Janet is a Major Projects Director. Janet has been a public servant for 22 years, mainly in roles dedicated to making things work better for people. She's spent the last 6 years in digital delivery and leadership roles, and is now trying to apply what she learned to support a large portfolio of major projects at the Department for Education - things like building and refurbishing schools, recruiting and retaining teachers, changing the way social workers are accredited, building the apprenticeships service and introducing new qualifications.
Applying digital ways of thinking and working to things that aren't obviously digital
How do you take the best of user-centred & digital cultures and practices and apply them to a whole government department? That's the question we're trying to answer in the Department for Education. In this talk Janet will share some of what the DfE has done so far, what they've learned and what they're planning next.
Janet will refer to DfE's transformation aims throughout her talk, which are:
- be user centred
- empower yourself and others
- make evidence-based decisions
- deliver end to end
Let me make sure it is working.
Hi, everyone, I'm Janet.
I work at the Department for Education;
I have been around in Government for a while.
My role in the Department for Education has
nothing directly to do with digital.
It is the first time I have had a relative like that for a
few years, I'm the Director of Major projects.
I'm responsible for the projects we do, the major projects in DfE.
This isn't working, it is where you learn,
what you learn, how you learn it, who teaches you,
all that kind of stuff, we have projects to
do with all of that kind of thing.
So, I'm going to talk today a little bit about
why I want to do that role, why am I working on
that and how we are trying to apply digital ways of
working for all of those things, I will talk about
why it is all about everything, not just websites.
Why that is, how we have been going about this
in the Department for Education and a couple of reflections.
So, first of all,
it is about everything.
I'm going to give up on that.
So, the point of digital, I reckon is to
change absolutely everything.
Do you disagree with me?
That is what the point s Martha had it write when
she wrote the report to establish DDS,
she call it evolution, not revolution,
make massive change, not on a few little things,
everywhere, I adore Martha, she was right on this,
as she has been on many things,
digital is a Trojan horse which quite
obviously is what that is.
And with a Trojan horse, you go in in disguise to say
- we are here it look at your website and then you go
- we are now going to change up all area shit.
But you have to make sure you get out of the
horse and that is what I'm doing by doing this job at DfE,
getting out of the horse and changing up all the shit.
I am I have now outed myself having been in
disguise in DfE for the last year.
I reckon there are four stages of digital Government.
I talked about this last year,
and if you want to see more about it,
you will see it there, I will give you a quick
summary of my reckon, you start with pre-digital and
there you care about social media, and
that is what digital means to you, it happens when I talk,
I get excited and my nose runs, so,
that is why I'm sniffing, in a minute I will get defeat and
get my tissue out.
For a moment, I will plough on as
though it is not happening.
Stay with me on that.
The next stage is rationalise, we care here about
improved content, maybe putting it all in one place,
making our forms work better for people,
very important work.
And we are all about doing things more
efficiently at this stage, we will move away,
remove duplication, do things, simpler,
clearer and faster, like we were trying to
do on gov.uk when we set it up.
But we are now thinking about how to
integrate all the things, how do we make
coherent end-to-end services.
It is welcome to see the steps that have
been made in DfE where I am from.
We have a set of people outside sharing some
of our work, too, talk to them.
This is one of our services, there are loads
all around Government doing end-to-end
coherent service design for people.
It is about doing things differently,
not more efficiently but differently and better.
It is also not what the point of digital is, for me,
the point of digital, the really exciting thing,
the reboot is where you reinvent Government
for this age, just as Star Trek was reinvented for
the 21st century, if you don't get, that blame James,
he thinks the Star Trek reboot is highly relevant.
I don't watch it.
That is how that goes.
The reboot is about doing different
things altogether, not about being more
efficient or doing things differently,
we are not talking being digital government,
just government and controversially maybe not
about services any more,
we are talking about absolutely everything.
Why is that important?
Because, if we talk about services,
we try to shoe horn everything government does
into service language and it doesn't make
sense to people who are doing things in
Government that aren't services like making laws,
making big policy decisions, doing regulation and
stewardship, like looking after the schools system.
And doing enforce am, you cannot say
enforcement is a service in the same way
that paying your taxes might be.
So we need to deal with all of this stuff,
not just the services.
We need it make Government into all of
these beautifully arranged,
slightly increasing in size words.
But I have not chosen them for that reason,
it is because they are the correct words to
describe what any organisation needs to
be like in the digital age.
It all starts, of course with putting users at
the heart of everything you do.
So, we need digital skills and tools and
ways of working for everyone in the organisation.
And I'm talking here about Government,
but my reckoning is it applies to any organisation,
if you are not from Government make a
leap of imagination and think about which
things we have in common as I go
through the rest of the talk.
So, I'm here to change everything,
just to make it clear.
I'm not here to mess around with
pixels although I like that too.
Why is that important?
Being user-centred is the starting point for all of this,
if you don't do that, you might as well not
bother being responsible for the other things,
you want to put users at the heart of
what you do, why is it a good idea?
You will do less things that are a
really terrible idea, you will realise they are
a terrible idea before you release them on
to unsuspecting humans.
Things will work better for people;
you will cock it up less and
there will be less waste.
So pretty obviously something that we
should all just do, I will have to give up and admit defeat.
Talk amongst yourselves, there we go.
It didn't happen, nobody saw it,
can't proof anything!
OK, so why is government not
just doing this, then?
Here are some reasons, one is the
way Government works is designed for the
pre-internet era, we do things like policy.
And what policy is, is having thoughts and
writing them down, in summary and
also talking to some people about them.
And we also, we are stakeholders,
deal with stakeholders and they are at the
forefront of what we do a will the in Government.
We invented stakeholders when we couldn't talk to
people at scale and do experiments with
people at scale because there was no
internet to do that on, so we have to talk to
stakeholders to tell us their continues about
what they thought the users they claim to
represent might need and lots of failure in
Government has happened because of that.
But it is an entirely rationale thing to do in a
pre-internet era where it wasn't possible to
talk to people at scale and get their views and
find out what is going on.
The other thing about Government is we
solve every single problem, in Government,
we reckon, can be do about things,
addressing things better and
handling stakeholders better.
And they are the fundamental skills of
a civil servant.
We reckon in Government if we do that well,
we will succeed.
By that I mean laws will be passed,
the consultation document will be published,
we will get away with it without being on
the front page of the Daily Mail.
The main metric of success in Government.
This all happened because of this era.
It is the 19th century, the point of ministers at that
time was to tell the civil servants what the public thought.
They had politicians who found out what
the public scared about and thought and
went to Whitehall and acted on it
faithfully and effectively.
So how is that working out
for you right now?
Representatives, what is going on there?
But that is another talk.
That is how government is designed and
now the humans who work in Government,
we don't know how to be user-centred.
Most of the people working in
Government have not heard of user-centred design,
it is because it is not relevant to them.
We have not made it relevant for them.
They think it is for people in jeans and
converse trainers in the digital team.
And we in the digital community have created
that impression, we have not told them how to
do it yet much it is scary particularly if you have
not been talking to your user very much or
thinking about them.
The first time you do it, it is a shit show,
they unleash all the complaints they would've
made had you talked to them all the way along,
you have to push on through that pain barrier and
it is painful and upsetting to discover that
what you have done didn't actually work for
people and is causing them pain.
These are all rationale reasons, I should say.
While I'm saying this, it is not to belittle anybody.
I'm saying the way Government works has arisen as a
set of rationale responses to the situation people
found themselves N when I was in GDS we
described the way Government did things,
and the traditional way was on the top.
The new way is from the bottom.
this is to do with transactions and services.
We said the way things happen in Government,
which is rubbish, is we make up a policy and
then we spec out the requirements and
probably outsource it for somebody else to
deliver it and then it gets delivered and
then we unleash it on users, I nearly said
losers then, a Freudian slip.
There you go, losers, and
see what happens then.
And what we say is, don't do,
that involve users all the way through,
find out what they need, these are users,
every single stage, find out what they need,
iterate, make it better.
I have got experience in different Government
departments and in how it works in practice and
also things that are not services.
These are policy people, happy, smiling.
What they are doing is making announcements or
helping ministers to make announcements
and having thoughts.
I should have said I am not a
designer but I like colours.
Don't judge me.
The problem with policy people is they
sit behind a very big wall which that clearly is one of.
A big thick wall with no windows or doors.
On the other side of the wall,
here are some people trying to
deliver services to people,
people working in Jobcentres, for example.
The problem here is, as you can see,
there is a big wall between these sets of
people and they have different world views and
come from different communities and
have different ideas about how to
do things, even if they want it
communicate with each other it is difficult.
The medium of choice, communicating
between policy people and
delivery people is a hand grenade.
That is the grenade of policy ideas.
What you do, you lob your grenade over the
wall and then you carry on making more announcements.
And the poor delivery people over here,
unfortunately have long-standing smiling
faces there - in case you don't get the detail.
That is when it works well, sometimes it doesn't,
sometimes when you make your policy and
before it leaves the policy arena,
it blows up and that is the front page of
the Daily Mail right there.
I have been there,
it is not a good place to be, let me tell you.
And when I say all of this,
I am not blaming policy people,
I want to be clear, it is not their fault.
The prime directive tells us that
everyone is doing the best with the
information that they have available to us.
I have to tell you it is true of policy people,
it is not just digital people who do
their best with the information available to them.
There are others who do that too.
So, that is why I want to
change absolutely everything.
We should all want to do, that right?
So, in DFE we have been trying to
And I'm going to explain a bit about
how we have done that.
The question we have been trying to answer is:
How are we going to make it normal and
uniform in our organisation for people to
put users at the heart of everything they do?
Not just when they are making digital services or
dealing with websites but absolutely everything,
make it law, doing regulation,
do enforcements, everything centre?
Define being user-centred as knowing who
all our users are.
The word "all" is important.
We two out of our way to discover the
full range of our users and we know their needs and
we can meet them.
We are using the words "users" on purpose,
we had a debate about whether to
say humans or people,
we haven't because we are trying to
differentiate between users who have to
use our services or have to interact with us
in some way or are affected directly by
what we do and other people.
For example, stakeholders,
they are not users, stakeholders have opinions and
it is important to understand what the opinions are,
and if you don't, then they will put you on
the front page of the Daily Mail and
you will be shut down, you need to
understand their opinions, respect them,
and deal with them but they are not
users or they may not be users.
The other people who are not users are ministers.
Ministers are not users, they very important,
they are political masses and they are
respected and we must talk to them and
we must respect what they want to do
politically and they definitely should not be
telling you what colour the button should be
on website, what is who we think the users are,
that is why we use the phrase users rather than humans.
That is contested within DFE.
We identified and what we tried to do is
distil our collective experience and that of
the wider community, lots of people contributed to
this online for which we are really grateful and
distilled it into six user centred practices.
Our theory is, you can apply these six user
centred practices born in user centred design to
anything at all, that you do, in an organisation,
like the Department for Education,
or any other bit of government.
I am going to run through these and
tell you what they are and tell you
examples of how we have applied these in other places.
Define the outcome.
Define the outcome from the point of view of your users.
Not as a vague statement of overall intent.
So, driving forward an ambitious social
mobility agenda is not a statement of clear intent,
it doesn't tell you what outcome you are trying to achieve.
Maybe you are trying to make it easier for
unemployed young people to get a job.
You can tell if you have done that or not.
Maybe you are trying to reduce the number of
people who are young and unemployed,
that is an actual outcome you are trying to achieve.
Unless you define that outcome,
you can't possibly be user centred because you
don't know what you are trying to
do for your users and will be led by
random bits of feedback and that's how
agile projects get lost actually.
So what we are trying to do,
this is really obvious,
you should obviously have this process set
out for anything you do, you should obviously
if you are going to do inputs you
should work out what is the process,
what are the outputs and the outcomes
we are trying to achieve.
It is astonishing how much of the stuff we
do in Government we do without having
mapped out this simple thing here.
What we tend to do instead is focus on
the inputs, the policy announcement is we are
going to spend 20 billion on the NHS,
isn't that great, I don't know,
I don't know what you are going to spend it on or why.
Or we focus on the outputs.
Universal credit is an example of this,
where we have said we are going to
build this thing called universal credit but
it's remained contested throughout the
life of that project, what the point is of doing it.
So, time and time again that piece of
work has been derailed by a change in
view about what we are trying to
achieve because we didn't define the outcome at the beginning.
First thing, define the outcome.
Next thing, understand your users.
Start with user needs,
and Matt referred to this earlier.
We came up with this when I was working in
NHS Digital but you can apply it broadly.
Not just the transactional need the person has,
but also their emotional needs and
that's really important in government
because very frequently the reason somebody
is come to government is because they are
worried about something.
So the reason parents want to find out
stuff from the Department of Education is
because they are worried about their kids.
The reason people want to go on the NHS website
is because they are worried about their health.
If you don't take account of that you
are going to mess it up.
Have practical needs and Sarah has been
talking about that, really fantastically well about
the different practical needs different people.
All of those combine to make up somebody's user needs.
But that in itself, that's necessary, but not sufficient.
I don't know about you but I don't see myself as a
collection of needs.
I can be a bit needy but that is not all I am.
There is are more to me than my needs.
As I managed to persuade James of this recently,
he's actually married me, idiot.
Nobody here is just a collection of user needs,
are you at any given time, you have a context,
you have a back story,
you have always sorts of other things going on
in your life, we need to go out of our way to
understand what is life like for the teachers we
are trying to help or for the headteachers or learners.
We need to work out what is already known,
which we often just ignore.
We often do this in digital teams in
particular because we think although somebody
has done some research it's probably a bit rubbish.
But analysts, and other researchers can
actually do useful things.
I have discovered.
We should look at it and find out what
nuggets there are in the stuff they have done.
Then we should work out how to
find out what else we need to know.
We miss out these steps, we dive in and
that is just being reckless and wasteful.
We have to be mindful there aren't any
new problems, not really.
Most things we are working on in
government or any organisation they are
not really that new,
loads of people have tried stuff before.
We should learn from what others have done.
This is my favourite one.
Test assumptions, what is missing here and
is missing from an awful lot of what we are
do is what are the assumptions that need to
be true for this process to actually happen as
I have laid it out here.
I have got a really lovely example from DFE,
but before I get on to that, this is my fact
about how much policy failure happens
because of assumptions.
If anybody here thinks they can prove that wrong, t
hen have a go.
It could be higher than that.
So individual learner accounts absolute
cataclysmic shambles of a thing that
happened in the Department for Education and
we are going to own that failure in the DFE.
What we did was we introduced a thing that
was supposed to help young people find
training that would help them get better jobs.
But what we did was, we designed an
entire scheme without really trying it out.
Well we did try it out, we did pilots,
they didn't work, we ignored it and just did it anyway.
The reason we ignored it was because the
thing had been announced in the manifesto of
the Labour Government in 1997.
If you have a thing that is announced in a
manifesto, you basically have to do it
even if it is a terrible idea.
That is how it works.
I don't know what we are going to do about that.
It's really bad.
Because you go to the Treasury and
say we did these pilots and they didn't work and
they say "manifesto ".
That's the end of the game.
They have mentioned the manifesto.
Individual learner accounts,
supposed to help young people,
the idea was you have an account,
you put 20 quid in and you get 100 quid to
pay for a course and to make it simpler and
faster you didn't have to do anything to
prove your identity if you were a learner,
you didn't have to do it yourself,
it could be done by the training provider and
for the training provider there was no
quality assurance at all, we wanted to encourage
more trainer providers to come on board.
Some of the assumptions we made,
young people with no skills will access the
internet independently, create an account,
seek out a course, sign up for the course,
submit a draft to get the money, get the money,
complete the course and then get another job,
so that is about 20 assumptions I have listed there.
Who wants to guess how many of those
we actually tested before we released the scheme?
Not a single one.
We also assumed because each account wasn't
worth very much money, that no fraudsters
wouldn't be interested and all of the,
who is going to defraud for £100, it's not worth it.
So, all the lovely middle-class civil servants who
never occurred to them to commit fraud did a
risk assessment and went nah,
noone will do that, the project was launched,
runaway success, hundreds of thousands of
people signing up, way beyond our wildest dreams.
There were a few teething problems.
We didn't bother to investigate them,
we just went teething problems, never mind,
headlines, manifesto tick, excellent and
then it became clear that people were
having problems and people having
problems with the quality of
the courses, people being signed up to
courses they weren't doing and we had to
shut the scheme down.
That's quite a dramatic thing to do.
It wasn't just the Daily Mail that got the
head lip on that one.
The thing that gets talked about a lot is that we
wasted £100 million on fraud,
because fraudsters basically without ever talking to
you just registered you for loads of
courses that didn't exist.
There were thousands and thousands of
nonexistent people on courses.
But the actual scandal was the other
200 million we spent was spent on
basically middle-class white people who
already had degrees and perfectly good jobs.
So not the target audience,
the people who are supposed to be benefitting,
only 10% of the people who actually access
this thing were those people.
So, 300 million wasted because of
Test your assumptions would seem like a good idea.
Four is involve users, not just observe them and
look at them in a lab, and do things to them.
And then release stuff on them.
But actually involve them.
So the teacher recruitment and retention strategy,
I wouldn't normally get excited about a
strategy but this one is so exciting because it's
got the best piece of feedback I
have seen about a piece of government strategy.
Here it is.
A noteworthy thing about this strategy is the
absence of bat shit crazy ideas.
Most fail at it and that is true,
as Matt said earlier, no bat shit crazy ideas.
We developed the strategy with people who
were going to be affected by it.
We asked teachers why are you leaving,
what is making your life hard,
what can we do to make it easier and
what would make you stay, and as a
result we have now got stuff we are
doing that is not bat shit crazy and
hopefully won't fail.
Observe actual behaviour,
I am going to skip over this,
this is an important thing but we do it for
digital services, we don't do it for
anything else and I don't know why not.
Here is a team I used for part of many years ago,
with Pete Gayle, and the final one is deliver,
test, learn and adapt.
What is interesting about this one is,
we want to it rate services we all do that
in digital services, but you can't do that
when you are building a school.
Well it turns out you can, if you have a
programme of school buildings which
stretches to several billion pounds.
And you are going to build hundreds over several years.
Because what you can do then is seek time you
build a school you can see how it is used,
so I will tell you a story about this.
When we first started in DFE our
first programme to build and refurbish schools we
thought it would be great to get
headteachers involved and paired them with architects.
What happened was some really good
schools were built very expensively and
some really unusual designs came out.
For example, one school wanted to have a
library and a running track and to
optimise space they made the running track go
through the middle of the library.
That's an actual true fact.
Other schools wanted to have partitions so
they could open up the classrooms and use big spaces.
They were never used.
They were really expensive live and made it
noisy in both classrooms.
So we stopped doing that and instead we
now do evaluation of each school after we
build it to see how it's being used and we
have developed patterns for how to build schools,
what colour to paint them, how to lay them out and
we use those for each school and the
result is they are better and they cost 35% less to build.
Which is amazing, talking about billions of
pounds here, talking about extra schools we
can build because of taking a union centred approach,
only for agility and websites,
that's not true.
So those are our six user centred practices.
How do we make that change.
You have loads of staff across loads of
sites doing loads of things and this is a Yapp triangle,
I love it, I use it 40 times a day at the moment.
It is like when I like a record and listen to it to death.
Next time I will see you, I will be Yapp triangle rubbish.
You have to choose your method accordingly.
To in DFE we really want to do this by agreement,
we reckon if you do it by setting rules and
people don't understand why and
haven't agreed to it, then they are going to
work around the rules and ignore them and
we have had plenty of experience of that
before in other areas where we have these
things called policy tests and
nobody ever does anything meaningful with them at all.
Tick the box, done.
For that reason, we have emphasised having
discussions and running experiments rather than saying rules.
We have had a structured conversation in
every single team in the organisation
going through those six practices and
thinking about how they might apply for that team.
We offered a whole load of
different training about half the staff in the
organisation took part in training on
things like how to do that logic model or
how to do user research and we find pioneers and
help them, which sounds obvious but
actually the experience of being a
pioneering government if you have never done
it is very lonely and the fear of the front page of
the Daily Mail is real.
And will meet you at every turn in the form of
what people saying what about this,
what about that and the other,
it is exhausting.
So we help pioneers,
we find them and go you are awesome,
how can we help you and
it makes a difference for pee.
That's how we have been trying to do it.
My two final reflections I am
going to make before I stop,
I might have time for questions,
I am not sure, the problem with this approach of
doing it by agreement is,
I am not super-patient.
I want it all to happen right now,
so I am a little bit like, for the love of God,
how long is this going to take,
and the answer is it's going to take generations.
There's no point in getting all impatient and
saying how long have we been at this, can't we get on with it.
This is going to take a long time.
It is a generational thing.
If we wanted to do it now,
we would have to set rules and as I said,
we think doing that prematurely
would result in people working around them and
ignoring them and it won't get the
change we want to happen.
So this is really hard for me because I am an
impatient soul and want to change
everything immediately but we have to be patient.
And my second reflection point is,
when we think of policy people,
we tend to diss them and talk about bowler hats and
idiots and do things like that,
that could be seen as disrespectful towards them,
though I did emphasise it is not their fault.
We pull this face at them.
We go, what, have you dropped on
your head as an infant, are you stupid,
that is the face that digital people,
that is the face policy people see when
digital people turn up in their converse trainers and
Mac books, saying haven't you heard of
user centred design.
People don't respond well to that.
Even if you think they are idiots,
they hold a lot of power.
They don't respond well.
Treat them like idiots, if you want,
if it is how you feel but they have a lot of
power and they will stop you doing things and
they can and they won't listen to you,
they get produced by people with
whacky ideas 100 times a week.
This is what they see when they see
digital people coming at them.
So you know, you see a bowler hat,
they see this on a skateboard they think
- you will be here for a few months and
then you will go back to, you know, Google,
or whatever, whatever, see you later.
So we extend a lot of respect and
generosity to our users but we do not he
extend the same respect, generosity and
patience to colleagues, every person here,
started from a position about not
getting being user-centred.
We have to respect other people who
have not got it yet and recognise they have not got it yet.
They are there for the turning and persuading.
So, those are my reflections, PS,
we are hiring in DfE, if you like what you see,
come and have a chat.
And that is me done,