Product development in the land of the free - Simon Baker & Gus Power

Product Development in the Land of the Free 24/09/10

High-level, easily digested

Personal experiences working with different organizations, large and small in a variety
of domains

We often try to deal with the problem at hand

We seldom step back and question the constraints and assumptions

As an experiment we worked with a few organizations to completely change their
approach to product development

© Energized Work Limited 2010 1


If you’re tweeting, these are the twitter tags for the session



Our purpose is to get and keep customers by providing products they want with great
quality of service



Our goal is to delight customers

Who’s working on something right now that is delighting customers?



Our goal is not do delight shareholders

Making money is mandatory

But fixation on profit and obsession with costs is a sure way to become detached from
customers

Making money is a consequence of delighted customers

Deming said: Profit in business comes from repeat customers, customers that boast
about your project or service, and that bring friends with them



There are lots of people who could be considered customers + they have different
needs

[ product is a customer ]

[ 16x16 icon ]



We must balance our deliveries to meet their needs

We need customers to share a vision, have common goals + provide consensus on
priorities

This isn’t easy

** Different languages [ folk-it | business process engineering ] + different places +
busy



There’s aren’t many delighted customers out there

But we knew this



Change is all the rage

IT is awash with improvements



It still doesn’t feel like it’s any easier to get stuff done

How many people here are pair-programming?
[ pair-programming health + safety ]

[ 11th hr conf call ]

[ 6m/27k to move desks ]

[ case of beer ]



Look what might be needed to get something out the door

[ 20 ppl in a room ]



We’re over-organized

[ cio org on a whiteboard ]



We’re separated by specialty

It’s common to have core platforms + infrastructure

With centralized teams providing services like build, testing and release management

Centralization comes at the cost of complexity – dependencies, bottlenecks, waste

[ stakeholder meeting ]



It’s difficult to understand the big picture

When all we see is part of the product through a window in our silo

We end up chasing local targets + creating local optimizations

[ velocity over value ]

[ software != product ]



It’s just all too complicated

Conway’s Law: Organizations are constrained to produce designs which are copies of
their communication structures



Most changes tackle symptoms

The underlying problems remain

People go back to how they used to work like zombies back to the shopping mall

[ nhs attitude to turnaround teams ]

Dealing with complexity goes in the too hard box



The trend here suggests that size does indeed matter!

[ click ]

Complexity is not a setup for success



The unspoken truth is that failing conventionally is the status quo



So we asked ourselves

In the land of the free where we can do whatever we want

What would be the simplest operating model we could use to delight customers?

[ describe diagram ]

Product streams are like small companies, running their own business autonomously



SIMON:

If we were to attempt this change, small continuous improvements only get us so far

We know there’s a better place to start from, then lets start from there

GUS:

A radical overhaul allows us to remove endemic waste

How would we get support for such a dramatic change?



We seize a crisis or create one

We were invited in by the CTO to rescue a project that had been going for 18 months
without visible progress

Core platform off-shored to India



It’s crazy to make such a big change without first understanding the system –
structure, people, process …



Executives must acknowledge a crisis exists

Speak plainly using real data + talk in terms of risks + costs, capacity, throughput +
inventory rather than utilization + productivity

Provide options

Focus on a single product because it limits financial investment + contains risk



To sustain a big change we need executive support on tap

Everything we do from here is about building executive confidence in their decision



We have to make the change quickly before any doubt sets in

Moving at speed allows us to test whether their support is real or not

[ built a technical capacity + used a scheduled office move to collocate with business ]

[ removed unnecessary stakeholders and eliminated dependencies ]

[ removed people from silos + broke reporting links so that they worked on the product
full-time ]

[ ignored existing software and infrastructure ]

[ set up environments ]



Treat everything as a PR event

Get noticed by as many customers as possible

Start with an important and visible activity that gives them value

[ built designs out into html so they could start playing with the experience ]

We worked with the CTO to sell what was happening upwards and out



Clock is ticking so forget grand strategy

Make a splash with something small

[ sponsor’s head was full of world domination - shipped a public alpha in 9 weeks ]



Produce continuous results that keep customers engaged

Shipped public beta in 17 weeks + launched in 20 to the delight of various customers

[ business sponsor was happy to have product live ]

[ editors were happy with their new publishing system ]

[ PMO + finance people were happy we didn’t blow the budget ]

[ cto was excited by the new approach ]

[ permanent staff revitalized with renewed passion for their jobs ]

Don’t be surprised if other people are polarized



To build momentum flow must be protected

Work continuously to eliminate waste + remove sources of special cause variation

[ cag, release management. i+o war of attrition ]

Build in resilience by growing generalized specialists

Invest continuously to achieve repeatability through automation



Process + structure changes without changes in peoples’ attitude + thinking is just
posturing

New status quo sets in quickly



Change starts with ourselves

There’s are no new ideas …

Only new behaviors



How do we know customers delighted?

How do we know if we’re getting better?



We learned we must continually rediscover our customers

We must listen to the voice of customers

If we don’t we rely on our assumptions + we stop delighting them



We need to measure how we are doing in terms of our purpose, which is to get and
keep customers

Visualize the data

Analyze the results so we can understand and get knowledge about the system



The simplest thing is to canvas the mood of customers

It’s subjective but it’s a useful litmus test



Also watch for trends in product-specific measures

170M page views per month and 5M uniques per month



As an overall performance indicator we use a simplified profit and loss

Everything is a direct cost. No assets

There is only the cost of capacity

What we’re essentially doing is managing by cash-flow rather than by balance sheet

It doesn’t replace normal accounting



[ green is the throughput ]

[ red is the inventory, i.e. cards that are completed but are queuing for release ]

[ orange is the various stages of wip ]

[ cycle time measured from exit from backlog to release ]



[ first 9 weeks we built inventory then it was flushed ]

[ sobering to visualize the money sleeping in inventory and wip ]

Sleeping money is calculated from Cost per Card



Is a measure of the ability sustain throughput + minimize failure demand

[ failure demand is rework when something doesn’t meet customer needs – debt +
defects ]

[ defects fixed immediately ]



Use these measurements to listen to the voice of the system

Separate signals from noise + validate assumptions to stop speculation



Owned by people doing the work – they’re in best position to know how to improve
system

Not to let us set targets



We’ve been talking about 1 product

How does this scale?



At the organization level there are lots of projects

Identified different types of product stream were identified



Realized conventional PMO, being cost focused + project-oriented

Wasn’t setup to manage portfolio as system of products

It had no technical input

Budgeting was financial exercise + lacked operational reality + didn’t consider
customer demand

[ 10% budget reduction applied equally ]



What’s the simplest governance model we need?

Business director providing direction + realizing business goals + liaising with senior
stakeholders

Program manager dealt with organizational noise

Change agents – transformation, delivery expertise + technical input to investment
process

Financial controller



Purpose is to improve the operational + financial effectiveness of portfolio



Rather than being cost-based the hub focused on the flow of value, delivery of quality
+ the elimination of waste

See where costs are generated

Strategy was to 1st increase revenue, reduce costs retaining capacity, cut costs +
capacity



Provided safe environment + nurtured culture of continuous improvement

Encouraged a friendly rivalry between product streams

Didn‘t compete for investment

Helped each other solve problems + Shared improvements

Rotated people to encourage cross-skilling + propagate knowledge



Champion of effectiveness over efficiency

Resisted enterprise initiatives from the wider organization

[ enterprise ]



We’ve used PDCA to test our delivery capability

[ 80% of project risk is Market risk that your building the wrong product for the wrong
price or to a group of customers you can't reach ]

Everything is a hypothesis until it’s validated by paying customers

By implanting a rapid iterative delivery capability into day-to-day business operations

We’ve enabled business people to continually test business plans | marketing |
editorial content

Quarterly product chartering with incremental funding based on market performance



There’s increasing furore in this country about big IT failures, particularly in
Government

[ petition ]

So it seems the ideal opportunity to send this message


Product development in the land of the free - Simon Baker & Gus Power

More Related Content

Similar to Product development in the land of the free - Simon Baker & Gus Power

More from AGILEMinds

Product development in the land of the free - Simon Baker & Gus Power