the offer better. She can branch off into different lines of conversation, and she’s kind of fun.
Having her on the page has increased opt-ins by about 20%.
However, Mysteria is NOT the new system. In fact, she’s just a toy compared to the new system. But the new system kind of grew out of her.
It also grew out of some conversations I had with some local internet marketing friends (Ray Edwards, Kirt Christensen, Frank Sousa, Mike Morgan, Darrel and Genhi Hawes, Becky Dielman, and Mari Smith when she visited our lunch group one day).
And, finally, it grew out of 7 insights I’ve had over the last few years.
In the remainder of this post, I want to share those 7 insights with you. Some will probably be familiar to you already, and some might have immediate practical application for you.
Then, in the next post, I’ll show you how all the insights came together to produce the new system. so that the whole is greater than the sum of the parts.
Insight #1:
You can get more money, from the same traffic, if you can convert more visitors into customers.
You probably already know this. If you’re reading this blog post, chances are you are interested in split testing, and already know that higher conversion rates lead to more money.
But it’s still an insight.
It also provides an opportunity to introduce a little diagram that we’ll build on as we go. These are the same diagrams I used to explain the system recently to some of my internet marketing friends in the Spokane area.
(They’re not the exact same napkins, but it’s basically the same presentation.)
visitor value is the product of conversion rate and customer value
The block in the diagram represents the money you currently get from a traffic stream. The up and down axis represents the number of customers per visitor (your conversion rate). And the left right axis represents the dollars per customer. The product of the two values (or the area) is the number of dollars you get per visitor.
If you increase your conversion rate, it’s like stretching the block along the up/down axis.
You can increase visitor value by increasing your conversion rate.
That creates a greater area (now extended by the dotted lines). And that means more money.
Like I said, that should feel familiar. So let’s move on to insight #2 . . .
Insight #2:
The offer is the most important thing to test.
I’ve been teaching the value of split testing for a few years now. And I’ve suggested that there are many things you can try to change on your landing pages in order to improve your conversion rates.
Headlines are good things to test. So are images, bonuses, testimonials, guarantees, P.S. statements, and so on.
But FAR AND AWAY the best thing to test is the offer. Now most of my readers have probably heard this in the past, but I’m going to guess that most readers don’t test their offers much. In fact, I’ll bet many readers aren’t really sure what it means to test an offer.
Mike Filsaime recently summed up Gary Halbert’s thinking on this topic. Halbert believed that the offer was the most important thing to test, and he said that there were four components to the offer:
- What they get
- How they get it
- What they pay
- How they pay for it.
I’ve found that, even among these 4 aspects, there is one aspect of the offer that is far and away the best to test.
Most marketers spend some time testing “how they get it”, “what they pay (price)”, and “how they pay for it”. And that’s great.
But the most important thing to test is “what they get”.
Here’s an example:
If you’re getting traffic from dog lovers, and you’re selling a “poodle grooming” guide, you might make a few sales, but you’ll probably struggle.
Now you can probably increase your visitor value by testing the price, delivery method, and payment methods. And you should test these things (along with headlines, and other things).
But you know what would probably make much more difference? Testing an entirely different product. For instance, maybe a general dog training guide would yield many more dollars per visitor from your traffic stream.
Now, mind you, if you’re paying for traffic by keyword, you should probably look for more targeted keywords, but suppose you get more general traffic for free. You’re not going to turn that traffic off. Instead, you should test your offer.
But even with narrowly targeted keywords it can be extremely profitable to test your offer.
Now, here’s a question . . .
why don’t we test
“what they get”
more often?
The number one reason is because we have only one product. And we are reticent to try offering something new.
Perhaps we’re emotionally attached to our product — even to the point that we wouldn’t think of offering an affiliate product in its place — even if the affiliate product yields a higher visitor value.
Perhaps we think that an alternative offer should also be something we created, and that it takes too long to create something new just for a “test”. Or we have “sunk cost” thinking. We already went to the trouble of writing the sales page, and creating the product. So, by golly, we’re going to make THIS offer work.
But the bottom line is, if we want the best chance for increasing our visitor value, we’re going to have to test our offers. And if we’re going to test our offers, we’re going to have to spend some time bringing new offers into our stable.
Now there’s good news.
You can offer new things
without developing new products.
Here are some things that would count as a new offer:
- Your old product presented in a new way, to appeal to a new market segment.
- An affiliate product for which you receive a commission.
- A new product you create after surveying your traffic to have some confidence that it really will increase your visitor value.
According to this insight, then, the question we should ask ourselves is: “Is there something else I could offer my visitors that they would respond to better?”
In the next post I’ll discuss how the new system helps you get the right offer in front of the right visitors.
Insight #3:
A Taguchi question leads to a TRIZ contradiction.
Alright, so the label for this insight sounds a little esoteric. So let me break it down. It’s actually pretty simple (yet profound).
The Taguchi question is something I shared in the “10 Steps” course. The TRIZ contradiction should be new.
(NOTE: The new system does NOT require the use of the Taguchi method, or of TRIZ — though they can complement the system, of course. So just plow through if you’re not familiar with Taguchi or TRIZ.)
Genichi Taguchi wasn’t the first to use orthogonal arrays to speed up experiments. But he was the first to use the signal/noise ratio the way he did. And it turns out that there are three different ways to calculate the signal/noise ratio, depending on the nature of your goal.
There are three types of goals. “Smaller is better” ( perhaps you’re aiming to eliminate an effect completely), “Bigger is better” (perhaps the upper bound is unlimited, and you want to maximize some value), and “Target is best” (you want a specific value, and don’t want to miss it on the upside or downside).
So the question arises: What kind of goal do we have when we try to improve visitor value on our landing pages?
One thing is clear, it’s not “Target is best”. We don’t say “I’m aiming at a 2% conversion rate, no more, no less”.
So it’s either a maximizing or minimizing problem. But which one?
Actually it can be either one. It depends on how you look at it. And both ways of looking at the problem can be very useful.
You can frame it as a problem of maximizing the number who take the offer.
Or you can frame it as a problem of minimizing the number who leave without buying.
And it’s funny. It’s really the same question — logically. But the way you frame it affects your focus, and causes you to come up with different ideas.
When you see it as a maximizing problem, you come up with things you can add to your page to make your visitors want to buy.
And when you see it as a minimizing problem, you try to figure out what’s driving them away, you’re just trying to keep them from going away empty handed.
Now I think the maximizing mindset is the more natural one. We naturally see this as a maximizing problem. but I’ve found the minimizing mindset to be perhaps even more fruitful. It gives me a lot of my best ideas.
The mindset determines the question you ask yourself when you look at your page and figure out how to improve it. And here’s the question you wind up with when using the minimizing mindset:
How can I change something on the page to keep people from leaving without ordering?
And if we take a clue from the last insight, an even more focused question is this:
How can I change the offer to keep people from leaving without ordering?
But this question leads to a TRIZ contradiction.
Because the thing is, some people are ordering. And if I change my offer, I might actually cause some of those people to go away without ordering. And I don’t want that.
So I want to change the offer (for those who are leaving the current offer on the table).
But I don’t want to change the offer (for those who are currently inclined to take the offer).
In TRIZ, a theory of problem solving and innovation, this is known as a “contradiction”. I prefer to call it an “apparent contradiction” because the point is to find a creative way to solve the problem. And I’m kind of old school about contradictions. Contradictions (true, logical contradictions) can’t be “resolved”. Apparent contradictions can be resolved.
Anyway, . . .
The point of framing the problem as an apparent contradiction is that we often get our most creative when stating the tension very starkly. So when we get it to the point where we have an apparent contradiction, we know we’re closer than ever to a solution.
So I thought about this apparent contradiction:
I want to change the offer
But I don’t want to change the offer.
And it turns out there is a very powerful resolution. I’ll reveal more when I get the next post written.
Insight #4:
Autoresponders Allow You To Increase Your Visitor Value, Because They Allow You To Present More Offers To Your Visitors.
If you use an autoresponder, and you collect people’s email addresses, you can offer them the same thing more times, and you can offer them more things.
Both of these things increase your average visitor value. Offering the same thing more times increases the conversion rate (more customers per visitor), and offering more stuff can bring more dollars per customer.
If you don’t currently collect email addresses and send out a series of messages to your leads, you are probably leaving a lot of money on the table.
There are some niches where it’s better to just go for the immediate impulse purchase and move on. But most of the time it’s worthwhile to collect an email address and use it.
With a list you can build a case for your product slowly — a must if it’s complicated. Also, you can offer it multiple times, with different framings. This alone could double your conversion rate. Perhaps you catch your visitors in different moods, or different circumstances. Or you just wear them down with repetition.
Finally, you can offer multiple products.
And offering multiple products can increase the average number of dollars you get from each customer.
The back of napkin diagram goes like this:
You can increase visitor value by increasing the amount the average customer spends.
Again, the size of the solid block represents your current average visitor value. And this time, we can make the block larger by stretching it along the horizontal axis. This happens when we can sell multiple items to our visitors, so the average value we get from each customer grows.
Now, look what happens if we increase our conversion rate, AND our average customer value at the same time.
If you increase conversion rate AND $/customer, you get a multiplier effect.
We get a multiplier effect, and really start to make our profits grow.
So, that’s insight number 4. Let’s look at insight #5 . . .
Insight #5:
People Love Quizzes That Come With Assessments
The best way (I’ve found) to get someone to opt in to a list is to offer them an assessment related to the offer, and give them a question by question walkthrough of their answers after they opt in.
Eban Pagan first got me started with this. In fact, he offered a free program called Psychic Sales Letter. I didn’t use PSL, because I had already created a custom survey script, and I modified that and used it instead. But the method works wonders if you do it right!
Eban claims that with targeted traffic (from a warm partner list with presell, I’m guessing), he achieves a 70% opt in rate using the quiz/assessment method.
And, in case you’re wondering, . . .
70% opt-in rates are not the industry norm.
Optin rates are usually much lower — especially if they come from adwords traffic instead of from JV partner lists. Recently I used a keyword on the Google content network that was a little broad (but not too bad), and my first attempt with a normal opt-in form (no quiz/assessment) got only 5% opt in with a squeeze page.
Perhaps you can relate to numbers like that.
By the way, my free adwords spreadsheet opt-in form averages about 55% opt-in rate when it comes recommended from someone else’s list. So it’s not that my opt-in page was particularly bad. I’m capable of building a good squeeze page. It’s just that the traffic stream wasn’t warmed up much, and they weren’t all that eager to sign up for what I was offering in return.
But then . . .
I put a quiz in front of the squeeze page (requiring the user to take an extra step no less) and the opt-in rate jumped to 15% immediately. Adding the assessment tripled the opt in rate.
And the way I integrated the quiz script with the assessment report, it provided one of the easiest bonuses I could have ever created. In addition to the free bonus I offered originally, I just offered them a question-by-question walk through of their quiz results.
And I have reason to believe that offering the question-by-question walkthrough alone — without the original free report — would convert even better. I haven’t tested that yet on this landing page, but I’ve seen a similar effect on another page.
Talk about an easy opt-in bonuse to create!
And here’s the thing . . .
The assessment tripled the optin rate — AND made them more receptive to my offer at the same time!
There is an art to writing the questions. But it comes pretty easily — with just a little effort.
If you haven’t used a quiz/assessment opt-in strategy, I highly recommend it.
And I’ll tell you the key: VALUE-BASED CURIOSITY, RELATED TO THE OFFER.
If you design questions that create curiosity, . . . and it’s “value-based” curiosity (in other words, the visitor thinks, “Gee, if I knew the answer to that question it would really help me out”), . . . and the questions ultimately relate to the first offer you want to present them . . . then you’ve got something very powerful.
You’ll get a higher opt-in rate, and they’ll also be more ready to buy.
Onward . . .
Insight #6:
Most Of Your Visitors Will Spend Some Money With Someone This Month — On Impulse.
Consider one stream of traffic to one of your landing pages.
For every 1000 visitors, how many buy? 10? 5? That’s pretty much par for the course for someone offering a niche info product and using Adwords to drive traffic.
If you’re lucky, extremely good, sell a different kind of product, or get your traffic from warm lists, you might do considerably better.
Either way, . . .
let me get you thinking about
how much room there is to improve.
Of those 1000 people, how many do you think will buy *something* on impulse on the internet this month?
500?
I don’t know, but 500 sounds like a good guess to me.
And how many of those will actually make more than one impulse purchase?
Many, I’m sure.
So out of 500 who will spend money on impulse this month, you’re getting 5 or 10 to buy your product.
You’ve already done the hard part. You’ve got their attention. Yet 490+ people who will buy something from someone else this month are getting away. And only 10 are buying from you.
In this “back of napkin” sketch, I’ve placed our expanding visitor value block in the context of our visitors’ impulse purchase budget for this month.
You can secure more of your visitors' monthly impulse purchase budget with the right system.
Your current share is represented by the solid block in the lower left corner (it’s probably actually smaller than what’s represented here). And you expand your share, as before, by improving your conversion rate and customer value simultaneously.
The big box represents the total amount your visitors will spend on impulse somewhere online this month.
Look how much room there is to expand!
Your visitors have deeper pockets than you think. And they’re ready to spend it on the right offer.
The new system will help you keep some of those people from getting away. And you can get bigger and bigger chunks of your visitors’ impulse purchase budget the more you build out your system.
Insight #7:
Even Very Targeted Traffic Streams Can Be Profitably Segmented
OK, this has been a long post, so let me make this brief.
I was bidding on terms like “how to flirt with women,” and was running ads in the Google content network. The ads I ran had titles like “How To Flirt With Women”. And the site is clearly a site for guys. And the offer is a book that teaches men how to flirt with women.
Not bad targeting right? I should be getting a good market to message match, right?
Well, it’s pretty good. But after surveying the traffic I found out that, actually, 15% of the traffic was made up of women who wanted to learn how to flirt with men.
Had I not surveyed them, I would have never known.
And you should know . . .
these weren’t
“fake” women either.
After directing these women to a list geared toward women, these women opted in to the “girl” list even more readily than the guys opted into the “guy” list. And they bought products geared toward women just as readily as the guys bought products geared toward men.
And these were women who clicked on an ad designed for men.
Maybe it doesn’t surprise you that people who aren’t in the market for your offer wind up clicking on your ad.
But you’ve probably thought there isn’t really much you can do about it, right? Those people will weigh down your conversion rate, and cost you money, just because they didn’t read carefully before they clicked. But that’s just the cost of business, right?
I mean, it would be nice if there were a way to intercept them after the click, and before they get presented with the wrong offer.
And it would be nice if we could figure out a better offer for them on the fly and get them to it without their even being aware of the process.
That would be nice. Catching just one mismatched segment, like the one mentioned above, would probably add 15-20% to our visitor value right away.
And who knows how many other kinds of people lurk in that traffic stream that won’t buy our primary offer, but would buy a different offer. Could we add 15% here, 8% there, 12% over here, and 5% over there — just by taking visitors who aren’t in the market for our primary offer, and getting them to an offer they are in the market for?
If we had enough offers, and could segment in a fine enough manner, we might only wind up offering the primary offer to 10% of the original traffic stream — the 10% who are most eager to buy our primary offer. We would get just as many sales from that 10% as we got from the whole original traffic stream, and we would redirect all the other traffic to other offers that are better matches for them.
If we could match it up right, we could possibly even get 10x the conversion rate this way. It probably wouldn’t work out quite that good in reality, but the principle seems sound.
It would be nice if we could intercept and redirect visitors on the fly like that.
It would be nice, but there’s really nothing we can do about it. Some mismatching is just the cost of doing business, right?
Wrong.
In about a week or so, I’ll explain how all these insights fit together into a system that will allow you to get value out of as many visitors as is possible with current technologies.
Best wishes, and stay tuned.
Sincerely,
Jim
P.S., There should be more to comment on in this post. Feel free to leave replies below.
October 21st, 2008 at 11:28 am
Awesome insights Jim.
One of the biggest insights I had in regarding Adwords was that the actual ad itself could be targeted differently using OATH (Breakthrough copy) to pull different segments from each keyword traffic.
This led me along the thought process of keyword and offer segmentation and I haven’t looked back since. I am eagerly awaiting your next post of gold.
C.
October 21st, 2008 at 11:35 am
I cant wait to see the new product!
October 21st, 2008 at 11:37 am
looks great. can’t wait to get the product
October 21st, 2008 at 11:44 am
More and more intriguing, Jim!
The Napkin Art is helping me get a sense of the possibilities … and the high payoff of implementing ideas like this.
Write fast!
Sheila
October 21st, 2008 at 12:18 pm
Hey Jim,
Some great insights here that – even for the seasoned pro – that are great reminders of what we all SHOULD be doing
My best guess on this that it’s a much *much* slicker version of Ebens Psychic Sales letter… with ability to segment quiz takers by responses, put them on appropriate autoresponder lists, and show them appropriate offers according to their answers.
If it is something like this, it’d be very powerful with broad untargeted traffic – such as the content network.
Now if you could incorporate MV testing into the landing pages, emails and sales pages…
George
October 21st, 2008 at 12:48 pm
Hey, Jim.
Your idea sounds pretty slick, along the lines of what I’ve thought about alot myself.
Would you say that the application of TRIZ helped you much in coming up with your new idea?
Marty Foley
October 21st, 2008 at 5:23 pm
Wow that was a long post to read. Can’t wait to see how this turns out.
October 21st, 2008 at 8:40 pm
Jim, you’re brilliant (just stating what’s obvious to anyone who knows you), and I can’t wait to see what you’ve come up with.
If you need a beta tester, I’d love to use it on a wealth management site. I have a feeling I would have quite a few clients who would buy your service/script
. Each and every visitor who becomes a client is worth $500 – $15k the first year.
You’ve got my email!
Sarah
October 21st, 2008 at 11:08 pm
You are onto something big Jim. I have always had an issue surrounding segmentation. So much of what you said spoke directly to me.
October 21st, 2008 at 11:47 pm
Very interesting..
Quiz people, taking advantage of their natural curiosity of ‘how they perform’ while at the same time letting them tell you what they ‘want’ and then direct them to different offers on the fly.
Am I close?
I think it was Halbert who also said ‘Find out what they want, and give it to them’…
We used a similar idea I learned at Eben’s Altitude event to suvey our list. Created a new product straight from the responses. The result — 24% conversions!
October 22nd, 2008 at 11:06 am
Your guesses are definitely “in the ballpark” now.
Marty — I didn’t actually sit down with the problem and apply TRIZ to it, but I think I was probably influenced some by the principles.
I’m using the reference here more as a way to focus the readers’ thinking than to recount how exactly the ideas came about.
October 23rd, 2008 at 7:37 pm
Jim,
Sounds very interesting. I’m interested in finding out how you’re pulling your segmentation off.
Burton