Seth Godin wrote a recent post about layering. Here’s an excerpt:

Someone asked me which post on this blog represented the turning point of its growth. The ‘breakthrough’ post. It turns out that there wasn’t one. Instead, there were 2,500 posts, one after the other, each building (and I was learning from each) as we went.

This reminded of Jim Collins’s analogy in Good to Great. He likens success to spinning a flywheel. Picture a massive metal disk, 30 feet in diameter and weighing several tons, mounted on an axle. You don’t spin it fast with one single heave; it takes many smaller shoves. It takes a lot of hard, sustained effort to get it up to speed, but once it’s moving it’s virtually unstoppable.

There’s a great chapter on hiring in Peopleware, by Tim Lister on Tom DeMarco. It starts by demonstrating how most managers would hire a juggler:

Circus Manager: How long have you been juggling?
Candidate: Oh, about six years.

Manager: Can you handle three balls, four balls, and five balls?
Candidate: Yes, yes, and yes.

Manager: Do you work with flaming objects?
Candidate: Sure.

Manager: …knives, axes, open cigar boxes, floppy hats?
Candidate: I can juggle anything.

Manager: Do you have a line of funny patter that goes with your juggling?
Candidate: It’s hilarious.

Manager: Well, that sounds fine. I guess you’re hired.
Candidate: Umm…Don’t you want to see me juggle?

At Red Gate, we heed the point that the book goes on to make. In an interview, we test the skills that people claim they have. If we’re hiring for a developer, we ask them to write code. Testers test something for us, sales people sell us something, we ask designers to design, and so on.

There is a chasm between the abstract and the concrete; between the meta and the specific. Ask somebody applying for a sales job how to sell something and they’ll speak fluently about finding out the customer’s needs, how to handle objections and how to close the deal. Ask them immediately afterwards to sell you a computer and they’ll jump straight in and try to sell you an iMac with not even a nod to finding out why you want a computer. No matter what they say should be done, most designers ignore users, testers don’t test boundaries and marketers don’t segment.

There is, however, a limit to this technique. It only works well when the essence of the job can be distilled and presented as an interview task. This fails when the job is more ill-defined; more amorphous. Project management, say. You can ask someone to code in an interview, but can’t ask somebody to project manage. You can’t present them with a group of people and a project behind schedule and say "here, project manage this for half an hour". You’re inevitably reduced talking about how they’d handle certain situtations, or how they handled them in the past, not actually doing the task.

I can see a few solutions to this problem:

• Avoid the problem. Always hire for these roles from within. This has the disadvantage that you often end up moving talented technical people into roles they’re not suitable or ready for. Hiring people externally is a good way to bring in fresh points of view. If you always appoint from within then you risk an in-bred culture / process.
• Accept the problem. Do your best, but acknowledge that many people you hire won’t work out and you’ll have to fire them. The disadvantages – commercial and human – to this approach are obvious.
• Only hire based on personal recommendation. If somebody you trust can vouch for the person you’re hiring then that removes a lot of risk. However, it also reduces the pool of people you can hire from. This is a severe restriction.
• Insist that the people you hire are hands-on. This isn’t a complete answer, but I hire people who have something to offer than pure management. They need to be able to demonstrate they can roll up their sleeves and contribute directly. I don’t hire people who stay aloof and refuse to engage directly with the tasks their team do. Project management isn’t (just) about Gantt charts; people management isn’t (just) about personal development, performance reviews and team meetings. These hands-on skills can be tested for, but although they’re necessary they aren’t sufficient.

However, I don’t think these solutions are acceptable. If you’ve got any better suggestions on how to hire project managers / development managers / support managers etc. then post them here.

Enjoyed this post? Subscribe to the RSS feed.

37signals do some cool stuff (Ruby on Rails and Basecamp, for starters). Over 1 million people use their web based applications. For me, perhaps the most interesting thing about them is their ethos. In particular, I like their forceful emphasis on design and simplicity.

37signals have a book – actually, it’s more of a manifesto – where they explain their principles and practices. Here’s a quote:

Your app should take sides

Some people argue software should be agnostic. They say it’s
arrogant for developers to limit features or ignore feature requests.
They say software should always be as flexible as possible.

We think that’s bullshit. The best software has a vision. The best
software takes sides. When someone uses software, they’re not just
looking for features, they’re looking for an approach. They’re looking
for a vision. Decide what your vision is and run with it.

You can read more from the Getting Real web pages.

Jason Fried, the founder of 37signals, has kindly agreed to speak at Business of Software 2008. I’m looking forward to hearing Jason speak. In particular, I’m interested in whether his principles are generally applicable, or whether they’re specific to 37Signals’ setup, culture and software. Do they apply to incumbents or just the to the upstart start-ups who want to topple them?

You can find out more details about the conference at the Business of Software 2008 web site.

Paul Graham tells us about the top 21 existing trends you should bet on…and the five you shouldn’t.

http://businessofsoftware.wistia.com/medias/eg2ltpi8yk?embedType=async&videoFoam=true&videoWidth=640

I came across the following quote the other day:

    Those who say it can’t be done should get out of the way of those who are already doing it.

Unfortunately people don’t always follow this advice. I was about to flip it round and write something about success, but if you want advice about success there are better people to listen to than me.

I do, however, know something about trying, and trying is often harder than succeeding: the leap between the status quo and trying to change it is often wider than the leap between trying to change it and succeeding. Often the opinions of others – especially those who we respect – can be paralyzing (what if they’re right? what if I’m being stupid? what will they think if I fail?).

Just to try, you need to listen to what other people say and then ignore them. Of course, this isn’t always true. And it’s not clear when it holds, except in retrospect.

I’m a sucker. Last night I went to the cinema for the first time in a while. The queue was huge. I was cunning and called the automated booking system from my phone. Its voice recognition system left me standing in the street shouting "I said Cambridge you motherf***ing piece of s**t" into my phone. The popcorn was twice a reasonable price and the place stank. The film wasn’t even that good. But that might have been my foul mood, or because my first choice had started before I could book tickets. Everything about the whole experience sucked, as it always does. I’ll forget and go back again in a couple of months though. Sucker.

It was a Cineworld cinema, by the way. Their share price has halved since they floated on the London stock exchange nine months ago. I can’t think why. I suspect it’s not only their customers they treat like crap either. I overheard a conversation between the manager and a potential employee. The manager was explaining how they’d pay minimum wage and how the employee would need to be on-call, unpaid, in case he was needed.

There’s a parallel with business software. It’s often expensive, unusable and buggy. It’s sold by complacent corporations. You use it only because the salesman let your CEO beat him at golf. You don’t like the software, you resent it. You, the customer, is at best an inconvenience to the profit of the corporation. A post by Robert, one of the Red Gate developers, illustrates the frustration that we all feel at times.

I believe that things will change. Software companies will need to treat their business customers more as consumers; sophisticated individuals who can choose to spend their cash on things other than your software; people who place value on qualities other than features. Design, usability and customer service become more important than features. The visceral, emotional reaction you provoke in your customers is as important as ticking boxes on a feature list.

Your customers will look at the software they use at home and wonder why there’s such a gulf between what they do at home and what they do at work. Why does Microsoft Money rock but your corporate accounting system suck? Why can’t SAP be more like iTunes? Apple, Microsoft (less so) and others are setting your customers’ expectations. If you don’t meet them then somebody else will.

At Red Gate, we’re on the path I describe. We’re not perfect, but every product we release pushes us further. We frequently get e-mails telling us how much people appreciate our approach. We don’t publish them all, but here’s a selection of a couple of a couple of hundred.

Finally, ask yourself this question. If you write business apps then are they as easy to use as flickr? Or Google Maps? Or Microsoft Money? Or even – setting the bar low – facebook? Why not, and what are you going to do to fix it?

Enjoyed this post? Subscribe to the RSS feed.

I toyed with calling last year’s Business of Software conference "The Web 0.2 Summit" and publicly not inviting Tim O’Reilly ("Tim O’Reilly not invited to speak at conference shocker"). I decided against it on the grounds that my sense of humour has a very poor commercial record. Plus I didn’t want to risk offending Tim O’Reilly.

Having said that, maybe this year’s Business of Software 2008 really should be "The Web 0.2 Conference" and we should continue to emphasize the eternal truths about the business of software. No bullshit, no fads.

What do you think?

I’ve started organizing Business of Software 2008. We’re planning on holding it in Boston in September or October. The web site is now live.

The big news is that I’m working with Joel Spolsky on this year’s conference. In fact, it’s called "Business of Software 2008 – A Joel on Software Conference". Together, I think we’ll be able to put on something very special.

Although the conference is still many months away we’re working hard on getting some top speakers. Joel, Eric Sink and Steve Johnson (the winner of last year’s Sofware Idol competition) are already confirmed speakers. We have more in the pipeline.

You can sign up for the conference newsletter at the conference web site.

Every Thursday and Friday the Mexican van rocks up, parks about 100 yards from Red Gate and opens its shutters. Just about everything they do is perfect. They serve fresh, tasty food. They have a metronomic process. Mannamexico’s owner, Luis, remembers everybody’s name (he must know hundreds). They have a loyalty card, and loyal customers. They’re the first food van in Cambridge to get a 5* health and safety rating. They restrict supply (they’re only there two days a week) to stimulate demand. Their customers rave about them. There’s even a Mannamexico appreciation society on Facebook (249 members and growing).The only negative thing about Mannamexico is the line of people outside. If you’re ever in Cambridge you should visit the van, or the eatery they just opened in Regent Street.

Do you see what Mannamexico have just done? They’ve just flawlessly executed a piece of word of mouth marketing. I’ve just written what I hope is a glowing testimonial and hopefully drummed up some custom for them. What Mannamexico have not done is explicitly exhort me, or their other customers, to tell people about them. There’s no ‘tell a friend’ button on their web site. They’ve provided an excellent product and the only marketing they’ve done, as far as I can tell, is to park a big, brightly painted, red and yellow van on the road where we work.

Word of mouth marketing is something that marketing folk love. When a friend or a colleague tells you about a product you’re more likely to try it out than if you read, say, a print advert. Getting people to recommend your product to people who trust them is arguably the best form of marketing there is. However, although word of mouth marketing is indeed a worthy goal, it’s not obvious how you get it.

Word of mouth is a bit like happiness. It’s something you arrive at indirectly. If you try to get long-term happiness by doing things that make you short-term happy then you will fail. You can eat chocolate cake or take crack cocaine but you won’t achieve sustained happiness. Happiness is a by-product that you’ll most likely achieve as a side-effect of something else, like dedicating your life to a cause greater than yourself, getting married, buying a dog, having happy parents or not being born in Scotland.

Similarly, the way to get word of mouth marketing is not to spend your time and money directly trying to persuade your customers to tell their friends. The best way to get somebody to do something is to make them want to do it. You need to make your customers unstoppable proselytizers of your product and service. The obvious first step to do that is to make your product and your service something which people are able to proselytize about.

Of course, even an excellent product will not sell or market itself. You still need to work hard to get your product into people’s hands. You’ll need to do Google adwords / run print ads / tape posters to lampposts / whatever works for you. You’ll need to nurture your best customers and make them love you.

But what you shouldn’t do is focus on word of mouth marketing as a direct goal. Do everything else right and it will follow.

Enjoyed this post? Subscribe to the RSS feed.

In 1978, marine biologists from Mexico and the USA joined forces to try to save the rapidly declining turtle population. They incubated turtle eggs and raised the hatchlings for 10 months. They then tried to imprint them with their current location, tagged them and then released them into the wild. If the imprinting succeeded then the baby turtles would behave like ones born in the wild. They would return some point later to form nests at their birthplace, to breed, to lay eggs in the sand and to start the cycle anew.

No turtles returned in 1979 so they tagged and released some more turtles. In 1980 there will still no new turtles or nests, so they released yet more turtles. By 1988 they had released more than 22,000 turtles into the wild. None had returned. They experiment was stopped, and the attempt to replenish the turtle population had clearly failed.

Then, in April 1996, a turtle nest appeared on the Texas coast. Five more were laid between 26th May and 5th June. New nests have been discovered since. Almost 20 years after the start of the experiment it had started to show results. It turns out that turtles don’t return to their birthplaces until after some 20 years, often having travelled thousands of miles.

You often find the same thing in marketing. You release your baby marketing turtles into the wild and wait for them to return. Nothing happens, so you release some more. Still nothing happens so you release yet more. Eventually, you think you’ve failed. Then after more time than you could possibly have anticipated, your marketing turtles return to nest.

Marketing is about sustained, hard work. If you run print ads, you need to run them over months. If you do local marketing, you’ll need to advertise in multiple places for a long time. It’s not about spending all your market budget in one big print full-spread or in a single superbowl advert.

So, release your marketing turtles again, again and again and be patient.

Enjoyed this post? Subscribe to the RSS feed.

I’m thinking about organizing Business of Software 2008. It’s still just a twinkle in my eye. I’d like to know who you’d like to hear speak / who you’ve heard in the past who was excellent. The only criteria are:

• Relevant to the business of software, even tenuously
• Interesting

They don’t have to be speakers who we could ever net (see my list below …), but they don’t have to be well known either.

My personal list so far includes:

• Seth Godin
• Edward Tufte
• Steve Ballmer
• Larry Ellision
• Steve Jobs
• Shai Agassi
• Vint Cerf
• Dan Pink
• Chip Heath
• Donald Norman
• Alan Cooper
• Jakob Nielsen
• Robert X Cringely
• Richard Stallman

Comments on my speakers? Who would you like to hear? Post here.

P.S. Subscribe to this feed to keep informed about the conference.

A quick reminder that there’s a Business of Software facebook group. It’s now an open group so anybody can join. It’s one of the mechanisms I’ll be using to tell people about BoS 2008 so if e-mail and RSS feeds are inconvenient or just too old fashioned for you then please join the group.

Matt Mason’s book on "The Pirate’s Dilemma" is now out. I was lucky enough to get hold of a preview copy of this a few months ago, and to talk (very briefly) to Matt after his presentation at Business of Software 2007. At the time, I thought Matt was onto something very big. I now know he is.

His is one of those big ideas that just clicks. Matt’s premise is that pirates perform society a valuable service. They show where the marketplace is broken. Pirate radio DJs were the bellwethers that created the path for commercial radio in the UK. File sharing and online music pirates opened up the way for iTunes. Matt puts it better himself in this blog post about other people’s property.

You should buy Matt’s book.

Eric Sink, founder of SourceGear, was one of the speakers at Business of Software 2007. Here he talks about why he founded SourceGear, marketing for geeks and other topics. His talk went down extremely well out with conference attendees, with 85% ranking his talk 4 or 5 stars.

An announcement before the holidays captured my attention: Red Hat named Jim Whitehurst, formerly COO of Delta Airlines, as its new CEO.  Maybe too hastily, I had a flashback to John Sculley’s disastrous tenure at Apple.

Whitehurst did some great things in pulling Delta out of bankruptcy and infusing employees with much-needed company spirit.  He is more than casually knowledgeable about open source and has definitely turned around opinions of people who’ve spoken to him.

But even if he is a self-proclaimed geek, does he have the background to succeed at Red Hat?  Are the company’s challenges in the near future operational or market- and technology-oriented? Do software companies need leaders who have a software background?

I’d like to hear opinions from the source — people in the business of software.

My day job is as co-founder and joint CEO of Red Gate Software. I’ve just been set up with a blog on the Red Gate blogs site too. In the future, I’ll post more Red Gate related content to the Red Gate site and more general stuff here. I’ll link between the two.

My first Red Gate blog post this year is a product-centric round-up of what we got up to last year, and what you can expect next.

If you’re interested, many other Red Gate people have blogs too.

Our assumptions bind and blind us. Often we believe them with no evidence, simply because everybody else does. They have no reason, no logic: they just are. This is as true in the business of software as it is in life.

They can be so ingrained that we’re not even aware of them. In the West, killing cows is acceptable but killing whales or dogs is not. Artifical additives are bad for you, ‘natural’ ones are not. Our intelligence is innate, measurable and fixed, and determines how successful we will be. Having children makes you happy. Terrorists are crazed, mad, brainwashed killers.

These assumptions are all commonplace. They also all have little or no evidence behind them. The cow/whale distinction is clearly arbitrary. Natural additives are often chemically identical to artificial ones, simply produced using yesterday’s technologies. There is no reason why ‘natural’ is better than artificial, and the boundary between natural and artificial is blurred. Syphilis is natural, penicillin is manufactured (and a good example of the blurring between natural and artificial). Assume that your intelligence is fixed and determines your success then you are more likely to fail than if you don’t. Having children makes you less happy, until they leave home at least. Most terrorists are normal, often middle class, well educated people who believe in a cause that they feel is worth dying for.

Challenge assumptions and you can change the world. Challenge the assumption that the banking system works and it fails. Earlier this year, people in the UK lost faith in a single bank and withdrew their cash. Only a £40 billion ($80 billion) state bail out saved the system.

We make assumptions in software and in business too. We often aren’t aware of these assumptions until we challenge them, and when we do challenge them things get interesting. Here are some assumptions, past and present, which are worth challenging:

There is no such thing as the business of software. Software should be free. In 1976, Bill Gates challenged this assumption. He’s done quite well out of it.

There’s no room for a new search engine. In 1999, Altavista had search sewn up. It was the unassailable king of search. Google challenged that assumption. Now the assumption is, yet again, that there is no room for a new search engine. Google has won the search engine wars. What if that assumption is wrong?

We should work 9 to 5, and be measured by our input not our output. Historically, when we were all basket weavers or pin makers, interchangeable cogs, fixed working hours made sense. People are now challenging this. In Brazil, Ricardo Semler is experimenting in industrial democracy. Workers set their own salaries and hire and fire their managers. When somebody tries to fill in a timesheet, Semler sends them away, not knowing what to do with the information. What if we all followed this example? What if we worked when and where we wanted, and if we were judged on what we produce and not the hours we keep? What if we removed all mention of hours to be worked from our contracts?

Only men aged 15-30 are interested in games consoles, and all they care about is graphics and speed. Nintendo challenged these assumptions. By cutting down on graphics and processor power they produced the only high level console that sells for more than it costs to make. By reducing the cost and widening the target audience, the Nintendo Wii is currently outselling the Playstation 3 and the XBox 360. The Nintendo DS, which is also aimed outside of the usual demographic (25% of users of its best selling Brain Training application are over 45), is doing well too.

Screen displays are expensive. Bill Buxton has pointed out that in 20 years time screen space will essentially be free. 25 years ago, when 300 baud modems were the norm, few people predicted that bandwidth would become essentially free and infinite. The same thing will happen with displays. The 32" screen I bought for over $2000 a couple of years ago now costs $800. Companies are already building factories that will ultimately produce flexible, plastic-based displays for one hundredth of the cost of their LCD counterparts. How will free, ubiquitous displays of arbitrary size change the business of software? It’s a question worth asking.

My product is so good it will sell itself. What if you’re wrong … ? And you are.

Half the money you spend on advertising is wasted, you just don’t know which half. It’s a neat aphorism, but that doesn’t make it true. Why not challenge that lazy assumption and try to disprove it?

China is an unstoppable economic force, and that’s bad. That’s what most people are saying, but is it true? For a start, China might seem unstoppable economically, but it’s facing enormous political and environmental problems. And if it really is an unstoppable economic force, what do an extra 1.4 billion potential customers mean for you? Is that really bad news?

Software is a people-intensive craft. At the moment it is, but does it have to be? What if we’re currently in the pre-industrial revolutionary stage of software development? Or the era – only 140 years ago – before the standard screw, when a bolt from one city wouldn’t fit a nut from another? Can software development become industrialised? And will it, and should it? And how will this change your business? Already, some are pushing software factories as the next big thing.

The Internet is here to stay. What if it isn’t, at least in a recognisable form? The world wide web has only been mainstream for 12 years. Much has changed in those 12 years, and much will change in the next 15. Will we still be using html, ajax and flash? Will a world of nearly infinitely fast connectivity, ubiquitous and free screen display, and computers 30 times faster than today’s still rely on http and html? Again, it’s an assumption worth challenging.

As I am, you are bound by thousands of assumptions about the way the world works. These assumptions range from the macro to the micro: they might be specific to your culture, to your country, to your company or even to yourself. Every once in a while it’s worth practising what Alan Bennett calls ‘subjunctive history’. Pick something you absolutely know to be true, something so obvious that you’re not normally even aware of it, then switch on Professor Farnsworth’s machine and ask "What If …".

Enjoyed this post? Subscribe to the RSS feed

Matt’s thesis is that piracy can be a good thing. Pirates are
bellwethers, flagging the existence of emerging markets. Rather
than fighting them, we should consider competing, or even co-operating,
with them.

Here’s Matt Mason talking about the Pirate’s Dilemma at last month’s conference:

It was extremely well received at the conference and it’s well worth a listen. You can find out more at Matt’s excellent blog.

At last month’s conference, Alberto Savoia gave a great talk about crappy code, what it is and how to avoid it. Alberto is a wonderful speaker with a lot to say (90% of the conference attendees gave his talk 4 or 5 stars).

Here’s the video:

You can read an interview with Alberto here:

http://blog.businessofsoftware.org/2007/09/alberto-savoia-.html

And visit the crap4j home page:

http://www.crap4j.org

Enjoyed Alberto’s talk? Then register your interest in Business of Software 2008.

In 1880, at the age of 27, our protagonist decides to become an artist. He has already tried, and abandoned, the careers of art dealer, lay minister, school teacher and book seller.

This is one of his first drawings, of a carpenter:

5 years later, our protagonist is on the way to mastering his craft. He paints these potato eaters:

4 years, and over 2000 drawings later, he paints this:

There are some interesting parallels with software development here. Here are a couple. One fairly obvious, one less so:

• van Gogh had to master his craft before he became an artist. He spent years perfecting his techniques and copying other painters before painting any significant works. Software development is a creative process as well. You too will need to copy, try and fail, and eventually master the fundamental techniques before you can produce your masterpiece.
• Put the 27 year old van Gogh in a sketch-off with me and there’s no way that you could tell from our drawings that he would end up one of the most influential and famous artists of the past 200 years and that I was destined for eternal mediocrity. The same thing holds for software developers. If you’re faced with two untrained, wannabe software developers then you have no way of distinguishing van Gogh from the muppet.

What do you think? Post here …

Enjoyed this post? Subscribe to the RSS feed