Thank you.
Good afternoon.
So if a typical website was a digestive system,
it would have no capacity to poop and that's painful.
In fact, digital suffers from a lot
of ailments and [INAUDIBLE].
Overproduction is one of them.
Sometimes I say, if you give a website
to a typical organization, it's like giving a pub
to an alcoholic.
It's happy days.
It's publish, publish, publish, publish
and then, a couple of years later, it's the AA meeting.
And it's, hi, I'm Gerry.
I have a 10,000 page website.
I used to have a 1,000 page website,
but I had no self-control.
It's just one more page, one more page.
And how do we deal with that problem?
And I think part of it is down to a core cause that
websites-- most websites and apps are designed by dogs.
And there's a reason a dog is a man's best
friend, because a dog thinks everything is a good idea.
No matter what time you want to go out for a walk,
the dog thinks that's a great idea.
So it's the dog that designs the website,
but it's the cat that maintains the website.
That cat is not going to review the content, not
going to take the stuff down.
Basically, wants to have life as easily as possible.
So I think we need to bring more cats into the design process.
I'm sure lots of you have dealt with migration over the years.
Moving crap content from an old system into a new system,
it's still crap.
You still bring the crap from one old system
into a new shiny system.
And it reminded me of the great migration in the Serengeti,
where one and a half million wildebeest and a half a million
zebras followed the rains.
And they cross the Mara river as they're heading towards Kenya.
And as they're crossing Mara river on this great migration,
what do they meet?
They meet, basically, the crocodiles.
And I think in any migration, we need lots of crocodiles.
We need content crocodiles.
That as that crap content is trying
to migrate into the new system, there's
these crocodiles going snip, snip, snip, snip
so that only a small percentage of it
actually gets into the new environment.
So we need lots of systems to deal with all this stuff
that nobody is very interested in.
So I want your opinion here.
This is Sims.
It's a very popular virtual game, create your own town.
About five or six years ago it launched a new version.
And it wanted to get people to register for that version.
So it offered some goodies, some nice things
to get if you registered for this new game.
And it tested-- AB testing.
So 50,000 people, it gave half of them
randomly this page, and half of them randomly this page.
And the objective was to increase registration.
Now, one of these pages was 40% more successful
at getting people to register than the other.
So if it was 0.4 of a percent, that
would be significant-- 4% very significant.
But 40% is a huge difference.
So it's the exact same offer, the image is different.
Basically, they reordered the content,
but you don't get anything different.
It's the same basic offer.
So I want you to trust your gut instinct,
because gut instinct is very important
when we come to design.
And I want you to trust and immediately say,
which page was 40% more successful at getting people
to register?
Was it A or B?
How many people say B?
How many people say A?
So that's about 70-30, would you say?
Maybe 80-20, somewhere in that territory.
The interesting thing is I've asked
this question about 200 times, at 200 conferences, all
over the world.
I've been to 35, 40 countries over the last 10 or 15 years.
And 99 out of 100 times, it's the exact same pattern
in the audience whether it's in Taiwan,
whether it's in Reykjavik, whether it's
in Oslo, whether it's in Dublin, whether it's in Brussels.
The exact same pattern, roughly 70% to 80% of people
will go for B. It really doesn't matter.
And actually, on this excellent website,
which has recently changed their name,
I forget what they're called recently,
but which test won is where I got this original data from.
80% of the people who came to that website
voted for B as well.
So before they could see the results, that they had to vote.
And these are marketing people, designers, developers,
programmers, right?
But you know what I'm going to tell you?
A was 40% more successful.
A was 40% more successful.
Most of our time, I've watched this over the years,
our gut instinct is incredibly dangerous.
We should not trust our gut instinct
when it comes to digital design.
The worst possible way to design a website
is to have five smart people in a room drinking lattes.
And the longer you leave them, the worse the website becomes.
This isn't randomly wrong.
If it was 50-50 wrong-- but I could give you 50 more
of these.
How do we get it so wrong, so often,
when it comes to our gut instinct?
That has to be worrying.
But of course, the bigger issue has
to be testing observation of what people actually do.
It's not a nice to have if we want to get the design right.
It's absolutely critical.
It's absolutely central if we want to get the design right.
Now I'm going to show you here a method
around identifying what really matters to people.
One of the key things, how do we identify
what is the crucial stuff?
And what is the crappier stuff?
There's top task and there's tiny task.
And this is a method on a survey that-- many things
that work on the web are counter-intuitive.
And they say, the greater the complexity,
the more things become counter-intuitive.
That we don't think things will work and they do work.
And this is a survey method where you bring together
the whole range of things that people might want to do,
the whole task ecosystem or environment.
And that is typically somewhere between 60 and 80 tasks
will define what's important in relation
to your health, what's important in relation to buying a car.
If you're using Microsoft developer technologies,
what are the key things you need to do there?
We've done about 400 or 500 of these
over the last 10 or 15 years.
And there's typically, within a task environment,
about 60 to 80, maybe sometimes 100 things,
that really define the actual environment.
And we assemble these 60 to 80 things,
and we ask people to only choose five from the list.
And then we give them this.
This long and randomize-- and that's about 60 or 70 people.
Now I'm not even going to ask you, would this work?
Because the vast majority of people when they see this say,
no.
It can't.
But actually half a million people
have voted on these surveys over the last 10 years.
Last year we did it for the European Union.
And we got 107,000 people to vote
which was more people than voted in the recent Russian
elections.
So it wasn't too bad of an actual result.
This method-- this specific method
where we, in essence, overload the people
and ask them to choose very quickly what
it is that matters most to them, is a great way of figuring out
a league table of importance.
The reason you have to give them the entire survey in one
go is because you want to look from the most important
to the least important.
Because most people can quickly agree what the top task is,
but then they all start saying their top task.
We have to clearly identify what are the low level tasks.
And when we did that survey with the OECD, about 2,000 people--
the OECD is a large NGO governmental type organization
that compares countries.
And says, oh, this education system in Germany
is doing better than this one in Ireland.
And when we do it, we get a pattern.
The yellow is the first 25% of the vote.
So four tasks dominate the first 25% of the vote.
Another six tasks dominate the next 25%.
And that's your top task environment, the yellow
and the green.
So there's 10 tasks.
And as I said, we've done this 400 to 500 times
in all sorts of environments, with consumers in Brazil,
with doctors in Spain, with engineers in the UK.
There's always a similar type of pattern
that actually comes out.
So [? 1,150 ?] to 75% and the final 25% of the vote
goes to 49 tasks.
So four get as much of the vote as the bottom 49.
The bottom 49 are your tiny tasks.
And the model of top task management
is that you should manage the top tasks first.
And make sure they're performing well, because they've
got the highest demand.
They are the things that the customers care most about.
Top tasks have a permanence about them.
They retain a permanence over time.
This was that John McCain 2008 presidential cycle.
Two dominant tasks when you come to the environment,
because when you come to these type of websites
you're a committed, politically interested type of person.
If you're a casual voter or an undecided,
you'll never go to these websites.
You'll pick it up on CNN, or Fox News,
or wherever you're going to look.
But when you come to these type of websites,
there's dominant tasks in the customers' minds.
And they're either donate or to sign up.
So they make those tasks really easy.
That was the McCain website.
This was Hillary Clinton in 2008, sign up as a supporter.
And perhaps the bottom submit was not
exactly the most appropriate label in the sign up process
because you don't want to be submitting, perhaps,
to Hillary.
This was Barack Obama in 2008.
It's just one task.
Get them to sign up first.
Get them in, get their e-mail address, get their zip code.
This is 2012.
The tasks don't change.
The way you deal with the tasks may change,
but the type of tasks, once you've identified top tasks,
they last over time.
They keep lasting.
And the principal is manage the task,
not the channel, not the environment, not
the specific format.
So this was Barack Obama in 2012.
Again, sign up.
The same top task emerging.
And we come to 2016, it's still the same task.
Join up and donate.
Two key tasks-- Donald Trump, join up and donate.
Bernie Sanders, join up and contribute,
or donate in the process.
Top tasks have a permanence.
The essence-- really identifying the essence of the environment.
The essence of the environment doesn't change over time.
In health, its check symptoms.
There's constant tasks.
In tourism and travel, it's special offers.
Getting here and around-- they retain a permanence over time.
How we deliver them changes, but the essence of the task
doesn't change very much over time.
So manage the task, not the channel, not the format
in the environment.
But naming the task, getting the wording exactly right first,
is as critical as identifying the overall task.
One of the key things the Obama campaign did from the get go
was relentless testing.
And they did this initially with three people in a bedroom
in Washington DC.
They didn't have a big initial operation, but from day one
they relentlessly tested.
So you see here, donate now is the baseline test.
So they tested, please donate.
And they tested buttons like, why donate.
And you see, not signed up baseline is zero.
Why donate, minus 27%.
You ask a stupid question, people leave immediately.
They don't give any money because these
are committed people.
You ask them, why donate?
They say, what a stupid question, I'm out of here.
They lost 28% of people by just asking the wrong question
at the wrong moment in time.
But donate and get a gift had a 15.8% lift over the baseline,
actually test.
Those were people who had not signed up.
But if you signed up, what was more powerful
was please donate because you had already got the gift.
So at different levels of the process,
different uses of the button were actually used.
And if he had already donated, so
if he had given money at least once, contribute worked better.
Contribute got lift.
Now millions of dollars were raised based on these things.
Knowing when to use contribute, versus when to use donate
is crucial.
So it's not enough just to know the top task.
You need to know how to word it, how to name it properly.
The DNA of digital is words.
The DNA of digital is words.
The way the web has been built-- it's been built
on an architecture of words.
So understanding the exact words can
have a huge impact on the behavior
of people and the success within a digital environment.
What really hurts the top task?
It's the tiny tasks.
Because when a tiny task goes to sleep at night,
it dreams of being a top task.
And when it wakes up in the morning,
it's all excited because it thinks it's so important.
And it thinks if only I could get on the home page.
If only I could publish more content, then they'd love me.
So the tiny task-- they're cute, but there's so many of them.
And most web people I meet, they are nibbled to death
by the tiny tasks.
They constantly combat them because they're
so hungry to get published.
When we do this analysis and we go out and get these votes,
invariably, it's the same stuff at the bottom of the list.
It's always the same stuff.
It's always the same organizational crap.
Organizations have a fabulous capacity
to create enormous quantities of stuff
that nobody is remotely interested in.
It never had a value.
It never will have a value.
So Secretary-General's speeches and activities
gets 39 nine votes out of 31,800 cast.
Oh, the poor Secretary-General.
Norwegian Hospitals-- they don't care about your organization
in Norwegian Hospitals.
They don't care about how you're organized.
And Scottish Enterprise-- oh, the organization chart.
It gets five votes out a three and a half thousand.
And all the effort that went into that organization chart.
And Enterprise Ireland annual report, six votes out
of almost 4,000 votes.
Only 6 votes?
Oh, that's terrible.
All the effort went into the annual report.
Look at the annual report of Innovation Norway,
22 votes on the 14,000 votes.
The organizational ego is embedded in the tiny task
on the bottom of the list.
These are the things that nobody is interested in,
except your boss, and your boss's boss.
So how do you deal with these sorts of scenarios?
So in the OECD when we did this analysis,
we got the customer to vote.
And we got the team, or the stakeholders to vote.
And it mapped reasonably well, except for overview
of what the OECD does, which the team
think is four times more important
than the actual customer does.
So the organizational focus-- and I've never
met an organization that said to me-- anywhere in the world,
nobody has ever come up to me in any organization and said,
can you help us become a little bit more organization centric?
Because we're too focused on our customers.
We need to focus more on ourselves.
We don't think about ourselves nearly enough.
We're always thinking about the customer.
Any one help us focus more on ourselves?
That's never happened.
If never will happen.
Organizations are fabulous at focusing inwards on themselves.
When we did Liverpool-- the city of Liverpool,
they did a mapping.
So we did the top task and then we
mapped the content production cycle to the top task.
So the more important the tasks were to the customers,
the less content they were creating for those tasks.
And the tiny tasks was getting all of the production.
So there's an inverse relationship
in most organizations.
The more important it is to the customer,
the less the organization is doing in it.
The less important it is to the customer,
the more the organization is doing.
Because the ego-- the ego of the organization
is so strong and, particularly, with in a senior management
level.
And what does that result in?
It results in enormous quantities of crap
that nobody is anyway remotely interested in.
This is the US Department of Health.
Last year they deleted-- they finally
got around to the AA meeting.
They deleted 150,000 files and nobody noticed.
150,000 files and pages and nobody noticed.
The World Bank PDF generator, which the World Bank
is-- one-third, one-third of their PDFs
are never downloaded.
Never download, never, not even once.
Never.
Microsoft has 14 million pages.
Microsoft.com has 14 million pages.
Four million of which have never been looked at.
That's the population of Ireland,
of pages that nobody has ever looked at.
I mean, it's ridiculous.
Why do we go to such effort to create
such quantities of so much stuff that nobody ever even looks
at once?
Something is broken.
Something is really wrong because the organization talks
about itself constantly and constantly
focused on its own needs.
And these are a key challenge in a digital based environment.
And part of it has to do with the hippo.
You know what hippo stands for?
The highest paid person in the room.
More people are killed in Africa by hippos
than are killed by lions.
And in fact, if you quantified that around the world,
a lot of people's careers are ended by hippos.
So how do we deal with the hippos,
because it is the hippos that have a fantastic capacity
to create crap, or at least demand that crap be created.
And of course, another animal that causes a lot of damage
is the seagull, as you've heard, who swoops and poops
on your project and then flies on.
So how do we deal with the hippos and the seagulls,
because they're killing our projects.
And they're killing a great customer experience.
It's often, the higher up the organization, the worst
the animals become that we meet up there.
But they often say, no, no, no, no, no.
Let's do this because it really won't affect-- let's publish
this, it really won't affect-- let's put this on,
it's an extra thing.
Of course, you having to do that takes away from your abilities
to be able to focus on the top tasks and make them better.
But let's take that for a moment.
Microsoft Excel, the top tasks there,
or a top task in the Help and Support
is, how do I sum a number?
Basic stuff-- how do I do addition and multiplication?
The problem is that with top tasks and tiny tasks,
there is word overlap.
There are only so many words in the English language.
So there's a function-- there's all these function pages,
or there used to be in Excel, which are for mathematicians
and they're formulas.
But they're called SUM function or IMSUM function.
They're called areas function.
They're called print function.
And you know what?
They're really popular.
For years they had loads and loads of visitors.
And they were saying, I didn't know
there were so many mathematicians around really
interested in the sum functions.
But of course, they weren't interested in the sum
functions.
They were trying to sum a number,
but they were finding the IMSUM function in the search
results coming true.
And that's a classic example of how the tiny task
content, because there is often so much of it,
and there's word overlap, clusters
either the find ability to search or to navigation
of the top task.
Tiny tasks affect performance of top tasks
when you do the analysis.
So they notice and they worked on it.
And what they eventually ended up doing
was they deleted all of the function pages.
And they put them into a single page, called math functions.
So the mathematicians could still
find-- so the tiny task could still
be found by the mathematicians that needed to find it,
but sum a number was on its own.
So you didn't have the overlap of the words coming through.
So we often need to prove the negative impact
of the tiny tasks on the top task.
That's one of the ways we can get to more focus on a top task
environment.
So when you're building up this list-- so
you're trying to prove initially that there
is a league table of importance from the customer.
So you're collecting all these tasks.
There's essentially two sources.
There are customer facing sources,
so they're all the research that you've done
and maybe you need to do new research.
You're looking at competitor websites.
You're looking on social media.
You're looking at your own website.
You're looking at analytics and overall search behavior.
So you're looking outwards into the world of the customer.
And there's the organizational sources.
You have to talk to the seagull's and the hippo's as
well.
You have to get their opinions because if you don't
get the entire ecosystem of the task,
then you can't prove what the tiny tasks are.
And they'll come back and say, well, you
missed the things that are important to me.
So when you're building task list,
it needs to reflect the outward focus,
but also the inward focus as well, in order to be complete.
But it should be roughly an 80/20 split.
80% of the list-- so if there's 100 in the list,
then roughly 80 tasks that are very much outward focused
and about 20 tasks that are inward focused in the process.
Some people say to me over the years,
isn't search and page views enough?
That will give us a picture of what our customers really
want to do.
Not necessarily.
Not necessarily.
Page views, and visits, and search, they
gave a certain window into the customer,
but you can miss very important things.
A couple of years ago we worked with Lancashire County Council
in the United Kingdom.
And we went outwards in our research.
And we were talking to citizens.
And we were looking on social media.
And we went beyond, out into the Lancashire community.
And one thing that kept coming up
was hospitals and health authorities,
that that was important to people.
The broad question was, in living in Lancashire
and dealing with the county council, what's
most important to you?
But when we brought that initial data back to the council,
they said to us, you can't use that.
That's not us.
We don't deal with health.
That's the National Health Service.
We're the county council.
And that's a classical thinking from an organizational point
of view.
But we convinced them to leave it on the list
and see what would happen in the vote.
And when we got the vote, it was the number two.
It was the number two thing that citizens
said they wanted from their county council.
Now they didn't start creating new content.
They started linking and collaborating
with the National Health Service in the process.
But sometimes, there will be tasks
that are outside of your organizational
thinking that the customer really expects you to do.
Now those tasks, because you don't do them,
won't necessarily come up in your search
or your most visited pages.
So keep it broad initially.
You need to think beyond your analytics.
So when we're doing these surveys,
we don't say, on our website what do you want to do?
We say things like, in dealing with your health, when
you're going on a holiday, in buying a car.
So we're thinking about it beyond the ecosystem
of even digital.
So you need to think beyond your analytics.
Search, often, is more symptomatic of a task
than actually reflective of a true task.
Search analytics doesn't always tell you
the true intent of the person when they're searching.
So in Excel, again, lots of people
were searching for remove conditional formatting.
That was a very popular search term.
So the reaction was, let's create a page.
Let's create a page.
Always create stuff.
That's always the initial reaction,
rather than what do we already have.
How could we manage what we already have?
So they created this page, remove conditional formatting.
And in test after test it failed.
It kept failing.
It kept getting very negative reviews and very negative
reactions from customers.
And what they discovered after a lot of pain, and trial,
and effort, was that remove conditional formatting
was a trigger of a behavior.
People would try and format in Excel.
They'd make a mess of it.
Their instinctive response would be
to search for remove conditional formatting,
but they didn't want to just remove conditional formatting,
they wanted to format text.
And when the page only told them how to remove it,
they were annoyed in the process.
So what they ended up doing was they deleted the page, removed
conditional formatting.
And they made sure that when people searched
for remove conditional formatting,
they found the page to format text
under a variety of conditions.
Success rate, satisfaction, jumped--
all sorts of positive jumps.
Search doesn't always tell you what is the true intent when
in a digital environment.
Search is often symptomatic of a top task,
rather than truly reflective of a top task.
So if you really want to define the top tasks,
you have to go broader than just looking at your analytics
and your search based behavior.
And you'll often-- great sources.
We did Microsoft developer technologies a number of years
ago-- Visual Studio.
And there was loads of independent developer
communities where they were discussing issues.
And you can find great sources out in social media,
and out in developer communities or different sorts of areas.
Organizations keep coming back and saying,
oh, that's all well and good, but we've
got our organizational objectives.
They must come first.
There's things we want to achieve in the process.
This was the Norwegian Cancer Society about six years ago.
And it's full of calls to please give us money
because it only is part funded from the Norwegian government,
and it needs to raise money.
So we've got an objective, we need to raise money.
So they say, that's why the page is constantly
calling to give us money.
We went out, we asked citizens of Norway-- obviously,
it was done in Norwegian.
This is just a rough translation.
What do you want from the Norwegian Cancer Society?
Or what do you want in dealing with these issues?
So treatment, symptoms, prevention
were the top tasks in the process.
And then what's at the bottom of the list?
Practically everything to do with donation and giving money.
So the bottom of the list is everything
to do with give money, yet you see legacy gifts, tax
deductions.
Oh, the annual report-- that annual report
is so popular around the world, isn't it?
[INAUDIBLE] and those sorts of things.
The organizations ego is in that long tail- those tiny tax
environment.
So they were in a difficult situation.
They said, what are we going to do?
Are we going to really serve the needs
of the citizens of Norway?
Or are we going to serve the needs of the Norwegian Cancer
Society?
And they decided that they would serve
the needs of the citizens.
And this is the website they came up with.
The top key cancers-- and immediately you go into those
and you get symptoms, prevention-- they absolutely,
relentlessly focused.
And of course, they deleted 80% of their content
as well because they discovered that they
had five or six definitions of melanoma,
because they were drilled into silos of research and funding.
So they deleted a whole bunch of their content.
And they absolutely focused on the customer task.
And you know what happened?
Their revenue grew dramatically.
Their donations grew dramatically.
Because just think about it, if you have an issue
or you're worried about some type of cancer,
and you go to a website like this, and it's shouting at you,
give us money, give us money, are you going to give it money?
But if it actually helps you, you'll give it money.
You're much more likely.
Solve the customer's problem, and the customer
will solve your problem.
But if you try and solve your problem,
the organizations problem, first, nobody
will be happy in the process.
Customer centricity is the best business case
in this age when the customer's so confident and dominant
about their behavior.
Same thing happened in this Norwegian bank.
This was a classical home page.
Banking in Norway-- marketing, lovely hero shots.
I mean, look at that hero shot.
Isn't that amazing?
That is dangerous.
I mean, if she keeps her hands up in the air-- she just says,
oh, we've got 3.9% interest on our savings,
or something like that, but she's never
going to live to see it.
He's going to be able to cash in on the pension.
Maybe he told her.
He's revving there at that moment.
He's saying, I'm going to cash in on the pension.
What a stupid image.
But anyway, most of these hero shots
are absolutely, totally stupid, right?
But marketers can't help putting--
because we have to put big funny images on the page-- of smiling
people, regardless of whatever the situation is.
And then people at this bank, they couldn't login.
They couldn't figure out how to log into their accounts--
the current customer.
And the reason they couldn't figure that out
is that it was called, net bank.
I can imagine the brander's and the marketers
coming together and saying, hey, how do we
give people a unique experience when they come to our website?
Everybody else calls it login.
Let's call it, net bank.
And people were commenting, where's the f'ing login?
There was constant calls, I can't figure out how to login.
So they were getting all these complaints coming back.
And they did something really radical.
They basically stripped everything away.
They stripped everything away.
Radical simplicity, focus on the top task.
Focus on login.
But they did something really clever.
And this is communication at the end of the task.
The best possible place to communicate with somebody
is after they have satisfactorily done something.
When they've succeeded at doing something they actually
came to do, that's the best possible time
to get their attention, or to make them an offer,
or to expand their horizon.
So when people would log out, on the log out page
they would place a single ad for mortgage or something
like that.
So they deleted 50% of the content.
And they got a 500% increase in visits to their product pages.
So when they stripped all the crap off the home page,
trying to get people to visit the product pages.
Because that's the whole purpose of those stupid images,
trying to get people to visit the product pages.
And they simplified.
And they helped people do what they needed to do.
They got a 500% increase in visits and inquiries.
And a whole increase in inquiries on new mortgages.
Focus on what people truly need.
In many ways, it's so counter intuitive to the ego
of the organization.
But there is a tremendous capacity
there to deliver value.
When you're measuring these environments,
there's two things to measure.
So once you've identified your top task, you want to measure,
can they complete the task.
And then the critical thing-- the best practice
is time on task.
So we actually did the National Health Service in the UK
a number of years ago.
And it's a huge, massive, big organization and website.
And the number one task for citizens was to check symptoms.
So that was the crucial task for the NHS website.
So we'd come up with examples of those tasks
like, you've got a stomach pain, you've got a headache,
you've had nausea for a couple of days.
What could that be a symptom of?
And we'd watch people as they try and complete these tasks.
And that's the key, observation.
So the first is to identify what the top tasks are
and the next is, basically, observation
through using WebEx or GoToMeeting.
Remote based observation is a very powerful way
of understanding behavior.
So we would give people these tasks in a remote setting.
We'd set up a meeting.
We'd select our sample carefully.
And we'd just watch them and see what they were doing.
So this person searched for the phrase, stomach pain,
as part of the actual task journey.
And the first result they got was
real stories, stomach cancer.
And it came true.
During treatment for a stomach ulcer,
Deborah Knifton was devastated to find out
she had stomach cancer.
At the age of 40, she had surgery to remove her stomach.
I just searched for stomach pain.
I mean, imagine if you went into your doctor and you said,
doctor, I'm not feeling all that well.
What's wrong with you Jerry?
I've a bit of a pain in my stomach.
Let me tell you about Deborah Knifton.
She came into me six months ago.
She thought she had just a pain in her stomach,
now she doesn't have one.
Would you go back to that doctor?
I don't think so.
So I showed this to the NHS.
And there was 80 people in the room from the web team
and they all went, oh, that's terrible.
But not one person in the room felt responsible.
They were graphic designers.
The blog editor says, I just encourage people to blog.
I'm not responsible for that.
Strangely, the search engine team
didn't feel responsible for search engine results.
They were just responsible for the search engine.
Nobody was responsible.
All of our responsibilities are around format's, or tools,
or organization centric activities.
We need to shift the responsibility.
We need to stop managing inputs and start managing outcomes.
The shift in the new digital world
is away from managing and measuring
what you produce, to managing and measuring
what people consume.
That is the core.
So the metrics need to shift because the metrics create
the behavior.
And we measure success based on what we produce.
We encourage glut.
We encourage production by the very metrics.
We talk about lots of visitors and lots of pages.
We reward based- we encourage bad practice.
We have to encourage good practice.
And the way we encourage good practice
is by focusing on the world of the customer.
They want to check symptoms.
Can they check symptoms?
These are the crucial questions we need to be able to answer.
Now three or four months later they send me an email.
And they said, things are getting better.
We've improved the search engine.
We've added type ahead to the search engine.
Again, classic feature-- we've added a new feature.
We've added a new functionality.
So I went back and I had a look at it.
And I went, stom-- and it said, cancer of the stomach.
So we figured out a way to get you to crap even faster.
You want wrong answers really fast?
Six months ago we gave you the wrong answer
and we scared the shit out of you.
It only took a minute to scare the shit out of you.
Now we can scare the shit out of you in 10 seconds.
Isn't that great?
Type ahead is no good when you give the wrong answer faster.
But again, managing what we're doing,
managing what we're producing, managing what the organization
is actually doing and creating-- we've
got to manage the outcome.
Manage the task because the task remains permanent over time.
It was check symptoms 20 years ago.
It's check symptoms today.
It can be checked symptoms in 20 years.
The technologies to help us check symptoms may change,
but the need of the dominant top tasks remain the same.
Here, Cisco-- we do a lot of work with Cisco.
And download software what was the top task
in their environment.
So earlier in the conference, Lara Hogan
gave this excellent resource, WebPagetest.
And I went to it and I said, OK.
Roughly how long does it take the Cisco home page to load?
So 3.6 seconds.
So when we initially measured downloading software in Cisco,
in 2010, there was roughly 15 pages.
So there was 15 steps in the process.
And it took an average of 280 seconds
to download a piece of software.
So if you look at that from the point of view--
remember we looked there a moment ago, it
was 3.6 seconds was the length of time on the page for it
to actually download.
So there was 15 pages, 3.6 seconds, so we got 54 seconds.
So 54 seconds are connected with page downloading.
But it's 280 seconds in total to do the task.
So basically, there's 20%, roughly,
and we've seen this in many other environments
as well, that's connected with the performance of the page
downloading.
But there's another 80% of time taken up using the environment.
So it's great that we're measuring
speed of page performance, of how it downloads,
but what about measuring all the rest of the time that's
connected with the task?
If we really want to measure time on task--
the top of the iceberg is the page load speed,
but most of the iceberg is the use of the page.
We shouldn't simply measure the page downloading quickly.
We need to measure how the page is used when it's downloaded.
So we need a model that measures the entire quantity of time
of the customer that's being consumed.
Because, typically, 80% of that time is in the use of the page,
either reading, or navigating, or clicking
on links in the process.
So the Obama campaign in 2012, they were relentlessly
focused on speed.
And they said, we made the new platform 60% faster
and this resulted in a 14% increase
in donation conversions.
Absolutely great, make those pages faster.
But they didn't just make the page faster.
They made the pages better.
The donation before optimization,
after optimization-- they did 240 a/b tests in a six month
period on their donation pages.
So they weren't just focused on how quickly it downloads.
They were focused on how easy and how fast
is it to get through and to help people make a donation.
And they estimated that that brought them
an extra 49% lift in donations.
They raised $250 million online.
So they focused on speed.
Not just speed of the physical page
downloading, but speed and movement
through the actual process.
Because most of the value in relation to the customer's time
is found in using the page, rather than
in downloading the page.
So in calculating time on task, it's
not just the element of the downloading of the page.
80% is in the actual use of the actual page.
Is it too long?
Is there too much to read?
Is it too confusing in the process?
So if you were interested five or six
years ago in Dublin weather, or Irish weather in general,
you'd have to be an optimist because it always
rains in Ireland.
There's two very special days in Ireland.
One of them is St. Patrick's Day and the other is summer.
So we're always looking forward to summer.
So five or six years ago, if you were searching for summer,
this is on Google, this is what you'd get.
You'd get, search results page, Dublin weather.
That's good because Google very focused on getting
you the right results, their whole orientation.
Well, if we just look here at Google
in comparison say, to Bing and Yahoo, from day one,
Google has been relentlessly focused
on speed-- on your time.
But they're not exclusively focused just on download page.
What they're really focused on is getting you the answer as
quickly as possible.
It's a different philosophy.
They're not thinking just about your time in the context
of a page downloading.
They're thinking of, how quickly can we get people the answer?
Because the faster we can get people the answer,
the happier they'll be with us, the more
they'll come back to us.
So a couple of years ago here is what you got.
Because you're not looking for search results
we shouldn't manage the formats.
We shouldn't manage the content.
We shouldn't manage the channels.
We should manage the task.
Manage and give people the answers to do that task.
Google is relentlessly focused on your time
and the top organizations are as well.
And today you can go, Dublin weather--
this is type ahead that works.
So in your browser you don't even have to get to the site
and it tells you-- of course, it has
to know the top tasks within weather.
16 degrees on Friday.
Now why didn't they do it five years ago?
Because they didn't have the technical capacities.
The metadata wasn't at an appropriate standard.
Today, the technologies have advanced,
either HTML 5 or otherwise, that allow that sort of thing
to happen.
So you keep thinking about the task, what
it is people are trying to do.
Then you look at the technology in the context of that task.
And you say, how can we use the technology
to allow people to check symptoms more quickly
or to check their weather more quickly?
This is, basically, the empathy scenario.
To really be customer centric you have to look outwards.
The greatest skill of the digital professional today,
in my experience, is empathy.
It's the ability to get out of that rut
of organizational thinking, get out of that world,
and get into the world of the customer.
And try and understand their needs
and understand their actual behaviors.
Now this is a tremendous skill set.
But in most organizations that I deal
with, this person, who is the customer
champion, how are they seen internally?
They're seen as the person who's always complaining.
The person who's always going back to programming and saying,
you'll have to simplify this.
They're seen as a trouble maker in most
traditional organizations because
the traditional organizations measure what is
important to the organization.
They don't measure what is important to the customer.
So the way we change-- because this person is a hero.
This person is the customer champion.
This person is saving the organization for the future.
Because if organizations don't change-- customers
have transformed, organizations have not.
That's the core thing.
Customers are so much more powerful today.
They've got, often, far better tools
than organization employees have.
So how do we transform?
We transform with the metrics.
And crucial to this is a balancing
between the ego of the organization
and empathy for the customer.
Design is a dance-- is a balance between the ego
of the organization and empathy for the customer.
And crucial to ensuring their success going forward
is evidence, not opinion.
Gut instinct is very dangerous in the new world.
We got to get evidence of what the customer is actually
doing, how they're achieving.
Are they successfully completing their task?
That's the basics.
But best practice then is, how long is it taking them?
Can we save them time?
Can we make it easier?
That sort of culture-- that sort of evidence
based behavior, based on what is actually happening
in the environment is the route to success
in the digital economy.
And you might call it evidence, not opinion, or perhaps,
evidence, not assholes.
Thank you very much.
For more infomation >> Jaime Wyatt: "Stone Hotel" - Last Call with Carson Daly (Musical Performance) - Duration: 4:22. 
For more infomation >> Jaime Wyatt: "Your Loving Saves Me" - Last Call with Carson Daly (Musical Performance) - Duration: 3:52. 
For more infomation >> Ben Hardy Spotlight - Last Call with Carson Daly (Interview) - Duration: 4:56. 
For more infomation >> Tika Sumpter Spotlight - Last Call with Carson Daly (Interview) - Duration: 5:08. 
For more infomation >> Barbara d'Urso spiega qual è stato il gesto di Mara Venier che ha fatto finire la loro amicizia - Duration: 2:13. 
For more infomation >> Katalog Tupperware November 2018 | Tupperware Philippines - Duration: 6:06. 
Không có nhận xét nào:
Đăng nhận xét