7 tips for improving decision-making at work

I've recently spoken with various people (mainly from UK local government) who say decision-making has improved at work during the Covid-19 pandemic.  They say they're being made more quickly, distributed lower down in the hierarchy, and without as much bureaucracy.  

This is all good news.  However, some of those same people were also concerned that, as we move out of crisis mode, some of the old bureaucracy, delay, or sign-offs might creep back in.  There's a strong desire to 'keep' the better way in which we've been making decisions as we move in to our next normal.  

In order to do that, I was thinking that maybe some methods or approaches might be helpful for sustaining some of the better decision-making that has emerged during Covid-19.  

A while back I tried to learn more about effective decision-making in the workplace.  Here are 7 tips based on what I learned:

1. The advice process

The book Reinventing Organizations by Frederic Laloux looks at pioneering progressive (or as the book labels ‘Teal’) organisations.  It researches practices relating to self-management, ‘wholeness’ in the workplace, ‘evolutionary purpose’, and using the metaphor of organisations as living systems.  

One practice relevant to decision-making is the ‘advice process’  – a way of distributing decision-making in organisations.  The book cites a number of organisations using the advice process, and it can take many forms.  In essence, any person can make any decision after first seeking advice from everyone who will be meaningfully affected, and people with expertise in the matter.  

The decision-maker must take the advice they receive into consideration.  With all the advice and perspectives the decision-maker has received, they choose what they believe to be the best course of action.  No colleague can tell a decision-maker what to decide, as long as they follow the advice process.  It allows anybody to seize the initiative, distributes authority, and creates responsible practice. 

2. Could Loomio (or its principles) help?

Some organisation are using decision-making software like Loomio to complement the advice process.  The process on Loomio begins with a discussion to frame the topic and gather input, host a proposal so everyone affected by the issue can voice their position, and then the final decision-maker specifies the outcome (automatically notifying the whole group).

The people consulted have one of four options when giving their input:

• I agree with this, and want to go ahead
• I abstain, and am happy for the group to decide without me
• I disagree and think we can probably do better, but I will live with this decision
• I block this decision as I have a strong objection and am not okay with it going ahead

Samantha Slade points out in her book Going Horizontal that this shifts the standard question from “Is everyone okay with this?” to “Is there anyone here who can’t live with this?” or “Is there anything here you can’t live with?”  It gives greater autonomy and responsibility to the person proposing the decision, and better clarifies when and why a person might decide to reject a proposal.  

3. Easier decision-making

I once had a chat about organisational decision-making with Andy Brogan, founding partner of Easier Inc.   

He gave me a few helpful pointers (which I hope I've represented accurately):

• In order to understand what's currently happening with decisions, you can get helpful insights from mapping the 'journeys' of decisions through different people in the organisation (e.g. when you have multiple sign-offs).  This gives you evidence of where decisions have been delayed, deferred, re-shaped, or where there’s been duplication or rework.  And that helps make the case for change. 
• Do we need perfect decisions that are slow and bureaucratic, or ‘good enough’ decisions that are faster and involve less work? [I think Covid-19 has helped trigger a big move to the latter, at least temporarily.]
• It’s useful to identify and agree on what the characteristics of a good decision are, and make this explicit and transparent by turning them in to a set of clear behaviours or principles.  

4. Intent-based decisions

Andy also reminded me of the book Turn the Ship Around! by L. David Marquet.  It tells the true story of how, as captain, Marquet turned his submarine from worst to best performing in the fleet in under a year.  He did this through creating a different approach to leadership – a ‘leader-leader’ model instead of a ‘leader-follower’ model.  Many of the ideas in the book are summarised in this 10 minute video.  

The book introduces the idea of intent-based leadership.  Instead of giving orders, the crew would say “I intend to…” (not “I request permission to…”) and he would reply “Very well”.  Rather than being a minor trick of language, it was powerful in changing the culture and creating responsibility for decision-making among the crew.  It helped turn passive followers into active leaders. 

 

They later extended the concept.  Often Marquet would have to ask questions before saying “Very well”.  To address this he asked the crew to consider what questions he would have, and to provide a sufficient explanation that would allow him to say “Very well”.  This forced all of the crew to think like a Captain: “Instead of one captain giving orders to 134 men, we would have 135 independent, energetic, emotionally committed and engaged men thinking about what we needed to do and ways to do it right”. 

Going slightly off topic, I think this could help in re-writing some of our post-pandemic working from home policies.  Instead of asking permission, people just say "I intend to work from home tomorrow because....", and then they do.  

5. Above or below the waterline decisions

Sticking with Marquet, I also like the way he frames decisions as 'above or below the waterline'.  

He uses the metaphor of a boat, where a hole below the waterline can sink it. On the other hand, a hole above the waterline will cause damage, but it’s not going to sink it.  Decisions that are ‘below the waterline’ could potentially cause harm to the organisation and might require greater control.  ‘Above the waterline’ is where greater freedom and autonomy can be given – particularly in relation to innovation.     

6. PLAN decision-making

Folks at Vanguard introduced me to this one.  I gather it originates from the Police, who use the acronym PLAN to guide decisions about use of force and respecting human rights.  It stands for:

• Proportionate – Does the decision have the right balance between achieving the desired purpose, resources available, and risk?  
• Legal – What does the law say we can or can’t do?  Does the law say anything about this?
• Accountable – Can the decision-maker sufficiently explain or justify their reasons for making a decision to someone else?  
• Necessary – Why did the decision need to be made?  How does it contribute to achieving the organisation’s purposes?  What would have been the consequence of not making this decision?  

Sometimes the acronym is extended to either PLANE (Ethical) or PLAN-BI (made on the Best Information available).

This can be helpful for front-line teams, to give them an autonomous framework for decision-making (e.g. for solving 'customers' problems on the spot).  For example, the manager will trust their team members’ decision as long as they meets the PLAN criteria.  

I also find the L for Legal useful in local government, where it provides some peace of mind for people who are concerned about compliance with statutory obligations.  

7. Create conditions for 'fearless' decision-making

If we want people to make good decisions, they need to do that in an environment without fear – that is 'psychologically safe'.

Psychological safety means a climate in which people are comfortable expressing and being themselves.  When people have psychological safety at work, they:

• feel comfortable sharing concerns and mistakes without fear of embarrassment or retribution.  
• are confident that they can speak up and won’t be humiliated, ignored, or blamed.
• know they can ask questions when they are unsure about something.
• tend to trust and respect their colleagues.

It means mistakes are reported quickly so prompt corrective action can be taken, coordination across departments is enabled, and ideas for innovation are shared.  It also means more effective decision-making.

In her book The Fearless Organisation, Professor Amy Edmondson gives a leader’s toolkit for creating psychological safety, summarised in the table below.  If you're in a leadership role, or have influence in your organisation, this toolkit might help create conditions for better decision-making.

Are these tips helpful?

The above isn't intended to be an exhaustive or authoritative list.  At the same time, I hope people reading this find at least some of it helpful.   

What other tips might you add to this list for improving decision-making at work?  Has Covid-19 led to improved decision-making at your work?  If it has, are you now starting to see some of the bad old ways creep back in?  How might we stop that?  

Please feel free to share this post with anyone you think might like it.  

Tackling backlogs - a conversation with Andy Brogan

Recently on the Next Stage Radicals Facebook group, fellow 'radical' Adrienne Rogers started a conversation about backlogs.  At the hospital where she works, lots of activity has understandably had to pause because of coronavirus, meaning backlogs are growing.

Backlogs will be a cause of concern for many of us at the moment.  This crisis could mean:

• Increased demand, like at a council's Welfare Benefits service I spoke with earlier on in this crisis, who had in two weeks received as many claims as they had in the previous two months.
• Fewer people to do the work, because of increased sickness, self-isolation, caring responsibilities, etc.
• Simply being unable to do the work, because it's unsafe or impractical.  I spoke with a council's Food Safety Team, who've been unable to do any of their scheduled inspections. 
• The exceptional priorities this crisis has requires of all of us, with people taking on different roles, or being moved to where they are most needed.  That same Food Safety team were speaking about the possibility of them doing coronavirus contact tracing, as they're already skilled in doing this for other diseases like salmonella.

During the conversation, Andy Brogan (founding partner of Easier Inc) shared what I thought was some really helpful advice about backlogs, in both business and human terms.  I thought others might benefit from this advice, and I was keen to know more, so Andy and I got together for a (remote) chat where I asked a few questions. Here's how it went.

Q1: If you’re a team with a backlog, how might you start to tackle it?

The first thing is to consider how we frame the problem.  It's easy to think of them as backlogs of activities or things to be done.  In healthcare and public services, we may see queues of referrals, assessments, appointments, and wonder "How will we ever get through these?"

So I'd suggest we start framing our backlogs as backlogs of people to help, not backlogs of things to do.  This has a few benefits:

1. In our backlogs there may be multiple items relating to the same individuals, families, households, or communities.  By framing the backlog as 'people to help' rather than separate activities, we can collapse out activities in to something much more effective.  This strips out the noise and duplication and gets what matters done much faster.  I saw this an an insurance firm a few years back.  Their backlog was around 10,000 items, but through a household centred lens it was really about 1,500 households.  Understanding this totally re-framed the operational response. We created small multi-skilled teams that had case ownership rather than task ownership, and saw the queue evaporate within a few weeks. 
2. Taking this human-shaped lens gives us the opportunity to focus on doing only what matters now, post-crisis. The things that were wanted and needed pre-crisis may look very different by the time we get back to work, so doing the activities that are sat in in our queues may not be doing what matters now
3. Another opportunity – especially in public services – is to recognise our relationship with people, families, households and communities may have changed.  Their capabilities, expectations, and openness or desire to embrace different methods of engagement may have changed. They may be more willing to use technologies like Zoom or social media.  The pride in our key workers and the spirit of responsible citizenship and reciprocity between people and public services that this crisis has encouraged may mean there are greater opportunities to support people to self-help, to help each other, to volunteer, to hold group consultations and so on.  We can capitalise on the experience we have all been going through without abusing it, and we can all emerge a little bit different, a little bit more capable and perhaps even a little bit kinder than before.

Q2: What information might be helpful for people to know while working on a backlog?

I might start by asking “who is in my queue?” instead of “how much is in my queue?”

I’d think about where it makes sense to look at the queues as individuals, and where it makes sense to look at them as ‘hotspots’.  By this I mean groups of people who share characteristics that may be important to how we can help them.  For example:

• people who are related or live together (where a household shaped response could be an opportunity);
• people who live in the same area or access the same community centres, e.g. schools, GP practices, sports clubs, etc (where providing or supporting a local response could help;
• people who have the same needs and requests as each other (where supporting them as a group of providing peer support could help)
• people who are likely to be capable of self-help (where written information, a conversation, or group session could set them free).

There might also be a simple basis on which to prioritise who to focus on first, for example who is likely to experience harm if we don’t help them within the next days or weeks?

Q3: What are some practical things a team working on a backlog could be doing – say on a daily or weekly basis?

There are a few things that I think are really important for teams to do.

1. Make the backlog visible at the team level.  Avoid running lots of little backlogs everywhere which end up hiding or compartmentalising the issue, and creates extra pressure on people to reduces their queues, which may not be the same as doing what matters. 
2. Have a clear priority rule for who gets helped and in what order.  Without this it's really hard not to get drawn into fire-fighting the 'hotline' requests which quickly pulls effort out of shape. 
3. Have clear definitions of what good looks like, both for the people your work exists to help, and with progress through the backlog.  For example, how would you want the people you are trying to help feel as a result of your help? When would you have helped 'enough'?  What would you want to be true as a team at the end of each day, week, month, etc?  What would you see if you were making good progress together?  Framing these things can help quickly reduce tensions, make sense of progress, and agree decisions about whether, when, and how to adjust.
4. Have regular check-ins, favouring frequent and quick so visibility and collective responsibility are maintained.
5. Have meaningful 'in the work' support, where someone is paying attention to what's happening and why, what's being learned, what may need adjustment, and where judgement calls need to be made or supported.  

This last element can really grease the wheels of progress and provides a great opportunity for rapid and dynamic learning and adaptation, as well as a sense of ‘being in it together'.

Q4: What challenges to you think a team might face?

Every context will reveal its own challenges.  A key one for management and leadership is to recognise that the old measures may not be helpful in the new world. Continuing, for example, to monitor things like waiting times for specific activities doesn’t really make sense in a human shaped – rather than activity shaped – world.  This will also drive behaviour back in the direction of ‘getting things done’ over ‘getting people helped’.

In some contexts a predictable and important tension can also be around where service scope should start and end. As we take a person-shaped lens we may see that what would really help people is something quite different to what we might normally offer.  Or it might be something that is beyond our normal scope of expertise. In such circumstances it really helps to have clarity and peer support:

• Clarity: do we have a shared collective understanding of when we have done ‘enough’? Can we afford for our customer to define this for us? Can we afford for them not to?
• Peer support:  is it easy for us to get help with decisions that may be at the boundary of what’s clear and certain? Have we baked this into 'the work', which will enable us to keep working effectively and at pace.  Or is it remote, hard to access, or hidden behind committees, email queues, etc – which will leave us juggling activity rather than getting people helped.

Q5: Can you give any examples of when you’ve helped people with their backlogs?

In pre-crisis times this was a common issue.  I’ve helped quite a few organisations and teams – public and private sector – with backlogs.  One of my favourite stories is about what happened after I’d left a pensions firm I’d worked with. They had a predictable problem every year during what they called 'pension season'. A flood of requests to top up pensions before a tax deadline would hit the firm and send queues into meltdown. It would take them a full quarter to recover their position.

When I worked with them we realised a more human-centric response to their work in general – not just for pension season – would mean using multi-disciplinary teams who had case ownership, much like the earlier example I mentioned. This naturally led teams to do a degree of cross-skilling. People in the teams spotted that simple tasks they could each do separately but not all do together, could be easily trained in and would create more resilience. 

When I checked back with that client 18 months after working with them I asked ,"How was pension season this year?" to which they replied "You mean pension week?"  The teams had cross-skilled so that each team member now had the specific skills required to do pension top-ups. This took the operational capacity to handle ‘pension season’ from being a few skilled people to a system wide capability. Pension season now hardly registered as a blip on their radar.

Q6: What opportunities might this give us to improve how we work once this crisis is over?

I think the big opportunity here is not just backlog busting nor even ‘just' better, more effective, more timely or more efficient service. All of these things are possible and valuable but the huge win is what it does to the workplace. 

Imagine a workplace where you:

• Have real collective ownership of the work that exists to be done
• Work together to stay on top of it
• Are supported to grow your skills in response to the opportunities you identify for adding more value
• Have a voice in what’s happening and how that’s affecting you as individuals and your shared mission as a team
• Have a greater and much more systematically supported connection to the purpose and value of your work because you are dealing in humans and families and communities, not in widgets, referrals, tasks and activities.

In a pretty profound sense, I think that what can seem like a fairly prosaic task – "tackling a backlog" – could in fact be a perfect opportunity to truly re-humanise work.  Don’t you?

Want to know more?

If you’d like to see more stuff like this, have a chance to chat with Andy, myself, or other like-minded folks, then check out Next Stage Radicals.  

We’re an open learning community made up of people with a passion to make work work better. There’s more on the website, or follow us on Twitter, or feel free to get in touch with Andy or I if you’d like to know more.  

How traditional funding would ruin something good

I've been doing some work looking at how we support people who experience domestic violence or abuse, and how we might improve that support.  

During this work, I've gotten to know a fantastic local place that helps women who've broken free of domestic abuse to stay free.  Don't take my word for how great it is – here's what some of the women I spoke with there said about it:

• "Of all the services I've dealt with, this is the only one that works."
• "You are actually helped and understood.  They treat you as a person as a whole."
• "It's welcoming, and you are listened to.  It's like a family working together."
• "When you experience domestic abuse and leave the relationship, you don't know what you don't know.  They take you through it one step at a time.  They help with all the applications."
• "You get peer-to-peer support.  It's compassionate, and survivor led."
• "It's informal.  There's lots to be said for a cup of tea and a chat."
• "They pull out every stop for the children.  They helped get my child diagnosed for ADHD."

They don't turn people away because of thresholds.  No one completes an assessment form.  They don't work to targets, KPIs, or other metrics.  They build relationships with people, understand what really matters to them, and support them in achieving what matters.  The support is 'bespoke-by-default', responding to each person as an individual.  They have a network of friends and volunteers on hand with all sorts of skills – a handyperson, locksmith, experts on the law, housing, immigration, benefits, etc.  The support isn't time limited, and some of the women have been coming for years.  Many turn their experiences in to 'gifts' they can support their peers with, or have become volunteers themselves.

They've created a safe, welcoming, compassionate, and supportive atmosphere.  There's a real sense of community among the women – at times I struggled to work out who were volunteers or paid workers, and who weren't.   

And it works!  In five years none of the women they've supported have returned to the abusive relationship. 

How might traditional funding ruin it?

In my naivety, I asked one of the people who runs it "Do you get any funding – from the council or health or anyone else?"  The answer was a resounding "No".  The place is funded unconditionally by a charitable organisation, through local donations, and through the work of its friends and volunteers.  

I started to speculate on how this place might be ruined if they were commissioned or funded in the traditional and prevailing way.

• Traditional commissioning starts from an assumption that competitive markets work for public services, giving greater choice and efficiency.  To understand why this assumption is so flawed, read this book chapter from Kathy Evans, Chief Executive of Children England.  
• This means, in practice, the focus is on price.  To compete and compare on price, the service needs to be specified.
• Specification means a standardised service response.  This is where things really start to go wrong.  People are people shaped – they don't fit in to neat little boxes.  A standardised response cannot cope with the huge variety of people's need.  As Mark Smith writes in his blog, "[Standardisation] takes away an opportunity to apply judgement and nuance, borne from experience and an appraisal of what matters to whomever needs help"
• Once you standardise you have functional specialisms or silos.  Each service or provider is commissioned to only do specific things, and therefore won't do others.  The people who come to that service are looked at and labelled through the lens of whatever that service is commissioned to do.  Then they're referred or 'sign-posted' to the next service, who repeats the process again. 
• When providers are commissioned only to do certain things, they introduce thresholds or assessment criteria – essentially gate-keeping.  The thinking here is there's only a scarce resource available, so you need to ration what you do.  
• But, when people get turned away (or referred, or sign-posted), it means they either present elsewhere in the system, or come back later when their needs are higher.  Thresholds can be a false economy, as when people get worse they are more expensive to support.  
• Instead of building trust and relationships with their partners, commissioners and funders are typically interested in accountability (which is a nicer way of saying management by fear).  It means providers becomes concerned with external approval, which undermines intrinsic motivation and responsible practice.   
• To hold providers to account, commissioners use monitoring and performance management.  They ask for detailed information about how workers are spending their time, what activities they're doing, or if they are working to the standard.  A damaging example is 'time and task' commissioning in homecare services.  
• Another way to hold people to account is by setting numerical targets.  I've written about the problems with targets in a previous post.  They change the focus from doing the right thing to making the numbers look good, even if that means doing the wrong thing.  
• The undesirable effects of targets can be made even worse by payment by results.  They are a way to financially reward providers for producing data about targets, and not for helping the people they exist to help.  

Imagine what the place I visited might look like if it were commissioned in the traditional way?  What would it be like for the women who go there?  What would it be like for the staff and volunteers?  Would it be any more efficient?

Why do we fund and commission in this way?

As business writer Aaron Dignan puts it, "The way we work is broken.  It was invented 100 years ago on a factory floor for a world that no longer exists".  Commissioning is a symptom of these outdated ways of working, grounded in mass-production and 'Taylorism', which results in treating people as though they are parts in a production line.

In the 1980s, public services started to embrace what was subsequently called New Public Management.  This meant running public services as though they were private sector organisations, being more 'business-like', treating citizens as customers, and attempting to use competitive markets to reduce costs.   This led to policies like Compulsory Competitive Tendering (CCT), later replaced by Best Value, which are is still with us today.  

What are the alternatives?

There are many folks who know much more than me about how to improve funding and commissioning practices.

My go-to person for this is usually Toby Lowe.  He's an academic, who's previously been a provider of commissioned services, and has done loads of research in to this area with voluntary and public sector organisations. With his colleague Dawn Plimmer, he's put all of this into a brilliant report called Exploring the new world: Practical insights for funding, commissioning, and managing in complexity.  You can watch him speaking about it in this 13 minute video.

Andy Brogan, co-founder of Easier Inc, has recently written an accessible 'how' to report.  It gives lots of insight and helpful advice on how to put what he calls 'Commissioning 2.0' in to practice.

Jo Gibson from Vanguard has been doing some great work in this area too.  You can watch one of her students speaking about how they changed the way they commissioned homecare services in this 20 minute talk.  Jo is co-hosting a one-day commissioning event in London on 5 May 2020 which is worth checking out.

Over to you...

Do you work for a an organisation that provides commissioned or funded services?  How accurately  does what I've written here reflect your world?

Likewise, if you're at an organisation that commissions or funds services, are you still doing it in a traditional way, or have you started to look at some of the alternatives?

I'd love to read your thoughts in a comment below.  And please do feel free to share this with anyone who might be interested.  

Review: Beyond Command and Control (in people-centred services)

"Paul's life fell off the rails.  Over a period of 10 months, he was subject to 179 activity records by public servants, involving 91 staff from 20 different teams.  He experienced 12 assessments and 11 referrals, leading to six hospital admissions with total stays of 81 days, while staff generated seven support plans.  At no time did anyone spend time with Paul to understand what mattered and what 'success' might mean to him."

This is from the people-centred services chapter of John Seddon's new book, Beyond Command and Control.  It sums up much of what I've found when looking at these types of services.

I'm already a fan of John's writing, and have read many of his other books.  It was one of John's books that first got me thinking differently about improving services.

I've kindly been sent an advanced copy of the book.  Rather than covering all of it, I'm going to attempt to review the chapter on 'people-centred services', as it's an area I've worked with in local government.

What are people-centred services?

This is a term John uses to describe services that respond when "people's lives fall off the rails in a variety of ways". For example, services such as social care, domestic abuse, and homelessness are some people-centred services I've worked with.

They differ to transactional services as they usually require a relationship with the person first, in order to understand them and their situation.  How an organisation responds to "can I book a squash court" should be different to its response to "I'm struggling with my debt and my health".

Sadly, as John describes in this chapter, we too often see a transactional response when a people-centred one is needed, causing all sorts of things to go wrong.

What's wrong with the current design of these services?

The chapter highlights many of the problems with the prevailing "command and control" approach to designing and managing people-centred services.  Here are some of them.

• Targets. I've written about the problems with targets before. They're often set by politicians, hide the truth, and lead to sub-optimisation: "the time taken to get 'on target' is actually a reflection of the time it took public-sector managers' ingenuity to work out how to at least appear to be on target by gaming the system". 

In the example of care assessments, meeting targets "simply covers up the terrible truth: the extraordinary time it takes people to get through the red tape, mind-boggling number of people involved, the forms, assessments, reminders..." 

• Functional specialisms.  If you approach a council for help you "may be seen by a dozen or more people" each "considering your needs through their own specialist lens". 

I've seen this myself.  People get assessed and referred all over the place – like in a game of pass the parcel.  Each professional is only allowed to do their 'bit'.  All the activity going on here costs so much money, without really helping anyone.

• Standardisation.  A one size fits all approach which means "services inevitably fail to match the variety of people's needs, providing some things that don't help and others that go further than necessary".  Care services are commissioned on a "time and task" basis, meaning you get "30 minutes regardless of whether 10 or 40 might be better".  

This two minute video powerfully highlights many of the problems with this approach.  

• Managing demand. Or as John puts it "rationing by another name: limiting services, finding excuses to turn people away".  This is done in the name of reducing cost.  But when people's needs don't meet the criteria, they are left for the condition to get worse, meaning they cost more to support later.  

This was backed up recently by some superb research by Dr Rick Hood and colleagues.  It found children "screened out" of social care services are more likely to re-enter the system later on, with more complicated (and therefore costly) problems.  This is an example of failure demand, an important term originally coined by John in previous books.  

*            *            *

The chapter also explains a number of other flaws with typical people-centred services, including budget management, cost-cutting, commissioning, and a belief in market forces.  

Is there a better way?

Yes.  John describes a more effective design.  It starts with understanding demand and doing only the work that's of value to the citizen.  This consists of understanding the citizen and their need in context, helping them establish what a good life would look like for them, discovering what they can do for themselves to live a good life, and lastly looking at what support they may need from elsewhere.  

By focusing on doing only this, and cutting out the forms, signposting, remote assessment, rationing, standardised assessments, etc: "Lives get put back on rails... costs of services provided fall dramatically... and... as individuals and families are straightened out... overall demand begins to fall away".  

There are examples of where this has been achieved from a number of places, including a county council, a Swedish municipality, and some encouraging work done in Wales between a commissioner and care provider.  

In summary

This is an important read for anyone who works with people-centred services, or is concerned about increasing costs and demand in the public sector.  

It gives a crystal clear account of the problems with the current command and control design, valuable insights in to why that design is flawed, and demonstrates an alternative that's better for citizens, better for people working with citizens, and one that greatly reduces costs and demand.  

For a perspective of other parts of the book – especially the chapter on Agile – I highly recommend reading this review from Bob Marshall.

You can find more about the book, including how to order it, here.

How to speak with customer-facing colleagues

"Farming looks mighty easy when your plow is a pencil and you're a thousand miles from the corn field." 

– Dwight D Eisenhower

One of my favourite things to do in a service improvement role is spending time with customer-facing colleagues (by 'customer', I could also mean citizen, service user, client, patient, etc.)  I used to work in customer-facing roles myself, and miss being able to make an immediate difference.  I also find it incredibly useful for understanding how a service can be improved.

At the same time, I appreciate that some people don't enjoy the prospect of going to the coalface.

In this post, I'm going to give some reasons why you absolutely should spend time with customer-facing colleagues.  I'm also going to offer some advice and a few questions you could ask your colleagues.  These questions have worked really well for me, from a service improvement perspective.

I'm hoping that, by sharing this advice, some of you become a little less reluctant to go to the front-line, and feel encouraged to give it a go.  If you already spend time with customer-facing staff, reading on may help you get even more from the experience.

Why do it?

If you're not a customer-facing worker – for example, if you're a manager or work in a support role (IT, HR, Policy, somewhere in the 'corporate centre', etc) I'd say it's essential you spend regular time at the front-line.  I'd also recommend doing it if you are a customer-facing worker and want to learn more about another team.  Why?  Here are just a few reasons:

• The legendary business writer Tom Peters has referred to a conversation he had with Howard Schultz,  former CEO of Starbucks*.  It's a huge organisation, yet despite the inevitable bureaucracy, hierarchy, meetings, and reports, Schultz says he visits at least 25 Starbucks stores a week.  He apparently said that "...we still sell one cup at a time, one customer at a time, one server at a time.  I need to see it, touch it, and feel it".  Peters calls this MBWA (Managing By Wandering Around).  If you're from a Lean background, you might call this the equivalent of doing a gemba walk or genchi genbutsu.

• As the picture below says better than words could, your customer-facing colleagues are the ones who have the greatest knowledge of what matters to the customer, and what gets in the way of helping the customer.  They often have the best insights in to how you can improve service.  I always find colleagues respond well when I spend time with them and seek their views.  They especially love it when I bring a leader from the organisation along with me – they appreciate the leader taking them seriously, and the leader becomes more highly respected as a result.  If you're not getting these insights directly, you're missing big opportunities to improve your organisation and build important relationships with your colleagues.  

• Human Factors expert Steven Shorrock writes in this superb blog post about the 'four varieties of human work'.  I've summarised and simplified these in the quadrants below.  Work as imagined is what you think people do.  Work as prescribed is what the policy or procedure says people should do.  Work a disclosed is what people say about their work.  None of these represent work as done, what Shorrock describes as "the most important yet most neglected variety of human work".  And you're not going to get anywhere near to work as done by reading reports, staff survey results, process maps, or fancy charts.  You need to spend some time 'in the work' with your customer-facing colleagues.  

Questions to ask customer-facing colleagues

The following questions have worked really well for me, from a service improvement perspective.  I initially borrowed and adapted some questions recommended by the folks at Vanguard.  When I ask these questions, people tend to really engage, and their answers give lots of information and clues about how and where the service could improve.

• Where does your work come from, and how fit for purpose is it when it arrives?  This one is good to ask people who receive work from someone else.  Referrals, for example.  It gives you some clues about the quality of work they receive, and how much rework they have to do to get it right.

• Can you show me what you do with the work when it arrives?  As a general principle, it's best to aim for 'show' rather than 'tell'.  This gets you closer to work as done, instead of work as disclosed (see above).

• Why do you do it this way?  This may demonstrate the rules, policies, and procedures that staff are expected to follow.  Or they may reveal workarounds they have to do to get around constraints.

• What gets in the way of you doing a good job?  If you're only going to ask one question, ask this one.  I find people really open up to this one, pointing out many areas where they are hindered in doing what matters to the customer.  Keep asking "what else?" to learn even more.  

• What wastes your time?  Similar to the last one.  I often find I start to learn more about the type and frequency of failure demand in the answers to this one.

• How much, and how often?  A supplementary question for any of the above.  This helps you to start to quantify the waste in the service.  For example, I once found that around 60% of referrals received by one team needed to be reworked.  

• What numbers does your manager pay attention to?  This one may reveal the impact of targets or other arbitrary measures.  
  

• How do you know if you are doing a good job? You'll often find out here if people are clear about purpose, or (linked to the previous question) if the focus is instead on trying to make the numbers look good.

• How does change happen around here?  This will tell you how people feel, in terms of being involved and engaged with change.   

I recommend trying to ask these questions naturally in conversation – rather than looking like you are coming around with a clipboard and a survey.  It may be worth committing a few of them to memory first.  

And most importantly, when you ask these questions, listen!  As my friend Jo Gibson says, "give people a damn good listening to".  Remember you're here to learn, not to judge or defend current ways of working.  As Dale Carnegie wrote, "become genuinely interested in other people".  And demonstrate what Amy Edmondson calls 'situational humility' – make sure people know that you don't think you have all the answers, be curious, and emphasise that we can always learn more.

Over to you...

What are your thoughts?  What other reasons are there to spend time with customer-facing colleagues?  What can put people off doing it?  And what do you think of the questions I suggested?

Please feel free to comment below, or share this on Twitter or LinkedIn, or with someone who might appreciate it.

*Yes, I know Starbucks don't pay enough tax and put too much sugar in their drinks.  That's not the point here.  The point is that the CEO of an undeniably successful and huge organisation found it essential to regularly spend time with customer-facing colleagues and literally 'smell the coffee'.  

Do 40 points mean your football team is safe from relegation?

This post is about football (soccer) and variation of data over time.  I hope you're a fan of both!  If you're not, you're still likely to find at least some of what follows useful in the workplace.

In the English Premier League, it's a commonly held assumption that when a team reaches 40 points they will be safe from getting relegated to the next league down.

I recently read an article from the Guardian newspaper that challenged this belief.  It suggested around 36 points would mean safety.  If you're not familiar with this 'rule of thumb', I recommend reading the article before continuing.  You'll find it here.

Like at work, when people start throwing single numbers around, I started to wonder:  "how have they come up with that number?"  "have they understood the variation?" and "what would it tell us if we put it in a control chart?"  I did that, and here's what it looks like:

As you can see, it's a type of line chart, and I've plotted it as a time series.  Each dot shows the points needed to escape relegation (one point more than the relegated team) each season.

Like just about any data, there is variation.  In some seasons more points would have been needed than in others.  The three coloured lines on the chart help us make sense of that variation.

Average line

The red line in the middle is the mean average.  Although it's useful, there are problems when people only report on the average.  In my experience, the average becomes the number.  This is how they came up with 36 points in the Guardian article.  But the average doesn't take in to account variation.

For example, what's the average of 1 and 19?  And what's the average of 9 and 11?  The answer to both is 10, but the average doesn't tell us how much variation there is.  9 and 11 are much closer together than 1 and 19, but the average alone hides this information.   The same point is nicely made with this picture.

An average of 36 points (or 37 in my calculation) tells us nothing about how many points are needed to avoid relegation.  Roughly half the time more than 36 points will be needed, and the other half fewer will be needed.

Upper limit and lower limit lines

These are sometimes called the UCL (upper control limit) and LCL (lower control limit).

Data points going up and down between these lines are 'common cause' variation – the normal changes in the points needed for safety between seasons.  You shouldn't pay too much attention to the differences between these points.  They represent the 'noise' in the data.  The 43 points needed in 2002-03 is just as likely as needing 31 points in 2009-10.

A mistake I often see people make is to pay attention to common cause variation, and act as if something out of the ordinary has happened when it hasn't.

These lines also help with prediction.  If you want to be confident your Premier League team avoids relegation this coming season, 44 points should be enough.  And as long as you've got 29 points you've still got a chance.  Anything less than that and you're most likely doing down to the next division.

You might have spotted a couple of data points – 1992-93 and 1994-5 – above these lines.  That's what's called 'special cause' or 'assignable cause' variation.  It's a signal that something is different.  How you react to this would be different to how you'd react to common cause variation.  With special causes you'd ask "what's different?" or "why the change?"

In this case, there was a change.  For the first four seasons on the chart, the league was made up of 22 teams.  After then, it goes down to 20.  From 1995-96 onwards, teams are playing fewer games and therefore accumulating fewer points.

To show the change, the chart should really look like this:

Here are some handy 'rules' for spotting other signals in control charts.  This is where the average line becomes particularly useful.

Should football clubs set themselves a points target?

Probably not!  When you set a target, making the number becomes the focus, rather than doing the right thing – in this case, the right thing for the football club and its fans.  It can also encourage a behaviour where people ease off when the targets is met, or looks like being met.  

In the 2012-13 season, Barnet Football Club were relegated from their division.  Their then manager, Edgar Davids was quoted as saying: 

"It's even more disappointing because we have reached all the objectives that the chairman set and reached the 51 points target but we've still gone down."

They would have probably have fared better if they'd focused on winning as many games as possible, rather than achieving an arbitrary number.

How does this relate to work?

This is all well and good when looking at sports league tables, but this blog is supposed to be about work and improving services.  With that in mind, there are some lessons we can take away from this post.

1. Be suspicious when people quote a single number – often the average.  For example, I've seen the average time it takes to process benefit claims become the only figure used.  It was about 17 days.  Managers thought this was good performance.  Customers and stakeholders were given this figure, and they came to expect that's how long they'd be waiting for.   But a control chart revealed the predictable variation to be anywhere between 0 and over 100 days.  The average alone tells you almost nothing about performance.

2. Be careful not to confuse common cause with special cause variation.  I was once at a meeting where a department's figures were all 'red' because they were worse than the previous month.  The manager was asked to go away, investigate what had happened, and write a report to bring to next month's meeting.  This was a complete waste of time.  I put the data in to a control chart, and it was just normal common cause variation.  The senior managers had unwittingly reacted to is as though it was a special cause.  The next month they were back to 'green' – possibly because of regression toward the mean.

3. Comparing just two data points tells you almost nothing – certainly not about variation.  Although not covered in the above example, we see performance reports that compare this month to last month, now to this time last year, etc.  They might have arrows or colours applied, to indicate if performance if 'good' or 'bad' in relation to a target.  People are supposed to make judgements or decisions based on this information, with absolutely no context.  Displaying the same amount of information in a control chart would look like this:

If you took this chart to a meeting, people would probably laugh at you or tell you to leave.  Yet it's seen as perfectly acceptable to present the same inadequate amount of information in a 'scorecard' or 'dashboard' report.

4.  If in doubt, plot the data in a control chart.  Or at the very least plot it over time.  This post has hopefully made it clear that data has no meaning without context, and that you need a way to separate signals (special causes) from noise (common cause variation).  That's why control charts were invented!

5. Don't use numerical targets.  They're arbitrary and make performance worse.  If you want to know why, have a read of my previous post on the subject.


Further reading

• I cannot recommend Understanding Variation by Donald Wheeler highly enough.  It will change how you look at and interpret performance information forever.
• Samantha Riley from NHS Improvement has been doing some fantastic campaigning for better measurement in healthcare, highlighting the many benefits of control charts.  You can find her on Twitter @samriley
• Police Inspector Simon Guilfoyle used to write a superb blog, and also a book, on using the right measures, measured right.  This is one of my favourites of his posts.
• Lean expert and writer Mark Graban recently wrote an excellent article on how to React Less and Improve More by Using SPC More Effectively.  His recent 15 minute talk on the same subject is also well worth a watch.
• Vanguard have written a clear and concise guide to understanding and creating control charts (or 'capability charts' as they prefer to call them).

9 reasons why targets make performance worse

"When management sets targets and makes people's jobs dependant on meeting them, they will likely meet the targets  – even if they have to destroy the enterprise to do it" 

– W Edwards Deming

If you're in the UK, you'll be aware of the Windrush Scandal. The Home Office threatened the children of Commonwealth immigrants (mainly from the Caribbean) who arrived before 1973 with possible deportation if they couldn't prove their immigration status.

Something that contributed to the scandal was the Home Office setting "a target of achieving 12,800 enforced returns in 2017-18".  Much has been written about whether the targets existed, and if certain people in charge know about them. 

This post isn't about that.  Today I'm writing about the problems with targets themselves.  When organisations set arbitrary numerical 'performance' targets, which people or teams are judged by. 

Here are 9 reasons why targets make performance worse:

1. They narrow focus, so people concentrate on doing only what achieves the target.  This isn't the same as doing the right thing for the customer or citizen. 

For example, schools would give disproportionate attention to pupils expected to get a C or a D grade, as their target is pupils getting C or above. 

In hospitals, doctors were diverted from treating seriously ill patients to ones with minor problems in order to meet 4-hour waiting time targets.  As a result "people were left for hours covered in blood and without pain relief".   

Or Police didn't act upon warnings about child sex grooming because they were told by management to prioritise robbery and car theft, as that's what achieves the targets. 

2. People cheat so they meet their targets.  Studies have shown this.

Staff at an outsourced Police control room were making 999 calls at quiet times in order to meet their target of answering 92% of calls within 10 seconds. 

Confession time – in my formative years I did something similar.  I was working on a call centre, and we had an average call handling target.  Me and a couple of colleagues worked out that calling each other and quickly hanging up was an easy way to meet our targets.  We were using our ingenuity to meet arbitrary numbers, instead of helping the customer.

3. Cherry picking. I was working with a housing repairs service.  Tradespeople were set targets for how many jobs they get through in a day.  This created a behaviour called 'sponge knocking'.  If a job looked like it would take a long time – and risk not achieving the daily target – they'd silently push a 'sorry we missed you' card through the letterbox, and scarper.

The UK government's troubled families programme was reported to have a 99% success rate of turning lives around.  However, investigations showed some councils were picking the families most likely to meet the success criteria, and not working with the ones that needed the most support. 

4. People lie.  For example, traffic wardens were fabricating evidence in order to issue parking tickets, for fear of losing their jobs if they didn't meet their target.   

5. Statistics are manipulated.  In the brilliant US crime drama The Wire they use the term 'juking the stats'.  This is brought to life in the 1.5 minute clip below (contains swearing).

This isn't just the stuff of TV drama.  For example, juking the stats has happened in the UK Police force, and at hospitals.  

6. They cause student syndrome (when a student only starts writing an essay at the last possible moment before it's deadline).  When you apply for planning permission at your local council, they'll have a government target to meet of 8 weeks to decide your application.  The chart below – from one council I worked at – shows the large proportion of applications were decided in week 7 or 8. 

It might be possible to decide the application earlier, but as far as performance measures are concerned, it doesn't matter as long as it was done in under 8 weeks. 

As a different council found: "to meet these targets, applications would get refused if they were unacceptable and the deadline was approaching. [This] meant that the applicant suffered delay, additional costs and frustration.... The Council in turn incurred additional costs either defending planning appeals or processing a second planning application”.

I once did similar analysis looking at complaints, with a 15 day response target.  No prizes for guessing which day most of the complaints were responded to on.   

7. Staff get sick.  Targets put people under pressure to meet arbitrary numbers, which causes stress. Police Officers felt "almost continually under threat of being blamed and subsequently punished for failing to hit targets".  This does not create an environment that's good for people's health. 

I've worked with call centre staff who had average call handling targets to meet.  They hated not being able to help the customer because their manager wanted them to end the call. 

At a target-driven DWP call centre, mental health-related absences tripled.  At the same government department, staff went on strike because "We are being prevented from providing good quality service because of unnecessary and unrealistic targets."

8. People do unethical stuff.  Targets are an extrinsic motivator (coming from outside the individual).  Studies have shown that extrinsic motivators undermine intrinsic motivation. 

Sometimes people do the wrong thing in order to avoid punitive consequences.  Like in this DWP example, where the employee is clearly upset.  They want to do the right thing, but at the same time they need to survive in a bad system.  

There are other examples, such as PPI mis-selling, where people are doing the wrong thing because the target is attached to a bonus (financial incentives act like a turbo-boost to targets). 

Then there was the Wells Fargo fraud in the US, where staff at the bank sold customers things they didn't ask for, in order to meet targets.  This also shows how arbitrary numerical targets are.  When John Stumpf, the CEO, was questioned about why staff had a sales target of eight financial products per customer, he said he chose the number because "eight rhymes with great". 

9. Targets can kill people.  A study looking at targets in the NHS found patients were required "to wait in queues of ambulances outside A&E Departments until the hospital in question was confident that that patient could be seen within four hours [the waiting time target]".  In one tragic instance, this practice led to the death of a 16 year old boy. 

*                                                 *                                                  *

I'm not blaming the people who are working to the targets.  When a manager, government, or someone in authority sets a target, people will tend to presume it's right.  The Milgram experiment showed us that.

People focus on meeting targets to avoid getting grief from their boss, to make sure they don't get sacked, or to secure a bonus.  We all have bills to pay, and can't afford to risk losing our income. 

To quote again from Deming: "A bad system will beat a good person every time".  Or to quote from John Little: "Targets turn good people in to liars and cheats". 

My question to anyone reading this is – given all the points above – why do people continue to think it's a good idea to set targets?

What are your thoughts?  Please feel free to comment below, or share this post with someone who could be interested. 

9 things to avoid doing when studying demand

The people at Vanguard have written a lot about studying demand in services.  Why you should do it, how you should do it, etc.  They even invented the term 'failure demand' – so I'm not going to repeat what they say here.  For a short introduction to studying demand, I recommend this one minute read from Simon Pickthall.

Instead, I'm going to share a few things I've learned that are not helpful to do when you study demand.  I've either witnessed everything in the list below, or made the mistakes myself. 

1. Do all the studying yourself.  As mentioned in a previous post, the main purpose of studying a service isn't for analysis.  It's to help people unlearn and relearn.  The best way for them to do this is by studying the service from the customer's perspective.  And studying demand is a great place to start.   

I made this mistake more recently than I would have liked.  The demand analysis became just another slide in a PowerPoint presentation.   It lost all it's impact, and definitely didn't change anyone's thinking about the service.

2. Guess what the demands are.  Get a bunch of managers together in a room.  Do some 'brainstorming' and –  based on opinions –  identify the most common type of failure demand.  Put these in spreadsheet, with actions, due dates, etc, against each type of failure demand.

Or, decide the most common types of demand in advance.  Make them in to a tick sheet, and give it to the people who receive the demand to fill in.  

Studying a service is all about learning and discovery – it's not about trying to have all the answers.

3. Use existing data.  Run a report from your CRM, using categories already defined by the consultants who implemented the CRM.  This is similar to the previous point, where you shouldn't assume in advance you know what the customer is calling about.  The difference here is you've paid someone else to do the guessing for you. 

4. Be constrained by the existing rules.  I was once with a team listening to demand for council housing repairs.  One of the most frequent demands – particularly around midday each day – was "are you still coming to today's appointment?"  The appointment was for a tradesperson to arrive at their home and carry out a repair. 

The existing appointment slot was 8am to 1pm.  For this reason, most of the team initially felt this can't be a failure demand.  The resident was calling during their appointment slot, not afterwards, so surely we've done nothing wrong?
 
They were viewing this demand through the lens of the existing organisational system, and all it's constraints.  Yes, no person has done anything wrong.  But the system could be improved to stop this demand happening in the first place.  For example, you could shorten the appointment slot.  Or, as some councils have done, ask the resident when they want you do carry out the repair, and turn up then. 

5. Don't involve the people who receive demand.  When you're sitting with them listening to demand, don't explain why your there or what you're doing.  Don't show them what you're writing down. 

It's important to show your findings to the people who receive the demand.  It removes some of the mystery about what you and your team are up to.  It's also an ideal way to validate your findings – show it to them and ask them if this looks like a 'normal' day.  Listen to what they say about it. 

6. Fear failure demand.  Look for excuses to categorise everything as value demand, for fear it will look bad or demotivate staff if you find too much failure demand.  If this happens, it starts to give you some clues about the organisational culture.  

7. Ignore the type of service you are studying.  I've made the mistake a couple of times of studying demand in a people-centred service in exactly the same way I would in a transactional service.   I've learned from this mistake now, and it's something I plan to write about in a future post.  

8. Gather too much data.  I've had people on my team before who were concerned they needed to collect data on thousands upon thousands of demands.  Yes, you need to be somewhat scientific.  But not to the extent of randomised controlled trial or anything else that requires similarly high levels or rigor. 

When have you studied enough demand?  When the people you're working with have learned what they need to learn, and when the demand has become predictable.  This means you're no longer seeing types of demand you haven't seen before. 

9.  Do nothing else.  I've seen demand analysis be treated as a one-off exercise, done in isolation.   There may sometimes be value in doing this, but you're missing out on a fantastic opportunity to improve the service if you do nothing else.

It's important to next find out how the service responds to value demand, and does what matters for the customer.  You'll then want to learn why the service responds in the way it does.  

*                                                  *                                                  *

How about you?  Have you done or seen any of these mistakes?  Are there any others you've seen that aren't covered here?  Please feel free to comment below, or share this with someone who might be curious.