• View
• Edit
• History
• Print

Back to books

Ch 2, Development is inherently different from production

The natural people manager realizes that uniqueness is what makes project chemistry vital and effective.

We pay far too little attention to how well each fits into the effort as a whole. The catalyst is important because the project is always in a state of flux. Someone who can help a project to jell is worth two people who just do work.

Ch 3, Work smarter

Trying to get people to sprint too much can only result in loss of respect for the manager. The best workers keep silent and roll their eyes, the others are workaholics. The most dedicated workaholic eventually realizes he has sacrificed a lot for less value and will feel devastated. He seeks revenge and just quits, another case of burnout. Everyone has a bout with workaholism now and again. The loss of a good person isn't worth it.

The process of improving productivity risks increased turnover. People under time pressure don't work better, they just work faster.

Ch 4, Quality

Managers jeopardize product quality by setting unreachable deadlines. The decision to pressure people into delivering a product that doesn't measure up to their own quality standards is almost always a mistake. Self-esteem is strongly tied to the quality of the product.

By regularly putting the dev process under time pressure and then accepting poor-quality products, the software user community has shown its true quality standard. Allowing the standard to be set by the buyer, rather than the builder, is what we call the flight from excellence.

Quality, far beyond that required by the end user, is a means to higher productivity. Quality is free, but only to those who are willing to pay heavily for it.

Ch 5, Parkinson's Law

Parkinson was a humorist, not an engineer. His law was a funny dig at big bureaucracy, which gives little job satisfaction. The law almost certainly does not apply to your engineering team.

Study: Projects on which the boss applied no schedule pressure had the highest productivity of all. The decision to apply schedule pressure is analogous to punishing your child. If your timing is impeccable so the justification is apparent, it can help. If you do it all the time, it's a sign you have huge problems.

A company itself is the true Parkinson's Law subject: organizational busy work tends to expand to fill the working day, which is why very mature companies are less fun to work for.

Ch 6, Software management

Myth: changing languages will give you huge gains.

A manager's function is not to make people work, but to make it possible for people to work.

There are a million ways to lose a work day, but not one way to get one back.

---

Ch 8, Overtime

Disturbing possibility that working overtime can only increase quality, but not quantity. To be productive, people may come in early or stay late or stay home. This is a damning indictment of the office environment.

There is a 10 to 1 difference in productivity among organizations. Many companies provide developers with a workplace that is so crowded, noisy, and interruptive as to fill their days with frustration. The top performers' space is quieter, more private, better protected from interruption, and there is more of it.

In the long run, what difference does it make whether quiet, space, and privacy help your current people to do better work or help you to attract and keep better people?

Ch 9, Saving money on space

The entire cost of work space for one developer is a small percentage of the salary paid to the developer. Space cost varies in the 6 to 16% range. With benefits and salary, you might be paying 20:1 ratio in the cost of worker to his workplace. Attempts to save a small portion of the one dollar may cause you to sacrifice a large portion of the twenty.

Open plan evangelists have resorted to proof by repeated assertion. No real data supporting the effectiveness of such plans has ever been presented, because they can't find any.

Ch 10, Use of time

One study suggests 30% of work day is working alone, 50% with one other person, and 20% with multiple people. The 30% alone time is when work is actually accomplished, during which the developer is in flow. It takes a while to get into it, and if interruptions are constant no real work will get done. "Your people bring their brains with them every morning. They could put them to work for you at no additional cost if only there were a small measure of peace and quiet in the workplace."

Ch 12, Interruptions

Workers don't care much about glitz. As long as the office isn't depressing, money spent on high-fashion decor is a waste. All too often, appearance is touted over functionality. Does the work place let you work, or does it inhibit you? Work-enabling office space is not a status symbol, it's a necessity you'll pay for either way. Either you shell out what it costs (bargain), or you pay in lost productivity (bad deal).

Management, at its best, should make sure there is enough space, enough quiet, and enough ways to ensure privacy so that people can create their own sensible workspace. Uniformity has no place in this view.

Ch 13, Order

People cannot work effectively if their workspace is too enclosed or too exposed. You feel more comfortable in a workspace if there is a wall behind you. There should be no blank wall closer than 8 feet in front of you. Your workplace should be sufficiently enclosed to cut out noises which are a different kind from the ones you make.

Give each group a place where people can eat together. Make the common meal a regular event.

Successful space allows one to function as a human creature, emphasizing his essence and not denying individuality nor inclination to bond into a team.

It almost always makes sense to move a project or group out of corporate space. Work conducted in ad hoc space has got more energy and a higher success rate. People suffer less from noise and interruption and frustration. The quirky nature of their space helps them form a group identity.

---

The final outcome of any effort is more a function of who doe the work than of how the work is done. Yet modern management science pays almost no attention to hiring and keeping the right people. Here's the correct formula: get the right people, make them happy so they don't want to leave, turn them loose. For most efforts, success of failure is in the cards from the moment the team is formed and the initial directions set out.

Ch 14, People

Getting the right people in the first place is all-important. Oddly enough, most hiring mistakes result from too much attention to appearances and not enough to capabilities (march to the norm). The need for uniformity is a sign of insecurity on the part of management. Strong managers don't care when team members cut their hair or whether they wear ties. But organizations have the tendency to trend toward uniformity, which is why big/old companies are more boring than startups.

Ch 15, Hiring

What could be more sensible than asking each candidate to bring along some samples of work to the interview?

Maybe you should use an aptitude test but hire only those who fail it; you should use them, just not for hiring.

The business we're in is more sociological than technological, more dependent on worker's abilities to communicate with each other than their abilities to communicate with machines. We suggest auditions. Have the candidate prepare a 10-15 minute presentation about some past project. Audience is potential co-workers. Hold a debriefing afterwards. You can find out this way if someone is naturally able to explain simple or complex ideas. This helps team morale as it removes more dumb luck from the process.

Ch 16, Turnover

Average employee longevity is 15 - 36 months. Average person leaves after a little more than two years. Cost to hire a new employee is about two month's salary or just under.

New employee timeline:

• 2-3 months = doing some useful work
• 5 months = full working capacity

Turnover feeds on itself. Companies with hugely high rates have these probably causes:

• co-workers are just passing through, no sense of long-term commitment
• management treats people as disposable, interchangeable parts
• no loyalty because of the organizational view

Ch 17, Methodology

When the work becomes completely deterministic, you will kill a team of human workers. The project's workers are familiar with the effort, and if a given direction doesn't make sense to them then it doesn't make sense at all. Big Methodology is an attempt to centralize thinking, and it implies that the workers aren't trusted to think. Here is what the fixed mold will bring you:

• a morass of paperwork
• a paucity of methods
• an absence of responsibility
• a general loss of motivation

Malicious compliance is the idea of working to the rule exactly, meaning doing it all by the over-burdening book. This is a form of protest that causes simple efforts to drag on forever. This happens where there is too much big Methodology.

It turns out that people perform better when they're trying something new, and this seems to occur almost regardless of the type of change. This is called the Hawthorne Effect. Most productivity gains for new-fangled management methods can be attributed to this principle. No one studies ten-year-old improvements to see if they are still worthwhile, because they probably aren't. Take advantage of this by making non-standard approaches the rule.

---

Challenges focus a team, but that's not what you remember or enjoy. That's the coming together. People work better and have more fun when the team gels.

Ch 18, Team synergy

Workers do not automatically accept organizational goals, they often seem arbitrary. Management seems to rely on "professionalism" for marching to the company tune only at the bottom where the real work is done.

In one example, a pep talk to a motivated, overtime-working team to focus their efforts caused them to all go home immediately. The topic was the importance of the team's efforts for hitting a record quarter financially. This made success seem trivial.

Teams don't attain goals, people do. The purpose of a team is not goal attainment, it is goal alignment. A more directed team is a winning team.

Signs of a gelled team:

• low turnover during the project
• strong sense of identity
• sense of superiority
• joint ownership of the product
• enjoyment of the work

Ch 20, Teamicide

You can't really build a team, there is too much outside your control. There is no recipe. You can grow a team, but it's an inexact science. Growing plants is the best analogy.

How to kill a team:

• defensive management (lack of trust, gating at every step, too much process)
• bureaucracy and paper work
• physical separation
• time fragmentation and context switching
• low-quality product due to cost-cutting
• phony deadlines
• clique control
• continual overtime

On deadlines, what the boss is saying is he doesn't believe the team will do any work unless under duress and has no respect for his workers.

The idea of teams at upper levels of management is a myth, as teams only exist at the bottom of the hierarchy. Managers are only bonded onto teams when they are part-time contributors/peers in the group. The higher you go, the less the team concept is a coherent one.

Ch 22, Team-oriented Management

"A job situation that hurts your self-regard is itself 'sick'". The person who calls in well is ready for work that enhances self-regard. Assignments will be an acknowledgement of competence and allow some autonomy and responsibility. You take a chance as a manager putting someone in that spot, and they may make a mistake you wouldn't. Too bad, you try to get the right person in the right place and then you can't second-guess. A person you can't trust with autonomy is of no use to you. This is the opposite of defensive management.

A boss putting his reputation into others' hands is empowering and brings out the best in everyone. If you've got decent people, the best way to improve the chance of success is to get out of the way sometimes. You could even send them away to a mountain cabin to work on a design for a weekend. Visual supervision is a joke for development workers, save it for prisoners. If you have a distracting office space, use any excuse to get your people out of it.

Here's an idea - give the workers some say in team selection by having them put together their experiences to bid on company projects. This is a similar effect to having the whole team sit in on interview-auditions.

Optimal chance to gel = everyone is respected for their natural authority, the thing they are good at. No forced authority.

Ch23, Chemistry

A cult of quality can help give a team something to rally around because it sets them apart. The trick is protecting them from the short-term increased cost of achieving it.

As a manager, you need to get each team member pulling in the same direction. Elite teams can conform to the norm in some areas, as long as there is something that sets them apart.

Try adding some flavor to your team to give the signal that it's ok not to be a clone.

---

Ch24, Chaos and Order

Present day standardization has achieved documentation consistency among products, but nothing approaching functional consistency.

Ch25, Free Electrons

Consider the value of self-motivated people who are given the freedom to define their own job responsibilities. They are inclined to do everything they can to help the company profit from their efforts. The alternative is that they head off and do their own consulting, having many more freedoms than your employees. This is the idea of the intrapreneur, and they are more loyal to the company than other employees.

The individualism valued in Western culture seems to disappear in the workplace too often. Some need structured direction, but those who don't should be left alone.

Ch27, Teamicide Revisited

Overtime that drags on too long (months) accentuates the differences in the team members in terms of how much each can borrow from his life to put into the team effort. This will kill the team. Managers use the demand not because they think it will increase productivity in the long run, but because it is a shield to put up against blame when the work does not finish on time.

Ch28, Competition and Coaching

Conflicts within the team over power and status will cause problems. Given the diversity of the work, most coaching is now done as peer-coaching and the manager must enable this by giving enough time, attention, and respect to each team member to limit negative competition. Well-knit teams experience regular back and forth knowledge transfer. We feel grateful to those who have coached us in the past and we pay it forward. But this will not happen in an environment in which the engineers don't feel safe. In the wrong setting, coaching or being coached is a chink in your armor to be exploited.

Teamicide by internal competition is done as follows:

• annual salary/merit reviews
• management by objectives (shared goal with worker input; measured results)
• praise only for certain workers
• awards and bonuses tied to performance
• actually, almost any kind of performance measurement

The point is, that any action taken to reward engineers differentially is likely to foster competition and managers must work to counteract this instead of using these measures as a crutch out of sheer laziness. The proactive manager instead uses investment, direct personal motivation, thoughtful team formation, staff retention, ongoing analysis of work procedures.

Ch 29, Process Improvement

Organizations that build products with the most value to their customers win. Those that build products that make the world yawn lose, even if they build them very efficiently. Process isn't worth a rip unless it's applied to projects that are worth doing. If process improvement is the goal, then scary projects get shifted to the back burner - but those are probably the projects worth doing. Real benefit brings real risk along with it.

The projects most worth doing are those that will level you down one notch on the process scale.

Ch 30, Change

Resistance to Change Continnum:

1. Blindly loyal
2. Believers but questioners
   1. skeptics
   2. passive observers
   3. opposed by fear of change
   4. opposed by fear of loss of power
3. Militantly opposed

The fundamental response to change is not logical, but emotional.

[Old status quo] + foreign element --> [Chaos] + transforming idea --> [practice and integration] --> [new status quo]

Ch 31, Human Capital

An expense is money used up and gone, like a heating bill. An investment is capitalized asset. Six months is often cited as the time needed to bring a new developer up to full integration in the team. However, for more specialized work, it could be about two years.

Wall Street applauds downsizing. Remember, the object of the exercise is upsizing, not downsizing. Layoffs are a sign that upper management has blown it. Companies that treat their people like as an expense instead of an asset suffer in the long run, and investment in human capital matters the most.

Ch 32, Organizational learning

Experience gets turned into learning when a company alters itself to take account of experience. Learning is limited by an organizations max ability to retain workers. With high turnover, learning is stilted or can't take place at all. The most natural learning center for most organizations is in middle management. The top is too busy focusing on mergers, acquisitions, branding, boards and stockholders. The bottom has no visibility or power.

Flattening the org chart by gutting middle management will certainly decrease learning. But for vital learning to occur, middle managers must communicate and work together, and this is very rare. Organizations in which these folks are isolated, embattled, and fearful will not be able to learn.

Ch33, Ultimate sin

Wasting people's time!

"Status ceremony" = chit chat, then go around the room, everyone reports on status individually, no engagement with peers. Status could be acquired in a less wasteful way, but this is about reassuring the boss and nothing more. Ceremonies to celebrate the bossness of the boss are a waste.

Ideal staffing curve: Attach:staffcurve.jpg

Overstaffing is a common problem too early in a project, and it's not designed or coordinate enough not to waste worker time. This is due to a time constraint or deadline and then someone blindly throws more people at a project to try and accelerate it, because that looks better than being wise about it.

Don't waste the time of your biggest investment, your people.

Ch34, Building community

What do great managers do best? Build community. A satisfying community will keep people. With investment in people, the people feel better, perform better, and don't want to leave.