https://MeclabsAI.com can write a powerful headline (with a guided headline pathway) and has 10 expert assistants (AI Agents and Apps - more to come) that you can talk with (like Conversion Pro, Landing Pages Copywriter, and Problem Solver). It’s totally FREE to use (for now).. ____________________________________________________________
In this Web clinic replay(http://bit.ly/HbxqEP), Dr. Flint McGlaughlin taught our live attendees how to get the maximum amount of customer insight from your A/B split tests. And through that customer insight, maximize your ROI across your marketing efforts
Learn more from Flint with a MECLABS online marketing course. Courses include certification in • Online Testing • Value Proposition Development • Landing Page Optimization • Email Messaging Optimization https://MECLABS.com/certification
all right Dan thank you very much and
we're back at it again I have uh just
saw a note come across the screen from
Christan uh and uh Christian thank you
for those encouraging comments we'll do
our best to deliver what you're
requesting the title for today's Clinic
is what your customer wants how to
predict customer behavior for a maximum
Roi it's a very interesting topic to me
it goes beyond testing it goes beyond
changing headlin and basic optimization
guidelines and it takes us to the place
where we have to think about what our
customer is actually thinking this is
the danger we as marketers have too wide
a gap we're trying to manage campaigns
rather than that sort of knowledge Gap
we have this Gap is our lack of
understanding as to what is unfolding in
the customer's mind in fact we don't
optimize
web pages we don't optimize email
campaigns but at meclabs we believe that
you should be optimizing the thought
sequence that is not possible unless you
can peer into the mind of the prospect
into the mind of the customer and ask
yourself what are they thinking what are
they struggling with what are they
afraid of as that thought process
unfolds there are there are conversion
Inhibitors and there are conversion
contributors a big part of the depth and
understanding that makes an effective
marketer is this sort of philosophical
insight into the operating software not
on your computer but in the mind of the
customer and I say all of that to share
with you what's going to happen today
I'll be joined in this Clinic with uh
with John pal John's been with me for
many years is one of our senior research
managers personally supervised the first
experiment we're going to talk about and
I want to move directly into the case
study with just one bit or one caveat
you'll notice on the on the screen
behind me we've we've kind of
Illustrated the marketer and the ideal
customer and uh we know that the
marketing effort is demanding it
requires time it requires energy and
budget significant resources we are
launching campaigns all the time being
pressured by sales pressured by
management and we're trying so hard to
discover how that we often slip past the
Y Factor how do I hit my deadline how do
I get more opens how do I improve my
conversion rate that question as
important as it is can get between you
and the most important point of the
entire mental uh exploration and that is
why why does my customer say yes why
would my customer say no why are my
customers not doing what I expected them
to do and the difference between
customer behavior and marketer
expectation a is the gap that we want to
address in today's clinic so keeping
that sort of as a context let's Dive
Right In to a case study you'll see it's
test protocol
4,067 it's a medical provider
specializing in treating chronic back
pain they're the sole providers of a
sort of uh minimally invasive it's uh
sort of innovative pain management
procedure we spoke about this particular
partner a research partner before we
worked with uh the key leaders at Roi
Revolution their analyst manage the paid
search campaign we talked about this in
a previous clinic but this is new and
what you're going to see will I think
you'll find quite interesting John pal
worked on this case study and uh there's
a primary research question now remember
it is not the hypothesis question it's
the research question and there's a
difference the research question and it
should Whenever there is a sort of
either or research scenario that happens
in single factorial that happens in
multifactorial the research question
begins with the word which but the word
which is only going to describe for you
a certain Choice it is not going to help
you answer the deeper question unless
you ask a different kind of or with a
different kind of interrogative so let's
move on and uh and let's look at the
test itself we asked you our community
of of subscribers and readers to help us
develop this test we asked you to submit
a paid search ad template that we could
use in the test and there were many
submissions the winner was Amy Harold of
GD nash.com and she submitted a
particular template we're going to look
at that template in just a moment but
you'll notice this uh connection here
between medical condition keywords and
AD abbreviated uh sorts of words that we
used in the in the actual campaign John
pal you might want to just come up here
with me and just stand with me as I
unfold this
point out anything in particular you
want to share with us about this
particular piece no we just wanted to
understand exactly what was in common
with all these particular keyword
searches and AD groups that were okay
all right all right so then here is the
treatment design and I want you to
notice that we have a variable and we
have a value the variable and this is
classic in any test design the variable
is the description in line one and the
value will be our various versions of
that description so there is a headline
it's specific to the ad groups condition
there are a group of ads and I'm going
to show you this in more depth if you
don't fully understand it yet it's
coming but there are a group of ads
focused around a condition and now what
we're doing is we're drilling down on
that first line and we're modifying it
in a series of versions to determine
which version would help produce the
highest lift John any you would add to
that no okay all right so let's keep
going then here is that key piece free
access to back pain resources and let's
look this would be uh one of our
treatments here is treatment two what's
the big difference here again that first
line that is always in your test design
your value I'll give you another example
just so you understand the difference
between a veriable and a value uh let's
suppose you were testing a headline
color your your variable would be the
headline color your value might be green
blue red in this case our variable is
the description and our value now is
compare available treatments let's look
again this is number three T3 treatment
3 focused on causes and solutions it
says learn about the causes and
solutions there again is the new value
and uh let's go on and look at treatment
four here it is how to recognize the
symptoms now you'll notice everything
else is the same we're just changing
this description this variable with this
series of values if you to look at all
these side by side here's what you'll
see for ads almost identical except for
that key change in that description free
access to back pain resources uh compare
available treatments how to recognize
the symptoms learn about the causes and
solutions now as you think about that
ask yourself in the audience which one
of these four do you think is going to
help us achieve our gain and I I want
you just to look at the four and vote
I'm going to walk over here John you can
see over my shoulder you're your votes
are coming in right now number three
says Grant number two says Karen number
three says
William number two says John number
three says Jake one says theia one says
Josh one says Tim three says kada two
says David one says Brian your lists are
coming in we're we're talling and
watching these votes as they come it's
very interesting keep voting tell us
what you think tell us what you think so
again these are the four
differences give us your vote 4 2 3 3 2
1 3 2 1 3 3 1 all right now as you think
about that let's just move right to the
results and think even more deeply here
then is the result of the experiment
you'll notice uh number three was
submitted by Amy and it scored very well
it certainly beat back pain
resources treatment options but it did
not beat the final version which was
symptoms see in bold on the symptoms
piece there's a 3% gain there's a
99.7%
increase uh or relative difference it's
a dramatic gain John any comments on
that in terms of background the just the
one thing that we noticed was that of
the two ones that did validate for a
positive difference yes the one thing
that they had in common was that it
seemed like the visitor was trying to
understand why they weren't just
interested in a solution they were
looking for answers to question what
kind of why why the solution would work
is that what you're saying or or or what
can I do where do I start why am I
experiencing this they're wanting to
understand okay okay all right so this
is uh marketers combining Art and
Science I just wrote a piece about this
from my book on how the two come
together they've done the science and
now they're in a bit of the art uh don't
be surprised that despite our emphasis
on science we're constantly stressing
that marketing also involves a bit of
mystery a bit of art the marketer has to
use art to bridge the difference between
what they know from their science and
what they need to know it's art that
helps you transcend the Gap it's
intuition it's perception it's that sort
of right brained activity to
oversimplify the processes of the right
brain all right let's keep going because
there's something deeper here we're
going to show you this and remember that
as we show it to you what we're trying
to achieve is uh is a learning for you a
set of principles that will help you
take an experiment like this and use it
to dig deeper into the mindset of the
person that you are trying to Market to
so back to this particular question we
began with this question and now let's
consider this experiment across three
critical principes these principles will
Define everything we do today and uh and
we're going to take each of these
three and uh and and and drill down
deeply to help answer the primary
question that we raised at the beginning
of the clinic so here's the first
principle all you need to know is enough
to predict let me say that with a little
bit more context customer research can
generate an overwhelming amount of data
but we can cut through this with a
simple maximum all we need to know is
enough to predict customer Behavior now
I want to I want to give you some peace
of mind marketer because if you are like
most you have this vague sense of uh for
boing this insecurity that you're not
capturing enough metrics data that you
don't know your customers enough that
you haven't done your persona research
and to some extent you haven't been able
to and it's not your fault but frankly
this this undone research hangs over our
head like the sort of democles and we
don't know how to escape it because no
matter how much we do it doesn't seem to
be nearly as much as we could do no
matter how much we know it doesn't seem
to be nearly as much as we could know
the problem with this is it will leave
you feeling as if you can never know
enough and it's frustrating it takes the
fun out of sitting down to think through
a campaign I want to suggest you that
the goal of a customer Theory or this
whole business of of doing customer
research should be disciplined around a
simple point and that is all you need to
know is enough to predict customer
Behavior now think about that for just a
moment and let's show you how that
connects with the case study that we
just talked about here and to do that um
we're going to go back and uh as we do I
think it's important that you understand
what we mean when we say customer Theory
the customer Theory because this is a
word used in our laboratory and by the
way we think and I would say this to
anybody on this phone that has a direct
connection with a SE Suite if you're a
CMO or even a CEO and sometimes on these
web clinics we have senior leaders from
some of the largest companies in the
world customer theory is at the heart of
your competitive Advantage if you
understand your customer better than
your competitors do you will be able to
introduce better products with higher
margin and with more Market traction
than they you'll find the greatest
companies in the world have a profound
understanding of their customer that
their competitors often cannot
match customer theory is an
understanding of the customer that
enables us to predict their total
response to a given offer now that's
what we're really trying to do is
understand enough I don't care about
where they live what their income
is what their uh religious preference is
if it doesn't enable me to predict their
behavior if that serves a way for
instance if that helps me isolate a
customer segment that I know has a
certain level of motivation and it's
useful information it's good but if not
I have a problem and so so I want to
know only what I need to know to predict
and keeping that cor of the background
let's go back to these four ads because
John had a critical issue here and that
is how do we take this learning and
translate it into customer wisdom John
you want to share anything about what
was going through your mind at this
point when you got back those early
experiment results well I wanted to I
wanted to beat what they've had before
this experiment I wanted to try and take
those learnings and apply them and I
didn't quite understand how um other
than to start exploring the question of
why what is it that they're looking for
at this stage of the pro process in the
micro yes process specifically because
it is fundamentally different than
another segment of their customers which
are looking for Solutions at a different
stage of the process based on different
keywords so I'm really looking to
understand what is it that's going to
motivate this group to get into the site
to find valuable information and then to
sign up so so you're looking almost for
the psychological trigger yes and and so
John now has to take that first
experiment and say how can we isolate
the psychological trigger that will help
us move them move them towards the
purchase down the micro path so we
designed an experiment and uh this
experiment is now taking into
consideration the previous experiment
it's reflected in a document that we
developed it's an actual tool called a
test protocol and you'll notice on your
version of it there's a
hypothesis followed up by a
question and you'll see in the research
side there is a primary question and
there is a secondary question the
hypothesis the question then leads to
the very variables that we're going to
isolate and test this is one tab in fact
if you look at the bottom uh you'll see
this is a live this is a live protocol
we just pulled up I want you to get a
sense of this we submitted this document
in 2006 for patent it was part of a much
more complex series of uh heris and an
approach to this problem it was just
granted last month and uh we're we're
we're pleased to see that but I want you
to know that so that you recognize
there's a lot of science here and on the
front side is a sort of uh planning
document you're looking at this planning
document this was used to translate the
previous test learning into customer
wisdom a new test designed like that and
and underneath that look at the rest of
the tabs at the bottom there's partner
information screenshots take me to the
screenshot tab someone if you can do
that we'll just move it to screenshots
or is that something I need to do there
you go these are all of the various ad
groups we're going to look at them in
just a moment so that you can compare
them then there's optimization
background information let's just slide
over and look at
that and uh this is all kinds of data
from optimization background then let's
look at uh the validity tab just a quick
uh validation calculation piece here
this is built into the documents design
by the way you can learn to use these uh
we have a online testing course no I
don't have a slide selling the online
testing course I'm just telling you that
there is a simple 4 to six hour course
you can take that teaches you how to
design a test properly it's part of our
research here it is not a major re
driver but it's something that our
audience needs and so we produced it for
them and we got an updated version
coming out very soon uh all I'd like you
to know is that you really move from the
planning piece over to the results piece
show me the summary which will be your
last tab there that's it and you can see
in this tab if you look at it carefully
this is now taking those original
questions and scroll down for me and uh
hit yeah make it larger you can do that
with a little arrow on the right hand
side there we go there we go all right
now scroll down here's the actual
results and all the things that we
discovered all the ways we protected
against validity errors uh listen so
many people are testing on the internet
and making decisions based on bad data
sets you've got to learn the
fundamentals of Designing a test just
want to leave it right here for a moment
in fact I think we'll go back to the
deck I want you to understand that this
document was help controlling the
process if you're on this call you say
wait a second I'm lost I'll never be
able to do all that I don't have a test
protocol I I want you to know that you
don't have to have this document or
specialized understanding to learn from
this Clinic something you can do right
now that you can take back and use to
translate data into wisdom to change
metrics from from from how much to why
so as I as I share that with you I want
to shift now from the protocol to the
actual test share it with you these are
the control and treatments that were
designed these are separate groups and
I'd like to break down how they helped
us isolate a learning John's going to
help me with that if you've just joined
this Clinic if this is the first Mech
laabs Clinic it will be dense with
content but we don't want to overwhelm
you we don't want to impress you with
how um our academic credentials the
history of the lab or how clever one of
our analysts might be the goal here is
to serve you it's not just about
marketing it's about the marketer
teaching you in a way that will Empower
you to go back and get a result so so
bear with this if this sounds a little
bit complex it's going to get simple and
actionable before we're done John uh
that's a quick overview I'm just going
to move through the groups real quick
okay any any context you want to give me
first before I do um no the only other
thing that we wanted to do and we stress
to the research team is to make it as
Fair as of test as possible really
isolate the variables so that we could
get a very clean result okay all right
so um we're moving and you can see in
the screen behind me that there's a
hypothesis uh for number two experiment
number two now this is control this
isn't treatment to this is the control
and this is the general hypothesis John
can you just summarize the hypothesis in
a in a in a heartbeat for us uh what we
wanted to discover and prove or disprove
was whether uh the symptom Focus would
outperform the control ads if that was
the trigger that we were um that's going
to get the customer and you said symptom
Focus so let me just drill down we're
saying there's a lot of ways to talk
about this treatment but if we focused
on symptoms would it produce a higher
lift is that what you're saying that's
what I'm saying all right so I'm going
to move on and uh I'm going to go to our
first attempt here's a treatment how is
how is treatment one trying to achieve
that John so if treatment one wins we
will learn that the symptom content
approach is most effective when used
only in the headline so headline if we
just put it in the headline will it
produce lift all right so in the
headline only do we get a lift that's
simple let's go to treatment two explain
treatment two for us um if it's only in
description will we get a lift not
headline so description will be the
first line underneath the headline is
that what we're talking about so now
we're trying to see a symptom Focus now
just to make the connection we realized
in the first round of experiments that
it was the symptom Focus that seemed to
produce the gain right so now we're
trying to extend that in our
understanding and that will tell us
something about how the customer is
thinking this is a hypothesis not about
ads but about what matters the most to
the customer with this stage in the
thought sequence so now we've tried it
in the headline now we're trying it in
the description line and let's move on
treat three this is for all of these ads
in this group what's the key here John
uh it's both headline and description
it's probably the most holistic full uh
approach to putting symptoms forward all
right so now we have it combined in the
headline and in the description so this
is a hybrid of the last two correct
correct so let's put it all together so
we have the control we've tried to
improve that with treatment one
treatment two and treatment three John
has already explained to you the
differences between these treatments I
think it's time for results but I really
want to hear what the audience thinks
which one wins one two or three tell me
quickly
audience Here Comes their
notes I'm expecting this and I think the
audience is going to get it right by the
way the general vote the first time
around was wrong 74% of the time when we
do marketer intuition and test it it's
incorrect but I think most of you are
getting it I think you're believing that
it's probably going to be the hybrid I
see a lot of Threes Melvin I think
Melvin's at a two and I see Robert
courageously holding on to one Robert
congratulations there are no trick
questions here may be a one um we're
just going to share with you what the
data says but but uh uh let's go on
Wayne Kesler says a three I see Cheryl
is a three Eric is a three uh and uh Tim
has a question mark Tim I don't know
what that means a question mark I don't
know how to I don't know how to rate
that on our on our scale uh and uh a lot
more threes coming in and uh and grant
grant what is ha mean ha ha apparently
he knows something we don't know all
right all right so so keeping that in
mind they're all laughing in the studio
we're we're we're looking at these and
we're going straight to results here
they
are so look at the
difference look at the
difference it was the
combination it was the combination T3
which I think most of you expected
changing it at the headline level and
the description level that produced a
further gain of
40.6% right what are your thoughts on
that John it's like I say to a lot of
Partners sometimes you can't set
separate um coffee you can't separate it
from the cream and sugar it has to be
combined to get the full effect that's
what we really wanted to understand here
and that kind of proves the point that
in order for them to truly get the the
symptoms approach to see it and to
connect with it and to click they needed
to see it both in the headline and that
first line in the description excellent
John thank you very much appreciate it
apprciate uh John is a rising star in
our group and I remember when he started
I I I'll tell him he started in Mech
laabs years ago as my assistant and uh
now a senior leader and and we're very
I'm very proud of him uh let me keep
going and and and I'm going to now take
this to three questions so stop for a
second say to yourself okay I don't have
all the perfect data I've got some paid
search ads I have some test results but
what's the general track I can follow as
an ordinary marketer to get more wisdom
so I can predict Behavior better so I
can achieve my Gap eyes so I can get
promoted so one day I can run the
company I can fire the mean salesperson
who's been abusing me and set this
organization right sort of that cause
and effect chain thinking about it that
way let me show you the three questions
that we think drive this kind of wisdom
the first is a research question it's
which which uh headline which approach
whatever that value is that you're
changing in the variable that's your
which question and a good research
question always begins that way but once
you get a result in you have to do what
John did you have to say but why and
when John asked that question John
noticed that the ads that were winning
were those that were focused on symptoms
this is in the first experiment which
led to a deeper question what does that
tell me about my customer and the answer
to that informs your customer Theory
what you can know in that way about your
customer can help you achieve a more
robust understanding gain that
competitive Advantage predict the
customer's behavior and so John then
postulated a new test that would confirm
what he was suspecting and he would go
right back then with the witch which
approach headline only description only
headline and description and that gave
him a new set of wise and he tried to
tell you about coffee cream and sugar to
explain what he's thinking which makes
him think about how do you penetrate
into the mind of that person with high
impact which brings him back to the what
question all of those connect and they
increase the wisdom that we're gaining
we worked with the ROI Revolution people
on this I appreciate Tim and his group
for being good partners helping us
implement this across their campaigns
and it had a major impact on this
organization's paid search so well done
for everybody that was involved in the
project now let's keep cashing it in
follow with me I want to keep teaching I
have two more principles and they get
easier to understand and easier to apply
as they go by the way before we're done
I'm going to show you what appears to be
a colossal failure on the part of our
own team an embarrassing set of
experiments uh those are coming uh
towards the end for now I'm going to
move you to this next piece to
illustrate my point you you see the
which why and what here it is an action
by the way that just kind of rounds out
what I'm trying to say cuz driving that
is this more robust customer Theory
which helps you refine this value
proposition in fact I just want to pause
there because I frankly that animated
after I got done saying everything I
could think of saying and now that I see
it on the screen it's rather clever and
uh and so I I want I want to bring it
back to you the the the which the why
and the what inform this whole business
of uh customer knowledge customer Theory
and the increased knowledge about the
customer the customer Theory translates
into a more refined value proposition
you may be in uh right now with your
team struggling to get the value
proposition right for a product
struggling to get it right for a whole
organization this is where the testing
process comes in the internet is more
than another Channel we'll talk about
that in another Point coming up but as
you keep moving through this series
you're able to fill in those gaps and
thus improve the articul
of your value proposition which brings
me to an example here's a group they had
a value proposition it was poorly
communicated they worked very hard on
solving it they came to classes they
came to training and finally we ran some
tests to help them solve the issue I
want you to see how the first treatment
we're not even going to look we're not
even going to look at what we did but
the first treatment produces a 200% gain
and you'd think you're done but no that
leads to that that cycle of which why
what so a new cycle goes we want to move
this now to templates without hurting
the actual performance cuz we've got to
get the content on a template and we
wonder what that customer is going to do
how they're going to react to that
particular presentation can we move it
without losing them we move it carefully
we get a minor gain we don't go
backwards we set a Best Line we go
through the same cycle again which why
what that drives another 30% Improvement
it's compounding it's actually hitting
the p&l now in a significant way and
that Cy sets us up at some point
remember the illustration from the
previous slide that that silhouette of
the customer and the blue line Rising as
we gained more knowledge at some point
we hit critical mass and that critical
mass translates to what you see here
which is gains all across the Enterprise
gains that are being captured from this
this kind of new understanding about
what's happening in the mind of the
customer that is what I'm trying to say
in simple form ground an experimentation
we've been learning this from 10,000
plus tests across landing pages and
paths like this this brings me to the
second
principle the internet is the laboratory
I want to explain that in more depth we
said earlier that customer
research can generate an overwhelming
amount of data all we need to know is
enough to predict customer
Behavior we'll say now that the internet
is not just another marketing channel it
is the primary laboratory for conducting
customer research this is what got me
involved years ago let me check my time
I've got to be careful because I could
wax eloquent on this point and and and
get us behind I was a philosopher trying
to understand value exchange and we're
releasing our catalog right now of
research over these past many years and
I remember thinking if the nodes
increase and the bandwidth increases
between the nodes we're going to see
human activity in real time measurable
like never before in the history of the
world but at that point I couldn't find
anyone excited about it I tried to get
professors at Cambridge to do research I
tried a lot of things we finally founded
this laboratory That Grew into what it
is today and at the heart of it is a
simple question why do people say yes
why do people say yes why do people say
yes and underneath that has been this
capacity to use the internet to cut
through philosophy do
science and answer the question now I I
I I don't want to stop doing philosophy
philosophy makes you go back and
question your warrant
it makes you question the operating
system that your software is working
within when I say your software I mean
our mental software our our our
Linguistics and all those ways we
approach marketing but science has
allowed us to measure to peer into human
behavior in real time and that's what
the internet can do for you so we need
to know how to do that how to approach
that de car this was lifted from one of
my books Dear and I don't know why it
says Source unknown because you didn't
have time did you I didn't have time you
need to put the source I'll give you the
source uh I actually had them fling this
on the slide just a few minutes before I
got up to teach I just mentioned it to
to Paul to know what people really think
pay regard to what they do rather than
what they say now that's a translation
from the French we'll get the original
Source make sure that's in the deck when
we go live the point I want to make for
you is way back then de car saw into
human nature and had an Insight that is
still profound and still relevant today
if you want to get inside of your
customer's mind don't rely on focus
groups or opinion research are focus
groups wrong no they're not are they
absolutely useless no they are not but
they are used so many time to justify
what an executive already knows they're
used so many times to to to validate a
new launch when in fact they don't
produce the desired result why because
it's not what people say that matters
it's what they
do an alignment between what we say and
what we do is important as marketers but
but the customer often doesn't even know
what they're going to do to a given
offer until it's presented to them in a
real life context and so you're going to
see on this uh
pyramid an approach that grounds the
intuition of a marketer which we think
is valuable I keep stressing to my team
that I've had 15 years of rigorous
emphasis on science inside of this lab
that might make someone think that we
believe the science component is the
critical and only component that should
drive effective marketing I do not think
so I have too many gaps in my knowledge
there are too many questions unanswered
both about the approach to marketing and
about what's happening in the mind of my
Prospect I have to use art I have to use
perception I have to use intuition to
bridge the
gap and by the way so do most scientists
in any Endeavor look at Coon's work on
that and all the subsequent argument
since the point I want to make for you
is that marketer's intuition is good but
marketer's intuition needs to then be
followed up with the process and their
opinion research can narrow down
Alternatives and provide you with
certain guidance that helps you design
proper experiments but in the end it's a
behavioral test for those of you that
say Well it can't be a behavioral test
um focus groups ought to be enough it's
always a behavioral test because whether
you call it a test or not at some point
you got to take it to Market and try to
sell it and whether it sells or not how
often how well is the ultimate
behavioral test I just prefer to fail on
purpose than by accident so I like the
first phase of that to be a test phase
as opposed to a make or break promise to
the management team so keep this pyramid
in mind and let's let's think about it a
little bit more here is the traditional
approach now if you're on the phone
right now you might be saying uh amen uh
to coin of phrase uh if you're in in
midlevel marketing and you can't get the
SE Suite to give you the permission to
do what you want to do and I'm sorry if
that's all you can do but for those of
you on the phone with the power to make
a change let me show you that this
traditional model is broken what you
often see is the business leader creates
the problem product it's pushed out
through the agency there's a briefing
that's done there's creative that's
prepared there's a media buy and it's
implemented across the marketing
organization and at some point in that
process down the road it's pushed over
to the online group often we see a
fracture the online group is not even
part of the offline group they're
separate but however you do it it comes
to to the to the online group too late
if you really want to use and I have
seen this this is not just the theory I
could point to major companies around
the world where we have done this and
I've seen a dramatic impact on their
results we preest the value proposition
but we do that by getting the internet
piece here right at the beginning when
Business Leaders are creating a product
they need to test the the
validity of their thinking using the
online approach to refine both the offer
in the form of the value proposition and
the messaging all around the offer
before it reaches the agency the agency
should be given guidelines
now if you're on the phone in your
agency and there are many agencies on
the line I'm not trying to take this
away from you in fact you may have to do
this at the beginning of your process to
enable your company to function properly
in some cases if the agencies would do
this first it would give you a major
advantage over your competitors number
one and number two it would solve a
problem inside of these major companies
who can't get the Business Leaders to do
it in cooperation with the online group
and in cooperation with the agency all I
would suggest for you is that you need
to view the internet as a lab and you
need to involve This research on the
front side and then follow up with your
creative and the implementation of your
marketing into all your other online and
offline channels the model you're seeing
on the screen right now this particular
model has yielded dramatic results all
over the
world so keeping that in mind um I want
to uh take you to one more Point quickly
and then we're going to get to the third
principle uh you can see this which I'll
just shift Paul chenny who helped
develop this put this slide together
showing you actual examples from our own
files and showing you how these lower
investment channels like the internet
can help you craft your messaging before
you get to the higher investment
channels and Paul I've already said it
but it's a very pretty slide so I had to
say it again and put it up there all
right I know that probably spent hours
on that yeah I really appreciate it so
so I'm on to the third critical
point this one is the one that I think
you'll find counterintuitive if uh your
time is limited I think this will be the
point that you might uh find your most
interest and actually quite practical
not to mention the fact that I'm going
to show you some embarrassing test
results so here is the third principle
the goal of a test is not simply to get
a lift but rather to get a learning and
even a negative test can translate into
positive results let's talk about that
here's a test protocol 1092 from our
library the company's a publisher of
electronic marketing information they're
experts in marketing they're selling
their expertise this is not us it's a
British company and the goal was to
increase registrations for free letter
they had a thoroughly optimized page it
was their highest producing page it is
on the surface ugly it did work they
came to us and said can you help us here
is the page now you can see the page is
simple you can see there's nothing
particularly uh attractive in fact if
you're an agency You' probably get fired
if you produced a page like this but it
produced significant revenue for them
and then we began a series of tests I'd
like to show you two treatments here's
treatment one it involves a lot of
things that we teach and by the way
you'll see the gray on here the gray is
because we're uh we're attempting to to
essentially anonymize uh the
organization but you'll see testimonials
Major Brands third party indicators a
different headline you'll see on
treatment two in fact let me just look
at some of the breakdowns these are all
things you've heard us talk about in
clinics clear headline emphasis featured
clients bolded key terms form films
reduced I mean friction an all the
things we talk about the conversion
sequence have been implemented on this
page it is uh a good piece of science
let's look at treatment
two again another approach headline is
even stronger added more testimonials
rearranged page to give a better sight
flow added a personal feel with images
and a handwritten signature uh we
improved uh the request with tell me
where to send look at these two pages in
fact look at them side by side right now
next to the control and as you look at
them ask yourself a
question which one of these is the
highest performer and why so I'd like
you to vote audience tell me here we
go I'm watching I'm watching I'm
watching some of you some of you uh
there we go two three three I saw
several ones one 2 2
three interesting interesting
interesting um there are
our uh guy you say C I assume that means
three uh hammed says three Lisa says two
Sally says one John says three Tom says
one all right looking at these three
side by side let's look at the actual
results what you'll notice is the
conversion rate for the control is
14.26% and treatment one it's
6.74% we managed to grow conversion by a
53% with our fancy treatment by the way
I don't think our treatment is very
attractive it's a test but it's
isolating variables and it's improving
many of them let's look at treatment two
treatment 2 is in front of you and
you'll see that we proudly improved our
performance moving from a - 53% to a
52% now in case it hasn't hit you yet
that's not a positive thing on the
surface it's negative
conversion rates went down way down now
I want to I want to I want I want to
challenge you for a moment because the
conversion weight went way down but I
would challenge you with this thought
the pages one and two are actually
better they're better
pages but they're not better performing
pages in the end it's not always the
better page that
wins it's the better me message and the
message is synchronized to the thought
sequence of the individual who is
interacting with your
creative we optimized the web page but
we failed to optimize the thought
sequence and that's because we didn't
have enough customer data yet now by the
way these were what I would call Planned
failures and in fact these tests were
not failing tests they were valid number
one and number two they produced
important learnings I really I really
want you to stop and think differently
about the testing process the goal of
these tests was not to get a lift it was
to get a learning and these tests helped
us understand something very important
about the thought sequence of the
individuals interacting with this
page so I share that with you because I
want you to start to think differently
about your whole testing process if you
get a maximum lift
it's going to be not simply because you
got lucky and ran a few chaotic test
designs and scored but rather because
you gained a significant or sufficient
understanding of the customer's thinking
process the enemy of Excellence is
adequacy and often what's costing us the
most is our successful test our
successful tests are costing us because
we think we've optimized the page after
all we got a 20%
left but listen to me if you proudly
announced to the management team that
you got a 20% lift on that page and then
because it's successful you move on come
back a year later and test it again how
much money have you lost if you should
have and could have got a 60% increase
on that page and now for the next 12
months you're leaking the
differential that's going on everywhere
it's going on everywhere and and so we
need to be careful that we're not
thinking about lifts we're thinking
about learnings and then We're
translating learnings into the greatest
possible lift that's what's going on
here we're learning something very
important about the customer's
process here's the
control and and what you need to
understand and what we suspected as we
started studying our data sets and
planning the test is that the customer
was in a different mindset when they
landed on this page and by going back
studying what was happening in the
channel we knew that they came here to
get instant access and they were all
already
sold if they are already sold if they're
that mature in the thought process then
everything we're doing to resell them is
only giving them more reasons to think
about why they shouldn't make the
purchase or shouldn't at least make this
next micro yes in the
unfolding uh journey of micro
yeses so the difference here is that we
have to match the page to the thought
sequence and instead of applying the you
might say Mech laabs researched best
practices to fix a
page what we were really trying to do is
use these tests to eliminate options to
understand how mature the thought
process was so we could create the page
that would appeal to them at just the
right point in the sequence of micro sis
and in doing so we discovered that the
best way to improve conversion was
absolutely counterintuitive
CU look uh and this isn't on my slides
but if you're taking notes you may want
to write this down there's only three
ways to fix a page this is
philosophy and I'm going to give you a
imminently simple translation of
philosophical principles they go all the
way back to Aristotle um but we'll keep
them very simple all that you can do is
is change something well let's put it
this way you can change something you
can add something or you can take
something away let's just think about a
page you can change the headline you can
add more content you can take more
content
away uh but you can't really do anything
else to that page it might be the same
with the graphics but including all the
elements you got three choices and
mostly what we do is change an ad change
an add change an ad all too infrequently
do we take away often taking away
produces a higher lift and in this case
here's what we did we reduced it
dramatically
now remember this was for a marketing
company they were already experts in
marketing they' already done all kinds
of optimization and they had produced a
a page that gave a significant yield
it's a very profitable group but now we
did something that was probably
completely unexpected we took their best
work we cut it in
half so look again at the control and at
the
treatment what's the difference between
these two
this a 78% increase I'm going to go back
I'm going to take the slides back
the treatment on the right produ 78%
more than the treatment on the left
What's Happening Here what's happening
is that we are dramatically reducing
dramatically decreasing the amount of
content so as to match where they're at
in the thought sequence which is I just
want to do this how do I do it as simple
as possible the text above the Box
required them to engage their mind and
make more meaning once through at the
point that you have Micro yes you want
to do nothing that slows down the
velocity of the cell the velocity of the
cell is a key to achieving maximum
conversion so we intensified we
increased velocity by reducing and in
doing so they got a lift what's the
summation of all this well it's it's
really this in most companies that test
sequence would appear to be something
like this we tried we failed we tried we
failed Eureka
we got it that's not the real way to
view this test we tried we succeeded we
tried we succeeded we tried and we
succeeded yet
again the gain comes and is built on the
other two experiments and you've got to
think about testing in that way and may
I say you've got to create those kinds
of expectations in your management in
your senior leadership so they
understand the testing process all right
I want to stop there for just a moment I
want to Mize these three cre principles
and I'm going to let you ask me some
questions I need you to help me Paul and
my team Stand By and I'm going to try to
answer some of the questions from the
audience here's the first principle
again customer research can generate an
overwhelming amount of data but all you
need to know is enough to predict
customer behavior number two the
internet is not just another marketing
channel it is the primary laboratory for
conducting customer research and number
three the goal of a test is not simply
to get a lift but rather to get a
learning even a negative FFF can trans
into a positive result uh while your
questions are coming in let me see here
did site visitors know about the offer
from TV and other media said guy the
answer is yes yes and then Derek says
check out the current page Dereck that's
a funny comment I haven't looked at that
page in years I'd like to check out that
page and we shall uh make sure that we
do that in our group John make a note to
Austin I'd like to see what it looks
like now um and uh and someone says how
do we formulate this is Daniel the
proper research question that leads to a
testable hypothesis resulting in a
learning now that's a great question it
deserves its own clinic at the very
least uh uh I'd like you to to fire that
question to one of my team members we
might talk about it in our blog Paul in
more depth well we've written I've
written an article on John come on up
here uh you might want to answer that
John just mentioned he's he's in the uh
Studio over at the seats and he just
walked up the front John has written an
article on that piece you want to give
us a brief summary John right I
basically wrote an article on the blog
on how to design a test in order to get
a learning um and a lift at the same
time and I've actually written two
articles I've written one that addresses
radical redesign so the more complex and
then the simpler version which was
published probably about a year ago or
so that goes through the entire process
of where do I start and then how do I
get to the point of test and right now
I'm actually working on one that uh an
article that is about test sequence okay
so um the broader version of that but
it's on there online just uh marketing
experence blog all right I want to
actually do more than that um I want to
get a link posted so that anybody can
get those audience would you like to get
access to those links and I I I I'm
seeing through the glass window of the
studio on the other side I don't know if
Daniel's over there uh Daniel if you've
got comments feel free to interrupt help
me know how to get them to that link
okay he said yeah okay so GNA tweet it
he's going to Tweet the link all right
so hash webcl Clinic we'll tweet the
links to you so that you can get to that
those articles that'll try to help
answer that question now Paul I need you
to be picking out questions for me um
because they're coming in so fast and
there's so many comments I I can't I
can't see them so let me make sure that
Daniel and the team are doing that
here's one right now okay I see it SEO
versus value prop if you're trying to
get 5% word density to rank well it's
not easy to write value prop statements
okay look um this is a classic we've
written on this as well and we've taught
on this how to balance the SEO demands
with optimization demands but one of the
pillars of conversion is the wrong
emphasis or misunderstanding of how to
use SEO you can and we can demonstrate
study after study particularly when it
comes to value proposition it's much
easier to articulate your value
proposition than it is say to play with
the overall design without impacting SEO
we have principles we're not saying it's
easy to balance the two but be aware
that you don't let yourself be held
hostage by SEO principles because it
doesn't matter if 10,000 visit the page
and your conversion rate is so
substandard that you're not gaining the
lift that you could now I'm not trying
to speak poorly of SEO because we do it
all the time and John has a comment he
wants to make on that I'm going to let
John comment on that but I'd like to
point out for you also that um that even
though you may have a challenge when it
comes to articulating the value
proposition that typically will help
your SEO not hurt it John go ahead just
to piggyback on what you just said that
one of the tests that we've talked about
probably hundred times in these clinics
that I ran with one of our partners when
we did the test um managing the thought
sequence and the value prop we actually
gained rank in natural search rankings
and increased the traffic so we went to
go test against that the SEO guy was
like whoa ho ho wait a minute don't want
you to start taking things away or
changing because you got me this nice
fat gain in in SEO because we well wrote
value prop statements yeah value prop
will often help your SEO all right um
and keep highlighting questions Paul for
me and and also Daniel you might want to
tweet uh a link also to the clinic on
balancing SEO and paid search uh SEO and
uh optimization that very problem we
addressed in a fulllength clinic where I
was teaching just like now I have a
question highlighted here from Scott
small it says do you think it's possible
to have one page be both an organic
landing page and a paid landing page I
think it's possible but why would you
ever do it right I mean if it's you've
got control of the URL then you might as
well send that URL to a distinct page
made precisely for paid search and you
do when it's coming out of Google or
somewhere so I would if at all possible
create separate landing pages from my
paid search any other comment on that
John no we did just that with a research
partner as well and we're able to see
gains excellent excellent very good all
right so Paul here's another question
this is from Shan Sean says can you talk
about application of customer Theory to
engaging charitable donations rather
than purchasing a product or service
actually Sean it's a great question I
had a conversation today with Tim cerak
who's an expert on this topic and he was
presenting a deck it was 112 slides long
showing one of the largest provider of
services to charitable companies more
than a thousand charitable organizations
how applying these principles can
improve two key metrics the first one
was um the first one was the number of
donations and the second was the amount
of the average donation Tim is not an
employee of Mech laabs he's a student
who's been at our Summits and Clinics
and been very careful to apply the
principles built a methodology around
that just for uh nonprofits now we work
with them as well but here's the point
you're still selling and you're still uh
facing some very significant challenges
like trust and credibility and there is
a a value proposition for every
nonprofit in fact yours is particularly
competitive and it might come to might
be boiled down to the answer to this
question if if I'm the ideal owner why
would I give my money to you rather than
to any of these other causes if you
can't answer that question you're going
to be struggling and the only way you'll
raise money is through the personal
influence and personal relationships
what you need is something more
compelling than that to get major
traction and we've done research with
World Vision and some of the largest
charities in the world and uh there's a
lot more we might say about that here's
what I'd like if some of you are
interested I don't know how many of our
listeners are interested in how to apply
these to nonprofits but send us some uh
an email or tweet that question to us
and we'll determine whether or not we
might want to have a special Clinic
sometime designed just to help the
nonprofit sector um we're running we're
going to take uh a couple more questions
quickly and um oh I need to be Paul just
reminded me to take you to this slide
and I'm going to keep asking questions
but really uh Tim is going to be at the
optimization Summit I just mentioned
he's going to be speaking on what he did
in that work again he's he's coming as a
he's not promoting mlabs he's coming as
a student talking and as a marketer
himself talking about what he's
discovered applying these principles in
that in that context I don't know if you
know about optimization Summit um You
probably do by now we held the first of
the we have many Summits around the
country for email and many things but we
heard the first optimization Summit last
year we sold out in fact we had double
the projections of our original
attendance we're going it to do it this
year in Denver
now you know something I can stand up
here follow a script I have none I can
stand up here and engage hype and cells
language I will not but I I am not
afraid to say with passion that this is
an a remarkable event all we're doing in
this Summit is revealing case studies
experiments there are no vendors selling
from the platform uh what you have
instead is either one of our analysts or
scientists sharing with you what they've
discovered or you have an actual
marketing practitioner who lives every
day just like you do uh By The Sword you
might say trying to achieve results
sharing both their failures and their
successes we had an overwhelming
response last year and I would encourage
you to get registered soon I don't want
to create a false sense of urgency but I
promise you we will sell out well we
we've got a special running and it's
going to be over in like three days okay
so Paul just said from the floor that we
also have this uh special running for
four more days or three more days in
terms of cost I I'll leave it at that
I'd like to answer another question but
but I would like you to come I would
like to meet the audience in person and
I'll certainly be there speaking and uh
the day before the day after and at the
event with that in mind I'm going to cut
back over here to one more question and
uh it says sales calls value prop the
positioning statement I don't understand
that sell calls value I don't understand
that how the positioning statement is
articulated depends on who it is for
same application the results refin the
Persona understanding I think I
understand your question let me just
touch it in brief when I I teach on
value proposition I'm summarizing um all
of our research for the last nearly 20
years on this and here's what I would
say there is a primary value proposition
that's the organizations but it has
derivatives it's interpreted into a
value prop for your for your prospect
types your product types and even the
processes like form buttons these all
have derivative value propositions
connected to the primary value
proposition we call them the four Ps
primary
product Prospect and process I'm getting
ready to do a major lecture teaching how
all of these come together in fact
that's the opening lecture at the
optimization Summit by the way where I
talk about how to combine these how to
create these and how to use them to
dramatically intensify the force of all
of your marketing efforts all right I'm
going to stop but I want to thank John
for being with us today it's good to see
you here John when you stepped through
my office those many years ago you
didn't know that you'd be standing in a
studio with me teaching kidding H uh he
a long ways I uh my office is a
classroom I have uh six seats behind me
and all of them are men like John that
are rising up in their abilities
participating everything we're doing I'm
talking about my personal office is
right in the middle of the building and
uh John um uh congratulations for all of
your growth in progress I want to thank
the whole audience for coming today we
don't charge for these clinics there's
more than $15 million with the free
research at marketing experiment.com we
just ask one thing tell somebody about
the clinic tell somebody about the
research uh that's how we have grown own
and that's how we're building this
community of marketers that are focused
on evidence that want to discover what
truly works thank you again we'll be
back in about two weeks with more
research for you
[Music]
oh