# Averages: Is it mean to use the mean?

There’s an often-quoted anecdote about the flaw of averages which recounts the story of Lt Gilbert Daniels. In the 1950s, the U.S. airforce gave him the unenviable task of taking the physical measurements of over 4000 pilots with the aim of finding the ‘goldilocks’ set of averages that would lead to better design of fighter plane cockpits. Define the measurements of the ‘average man’ and you can design the average cockpit and most pilots will fit it.
Won’t they?
Apparently not.

## “Out of 4,063 pilots, not a single airman fit within the average range on all 10 dimensions.” Todd Rose

In any statistical analysis it is second nature to grab the mean average as the overview stat of choice. The ill-fitting cockpit, however, is a reminder that in many circumstances with real-world data, the mean average might not actually describe anyone or anything real.

This plays out in fundraising data all the time.
‘Could you give us the average gift from this activity, please? We want to see what people raise so we can try to make fundraising more effective next time.’

Well, the short answer is no. I can give you the mean average gift from a fundraising activity, and that information will help you forecast likely income from running such an activity again (assuming similar circumstances). It will not show you anything interesting about how people are behaving. Therefore it is unlikely you will be able to influence that fundraising behaviour in the future.

In these circumstances, rather than providing the single figure to explain a distribution, I encourage clients to look at the shape of the distribution itself. A histogram of donation amounts, or a graph of banded income will nearly always show a positive skew – a bulge at the lower monetary end and a long tail of rarer larger amounts at the upper end. Plotting amount given by number of givers looks like this:

Take the mean average and you don’t see the shape, you get a number in the middle. It doesn’t tell you that the bulge exists, where it is, and it doesn’t tell you that the larger amounts are outliers. A fundraising activity where most people give £10 could have an average gift of several hundred pounds due to a few people giving £1000, which could be hugely misleading.

Using the median (the indicator of the middle that literally points to the number half-way through the data) is less likely to be skewed by the outliers and therefore is more useful in showing a truer picture of average activity – what people are potentially capable of raising. As you can see, it’s closer to the bulge than the mean.

The mode (the indicator of the exact amount that the highest number of people gave) will show you the where that bulge is. It will often be useful to think about in terms of a psychological anchor that people are rooted to. A fundraiser can use this anchor to look at whether any of their marketing is driving people towards that anchor (by use of a prompt) or whether this is a window into a wider cultural anchor that it might be worth working with rather than against.

As with many things in the realm of insight analysis, the advice is to go into investigations with an idea of what you really want to understand. It is good to think about about what an average is telling you and what you might be able to do about it.

Are you creating a cockpit that will fit nobody? Is it better to get the measure of different people and have some different fits of cockpit, or one cockpit design that is adjustable?  In marketing terms could you be designing different messages for different audiences? Or might you design a general fundraising proposition that could be customised to different audiences?

One-size does not fit all and in some cases it fits nobody at all, whether that’s for cockpits or marketing messages. Don’t be mean – no supporter you’re talking to is merely average.

I very much recommend this article taken from Todd Rose’s book The End of Average.

A footnote on gender:
An interesting addition to the pilot story is the what happened with an equivalent discovery of the mismatch between average body measurement and real body shape in women. Whilst the pilot finding led to acknowledgement that there was no ‘average man’ and cockpit design needed to adapt accordingly, the interpretation of the finding for women was rather different. In this case, it was the opinion of experts of the time that the average ‘ideal’ for women’s bodies was not wrong, women were. They should get in shape quick-smart.

It’s never a simple question of deriving ‘fact’ from ‘data’, it’s the interpretation of meaning that counts. To that end, it matters who’s doing the interpreting, who gets to tell the story, and who faces the consequences of conclusions.

# Insight Analyst vs Software Developer: Puzzles

At home, my resident software developer and I discuss data, tech and coding among other more domestic topics. These other topics include: what we’re going to have for dinner, why the purchase of a new shed was never part of the plan of action prior to destruction of the old one, and whether or not socks can ever really be said to be ‘temporarily resting’ on the living room floor; a whole other blog is needed to tackle these questions.

When discussing technical matters it’s interesting to note what we agree on, and where our mindsets differ.

One Christmas, we’re at my mother’s home with my brothers and, as is customary in my family, we are lapping up various puzzle supplements from the Christmas newspapers. I alight on a Futoshiki and get stuck in, and then get stuck.

#### Futoshiki according to Wikipedia: The puzzle is played on a square grid. The objective is to place the numbers such that each row and column contains only one of each digit. Some digits may be given at the start. Inequality constraints are initially specified between some of the squares, such that one must be higher or lower than its neighbor. These constraints must be honored in order to complete the puzzle.

Got it? Right, so I’m stuck and I can’t find the next logical deduction in the sequence. I ask software developer (SD) if he wants to help by taking a look at the Futoshiki with me. Despite previous suggestions that slightly more maturity and slightly less borderline racism might suit him better, SD replies ‘I think you’ll find it’s pronounced Fucko-shit-o’. Moving swiftly on, we establish that SD is not interested in puzzles of this kind. He finds them frustrating and feels no satisfaction upon completion, concluding that the only thing that can be learned from puzzles is how to get better at those puzzles, something of no use to anyone. He does however enjoy coding problems, and therefore looked at our puzzle problem as a classic case of ‘if something is worth doing, it’s worth coding a computer to do it for you’.

Thus started the race. I was to continue to try to solve the puzzle I was stuck on and complete it before SD wrote a programme that could solve any and all 5×5 Futoshiki grids. I won in terms of speed, although I had to use brute force on the logic at one point which felt like cheating. Unable to make a deduction I built scenarios for each of three options, following through until sure that two would fail, therefore confirming the correct path with the third. SD won in terms of comprehensiveness. His code used the brute force of computational power to run through every option for every row in a puzzle until it found the solution, thus providing a generalised solution. From my perspective his solution completely misses the joy of working through a puzzle yourself, from his perspective he had all the same fun in solving his own which also resulted in a concrete workable thing at the end of it.

Neither of us need to stop there with Futoshikis. I can continue to come to each puzzle fresh, perhaps even repeating ones in order to reduce the number of deductions needed to solve it. I could also move on to solving larger grids. SD could also tackle larger grids and the task of generalising a solution that solved grids of any size. At this point, victory potentially swings back in favour of the human analyst. I should be able to use the same deduction techniques again and again, larger grids will take more time but will still likely be doable until such time as I become bored or die. SD on the other hand will have to radically alter the brute force approach. Whilst computational power elegantly outweighs human endeavour for smaller grids, the processing time needed for such an approach on larger and larger grids multiplies exponentially. This means at some point he will have a large grid that, while still solvable, would take all the time between now and the heat death of the universe for the computer to complete. To avoid this, SD would need to figure out a far more sophisticated algorithm, one that (arguably) mimics human reasoning to a greater extent; some sort of artificial intelligence. So perhaps next Christmas, as I enjoy bending my brain with my Futoshiki, SD will try creating an artificial me to solve all possible ones.

# Deconstructing the mathematical bridge

For music students everywhere, musical analysis is incredibly useful for many purposes. You can pick out different structures from a musical work, pinpointing particular styles and influences. You can learn to recognise and recreate a composer’s sonic signature in orchestration, rhythm, melody and harmony. Identifying and contemplating these features aids and increases the overall appreciation and understanding of musical works by providing information that connects us back through time to historical context, biography and musical purpose. And, importantly, these are aspects of music that you can hear.

When I was studying for a music degree, we were taught Schenkerian analysis, which I and my closest peers found abstract, boring and pointless beyond its performance as an intellectual exercise. I’m not saying it is pointless, I’m saying it felt pointless and reductive. The joke about any Schenkerian analysis is that what you do is take a great classical work (let’s say Beethoven’s 5th) and reduce it down to the kernel structure of Three Blind Mice.

How neat, how elegant, what a charming coherent unity it all has! Gah, I hated it! How can this possibly add to my understanding when I’m left with so much less than the sum of the parts I started with?

A tutor sympathised with my frustrations, recounting the story of the mathematical bridge in Cambridge. The bridge is hundreds of years old, originally designed by Isaac Newton. Very unusual in design it is made entirely of straight timbers arranged into an arch via some very sophisticated engineering. So ingenious was it that it held perfectly in place between two buildings by its unique balance of tension and compression alone. There it remained for many years until some inquisitive mathematicians wished to understand the design better. They dismantled the bridge with the intention of putting it back exactly as it had been. They failed. By taking it apart and not reaching a full understanding of how it worked, they were unable to restore it to its former elegance and today the bridge is held in place by rivets. Too much analysis can indeed spoil things.

As I start again in psychotherapy I find myself thinking of the mathematical bridge often. Will I discover a new and greater understanding of myself allowing for appreciation, healing and positive change? Or will I be reduced to some disparate sum of my parts that perhaps won’t add up to the whole I started with. Nobody wants to be Three Blind Mice after believing that they might be or have the potential to be Beethoven’s 5th.

What comfort then to discover that the story of the bridge is a myth? The original design (which was nothing to do with Newton at all!) always had rivets in it; there was simply a time when they could not be seen. The bridge has been dismantled and rebuilt twice allowing for maintenance.

There’s a bittersweet quality to myth-busting. The magic of the story that resonates so strongly has to be true, no? As Stewart Lee often ends his ridiculous flights of fancy: ‘This story is not true, but what it tells us is true’. The mathematical bridge myth reinforces our deep fears about self-discovery and change, helping us hide, that’s why we like it. It’s less romantic to go in search of the rivets to tighten them up a bit, but perhaps that’s what the bridge really needs to keep it in place.

# Painting pictures with data: Emma’s year of adventure.

How’s this for a year’s project?

THE RULES:

• Do something new each day

• Do it with someone else

• Document it

Pretty daunting, huh?

That’s why I’m not doing it. But Emma Lawton is. That’s incredible. What’s more incredible is that in 2013 at the age of 29, Emma was diagnosed with Parkinson’s. And, like so many people with Parkinson’s, she radiates a determination to focus on the things she can do more than the things she can’t.

Late last year Emma sent a call out at the Parkinson’s UK office to anyone who wanted to spend a lunchtime with her teaching her something. I wrote back and said she should come and geek out with me and Myuran over some analytics. She said yes. Hooray! A win for spreadsheets!

In prepping for Emma’s session, it was tricky to decide how to approach ‘analytics’ as a subject for an hour’s chat in a way that would do anything other than scratch the surface. It’s a vast topic that can easily get quite boring about stats if that’s not what a person is interested in. Instead of that, we did what we often do with clients and chose to focus on the person rather than the ‘data’. We took inspiration from Emma’s up front questions to us which were all about how important we regarded learning new things. Combining that idea with the knowledge that Emma’s own professional background includes design, we decided to play around with dataviz.

It turned out to be a fruitful starting point to ask Emma about her own project, regarding the year she was planning out and acting on as a potential dataset. She told us she had a planning spreadsheet that she was very proud of and would we like to see it (would we ever – hooray for spreadsheets!) Then we started asking questions. What did she really want to achieve by doing this? What kinds of stories would she want to be telling people? What pictures might she paint to illustrate those stories? What information might she collect that could form the material for those pictures if we thought of them as graphs or infographics?

## Analysis can be defined as the summarising and visualising of information for the purpose of gaining insight.

Whilst Emma is writing a blog post for every activity filled with rich qualitative information about her experiences and thoughts on each, summarising and visualising the project as a whole after a year is more challenging based on reminiscence and journalling alone. That’s where data capture comes into play. By noting down a few pointers in a set format for every activity in a spreadsheet (hooray!), she can build material she can later mine for patterns, patterns she could draw. She could capture almost anything about her activities, but based around the ideas she gave us that the project was about people and about how doing this stuff made her feel, we kept focus there. The data capture includes names and dates, information on the activity, how the person she meets is connected to her and various measures of how she feels about the activity including how new, challenging and satisfying it was plus the brainwave of a dropdown list of emojis for overall feel.

Is this reductive? Yes.
Can we capture the richness of Emma’s emotional experiences with dropdown lists of ratings and emojis? No.
Might it allow the viewing of wood rather than trees? Yes.
Might it reveal something of her experience that might otherwise remain hidden? Yes.

Meeting, speaking and sharing ideas with Emma was a joy and I felt that in that hour we explored in microcosm what we do on any project with any client. We meet people where they are and focus on what they tell us is important to them. We then work with them to think of how data can be useful as a lens through which to look, not to the exclusion of other lenses, but in addition for the provision of a different angle.

I can’t wait to see what Emma’s year ends up looking like to her.

Catch Emma’s analytics blog post at the fuck it listAnd do explore the rest of her adventures. Perhaps there’s even something she could come and learn with you?

# Why ‘I don’t have time’ is a lie to yourself and others

At work, time poverty is a lie, an illusion. In a company where everyone works the same hours a week, time is cancelled out as a factor in the productivity equation. Time doesn’t exist.

So what truths does that leave?

## ‘I don’t know what I’m trying to achieve’

Sounds simple, but your goals and objectives need articulating on paper, out loud and revisiting often.

Why are you at work? What are you doing there? What will be produced or delivered to prove you did something? What changes will have been made, how will things look, what will things feel like when you’ve done what you’re doing?

## ‘I take on more than I can do’

Workload is a real issue. A volume of work that is unrealistically high will hamper productivity. Knowing your objectives will make this one easier. Making conscious choices about what you say yes and no to is then possible.

## ‘I’m not clear on my priorities’

So you have said no to some things but still feel time poor for what you have on your list and you’re stuck on how to prioritise. Knowing your objectives also makes this one easier. When you know what you’re focusing on, you can rank different tasks, choose to work on the important ones. Urgent is not the same as important and you can choose to work through that issue too.

Admit and accept that when you say
‘I didn’t have time for that’,
you actually mean
‘That is not my priority right now, I chose this other thing instead’.

## ‘This is taking longer than I thought. I don’t actually know how to do this task’

Trying to complete tasks that you do not possess the skills, knowledge or experience to successfully execute is time consuming. And unnecessary.

• Identify what’s needed that you don’t have.
• Swallow your pride and ignore any inner voice telling you not to ask for help.
• Reach out and ask questions.
• Accept help offered, learn what you need to, or share the task with others.

## ‘This is taking longer than I thought. There’s likely to be a more efficient way to do this that I haven’t explored’

You do possess the skills, knowledge or experience to successfully execute your task. But the way that you’re doing it is time consuming. Open yourself up to new methods, or solutions for automating parts of the task you’re trying to do.

Creating efficiencies or automation itself takes an investment of time and effort. But that up front investment is repaid on every occasion you repeat the task and reap the reward of the time saved then.

Creating more time is an impossible problem that nobody can solve.
Good news is that you can solve any and all of these other problems.

# Who, what, why and Venn

Vacation is a good time to contemplate life and career, fulfillment and frustration. It’s a good time to assess if you have the right balance in your life, and, if not, what might be missing.

And what better way to make that assessment that using the favourite of data visualisers everywhere – the venn diagram.

I enjoyed examining this and relating it to my own situation. I even found myself unpicking some of it: aren’t mission and vocation the wrong way round? Depends on your understanding of the semantics I guess. Also, what does it mean to be excited at the same time as complacent and uncertain? What kind of people actually find themselves occupying this space? It reminded me just how powerful venns can be for exploring concepts.

What I loved the most was trying to decide the tweaks I can make to draw me closer to the middle. For me, it’s allowing myself to ensure that I continue doing the things I love, and not be drawn too strongly out of the LOVE circle and into that space between profession and vocation. I think this happens to me sometimes. In future I’d like to remember to make choices that are enriching for me too, rather than always fall into the ‘pleaser’ habit of prioritising what others are asking me for.

# Data vs Insight skillsets

Growing any analytics or insight function requires hiring the right skills for the jobs. Those jobs include data jobs, analysis jobs, and various shades of soft skill jobs like stakeholder development, project management, communication, influencing, storytelling and the list goes on.

Whilst there are some unicorns out there whose talents straddle all things techy data as well as business knowledge, insightful thinking, and fabulous communication skills, these people tend to be rare and expensive. We’re all in search of the ‘geek who can speak’!

Teams need the right mix and any strategy for growth must assess what, and therefore who, is needed.

In thinking through this recently for our own planning I found myself drawing up this skillset comparison. Before anyone gets insulted, I’m not saying people on one side can’t do the things on the other. But it is my opinion that most people will fall more strongly one side or the other. I myself am firmly on the insight side of the equation. I do have strong technical skills and could even code up a model if I had to (probably!) But the work overall benefits from me delegating tech tasks to someone with a strong data skillset and concentrating my efforts on the insight tasks.

It’s tough to know how best to develop yourself in this sphere. It’s hard to keep developing in depth in all areas. I often feel I want to increase my techy skills but it’s equally important that I develop in line with my strengths which are for planning, project management and people stuff.

# Quantifying personal transformation: the altMBA

I’ve just finished my run at the altMBA. I recommend it!

I had high expectations going in. In fact, nothing sort of transformational would have met those expectations – especially for the money! (Although I was very fortunate to get a contribution from my organisation – thank you Parkinson’s UK.)

## What happened?

The stats
4 weeks
~111 students in my full international cohort
~25 students/coaches in my timezone cohort
~ 4-5 students per learning group – assigned to a learning group per week for discussion.
17 channels on slack where we discussed topics and supported each other
13 learning group meetings
13 projects shipped (more if you were mad enough to do extra!)
13 reflections on what was learned from those projects
~70 comments made by fellow students on my work to help me learn
5 1/2 books read (7 were sent so I’ve still got some work to do there)
~1000* online articles read and videos watched, countless more on my follow up list
3 timezone cohort meetings online
1 special online workshop
1 meet up in person with a fellow student
1 coaching session 121 where I cried because things were so challenging to manage
2 eyes opened to my own ideas, motivations, strengths, weaknesses and blind spots
(*I’m not going back to count them, so this might well be an exaggeration…)

Items I can’t count
New connections made, friendships forged, messages sent
Ideas, experiences, vulnerabilities, stories, troubles and successes shared

Areas covered
Thinking frameworks flexible enough for work and life
Exercising the creativity muscle – in business and life
The power of empathy and sonder – in marketing and life
Embracing vulnerability
Getting stuff done to deadline, no excuses
Working with others generously, but not letting them off the hook
Giving and receiving feedback well
Thinking big
Identifying the barriers and laying plans to do the hard parts first
Much, much more – it would take a book to flesh out the things I learned

Value gained
Immeasurable and keeps growing every day with practice of the learnings

## What now?

I wanted to feel changed by this experience, and changed I feel. As well as tired, stunned, nervous for the future and excited for the future. A little time is needed to digest everything that has just happened. But then it will be back to work with new energy, new optimism, new ideas and a new approach.

# Tell your story! Easy to say, hard to do. Here are the tips that help.

Storytelling is vital. Whether for making data accessible to an audience or for any other reason.

We all know this, we’re told it all the time, mostly by Seth Godin! We’re surrounded by stories and we all realize that to create connection, to persuade, to make change, we have to tell our own. And to get anywhere, we better get good at it. So why is it still so damn hard to do it well? How come I forget how to do it every single time I come to write?

‘What do I want to say?’, ‘what product do I want to tell people about?’, ‘what’s my opinion on this?’
Working at a charity, we fall into this internal thinking trap all the time. We have so many messages we want people to hear, so many important pieces of news to impart, so many actions to invite people to take. And for communicating insight we get caught up asking ‘what do my findings say?’, ‘what my recommendation?’, ‘how can I make this graph just a little bit prettier?’

But to start with the external means to think of the reader first. Who are they? What might they be interested in? Why should they care about what I’ve got to say?

Every story needs to start from the outside and go in. Every story is a fruit that the reader peels back the layers of to get to the juicy, tasty bit in the middle. If you’re like me and need a helping hand offering up the best fruit then look no further than Bobette Buster’s handy how to:

It’s all gold but here’s the best bits that are easily actioned every day:

1. Tell your story as if to a friend, no matter who you’re talking to.
2. Choose a ‘gleaming detail’, the one ordinary moment, object or metaphor that embodies the heart of what you’re telling a story about. Hook onto the senses: what does that detail look, sound, smell, taste or feel like?
3. Be vulnerable. Dare to share the doubt, the anger, the surprise the joy.
4. Bring yourself. It doesn’t have to be about you to include you or your opinion.
5. Let go. Less is more.

# Insight to Outcomes

This is a new outline that I and my analyst colleague Myuran have been working on as our department goes about their yearly planning cycle.

In setting up our insight function from scratch, one of the challenges I’ve found is how to properly evaluate the work and its impact. Often we have to chase our clients for feedback on what we delivered and ask them what they used the information for. It often takes time for insight to reap benefits, so it can be hard to measure tangible value early on.

However, this year we’ve drawn up this mapping to help us demonstrate what we achieve via different pieces of work and what the results feed into to. We aim to use this dataviz to help as context in future briefing and get clients focused on the decisions they want to make. Also, our client teams can use this themselves to map out the data-driven decision-making that they are taking care of themselves without coming through the insight team. That way we can demonstrate the ways that the department is becoming more insight-driven in its decision-making.

We’ll start using this more pretty soon and will no doubt find out if we’ve missed something!

(dotted lines around audience attitudes indicate that we don’t really do a lot of qualitative research ourselves at the moment, our insight team is quant based but we know that qual is important for the fuller picture on audiences).