Spolsky: Setting the Right Priorities | Hacker News Reader

Spolsky: Setting the Right Priorities

inc.com

105 pointswyday35 comments

A lot of software developers have critical priority inversion, focusing on initiatives which will almost never add value. Without specific reference to where Fogbugz was back in the day, I think a lot of software developers often prioritize improving the product when that should not be #1 on their list with where their business is right now.

In particular, if you have a working product which is salable, and you have a website which is getting 5 visitors a day, put down the IDE and start marketing. Better mousetraps do not automatically collect more mice -- you need to locate the mice, preferably a bunch of ones which enjoy eating whatever your bait is, and then put the mousetrap by them.

(There are circumstances in which development supports marketing -- working on your viral loop for certain species of web apps, or adding something which greatly improves your perceived value in your target market. Don't spend three months to add another bullet point to a web page nobody reads, though.)

It's not that easy. If you do marketing and other "business" type stuff, your body will demand some sanity and just good ole programming. I can completely understand people being "sidetracked" implementing novelty code they will never need. They're not wasting time, they're recharging their batteries: sometimes you just want to work on something tangible, verifiable, and get a quick satisfaction fix seeing it working.

Lately, most of my work is CSS, input validation, architecturing URLs, and reading as many web design "top 20" blog posts shopping for UI elements. Pathetic. However, between doing that, I have also written a completely unnecessary server-farm management package for my site, completely with load-balancing, health monitoring, backup, change propagation, hot configuration and other stuff I probably would never need (let's just say that if I ever needed it, I would also need a physical server rack.)

Of course, you have to do some marketing. But I think that if you're investing time in marketing to be able to survive, then you're doing something wrong ... you either chose the wrong market, or your product is not good enough.

From my experience, improving your product (which is not equal to adding more features) is the best way to get new customers. That's because marketing that's not properly targeted is as good as no marketing, and let's face it ... technically inclined people suck at it. Even Spolsky, which is what I call a marketing drone, in his interview on "Founders at Work" confessed that they got more customers from incremental improvements than through marketing plots.

On the contrary, I think the main problem of software developers is that we rarely finish what we start.

Marketing is an engineering process. Technically inclined people do not have to suck at it. To the extent that many do, it is largely because they've been taught to undervalue its importance, treat it as an afterthought, and dismiss properly conducted marketing as something which is for weak and stupid people.

> dismiss properly conducted marketing as something which is for weak and stupid people.

I didn't say that. I think it's dismissed more because the majority of marketing campaigns send out a dishonest message, and because of that customers are getting more and more insensitive to such campaigns.

And I also don't think it is an engineering process in the traditional way (although it's fashionable for everything that has a feedback loop to be called engineering these days). Marketing is more about social interactions, being more inline with humanities / social sciences ... and it takes a certain kind of skill-set to be good at it, and I've seen too few technical people being good at it.

You're thinking of advertising rather than marketing as a whole.

Think of marketing as the social component of design. You wouldn't start coding an application before you had at least a rough design of what its functionality would be, right? So why would you start designing a product before you identified who is going to use it and what they will be using it for? Marketing answers those questions.

In rough, marketing identifies your market: Who will be using your product and why. Your application / business is then built using that knowledge.

"From my experience, improving your product (which is not equal to adding more features) is the best way to get new customers"

How can customers find your product and understand what it does if you build stuff and don't market it?

> How can customers find your product and understand what it does if you build stuff and don't market it?

I already said that ... you have to do some marketing, but you shouldn't depend on investing marketing resources for survival. That's a big difference. Good products are their own marketing, and you can rely on a networking effect ... people usually tell their friends when they've found something hot.

And I'm not even talking about viral web 2.0 apps here ... I have a management app specific to pharmacy-shops (doing a couple of crazy stuff others haven't thought about) and it sells itself (I only had to do marketing for the initial 5 sells or so, and now after 4 months I have 30 customers ... which for the price I'm charging is really good).

1. You assume customers talk. Bad assumption in many, many markets. Why do you think chasm tells you to choose a niche with a network effect? Most markets aren't like this.

2. Network effects have specific rules to be useful. Most businesses can't take advantage of them, especially if they are charging money.

3. 5->30 is not a good basis to judge this on. You are still in the early adopter phase, and luckily for you this market must have a high word of mouth quotient.

The problem with just assuming things is that it stops you from looking objectively at your market, and you end up relying on luck. Most businesses I know want a repeatable sales process, not one based on faith.

Three years. YEARS. It took them three years to figure out how to grow Fog Creek. Jesus, what a relief. This article has made my week. Thanks, Joel Spolsky.

Agreed, probably the most powerful line in the entire article:

The first three years of Fog Creek were like this. We tried in vain to find the magical sales and marketing formula that would make our software successful.

When I read Joel's articles I know I enjoy them, I know they're interesting and I learn from them but also that I'm reading marketing for his products and brand - its win-win and I subscribe to his marketing output along with everything else he publishes willingly.

I'd like to read about the impact "Joel on Software" and his celebrity has had on his success/sales though.

Its almost like you read the article and he discusses "this made my company successful", "this was a mistake of mine" but talking about his blog and celebrity is seemingly off limits, to a layman like me its a huge factor in the appeal of his products but it goes unsaid in his articles - as though it had no impact.

I wonder if there's a worry openly discussing the marketing benefit to Joel of us reading his articles would be almost breaking down a wall and change our perception of them.

Or am I wrong, has Joel (and/or Jeff Atwood) discussed openly the difference their blog-marketing has made to their business success?

I'd like to read about the impact "Joel on Software" and his celebrity has had on his success/sales though.

within epsilon of 100%. I love Joel, his articles, and the Stack Overflow podcast, but nobody would have heard of him (or FogBugz) if he weren't such a brilliant marketer. He established FogBugz and Joel Spolsky as pretty big brands in the programming community, and made a previously unknown software company into what is now seen as one of the elite places to work in the industry.

He's got the technical chops to back it up, but that doesn't matter - he is amazingly good at marketing, and he really gets the power of brands.

I remember reading Joel's articles a couple years ago when I was studying for job interviews, and thinking, "Wow Joel sure is smart - Fog Creek must a really badass company." I had no idea what products they made. Now I'm working on my own company, and one of the first things I did was sign up for the 45-day free trial of FogBugz.

The date of the article is sep 1, 2009.

That's 6 days from now.

I'd say that's the publish date for the print article. Don't you just love old media ;-)

And they wonder why they're dying...

I read this article last Tuesday at the auto mechanic, after I finished the book I brought. So it's probably the September issue, which comes out before September.

Joel can see the future, so can Inc. ;-)

Wait, you're saying this article, which is about the past, somehow requires prescience? So the now-past is different from the future-past? Can we compare the now-past with the past as it will be on 9/1 and somehow glean a way to make piles of money?

I swear I've read that article before, so it's recycled anyway.

I think it's a shame that this article shortsells the importance of promotion and sales. I wonder if it was perhaps a backlash from Joel to the sales driven culture of Microsoft.

Promoting and selling your product is a discipline that forces you to answer some critical questions about what it actually is:

1. Who is going to buy your product - is it SMEs or individuals. If it's SMEs which ones in particular, how do you phrase things like your website copy to be meaningful for them

2. What do they want the product to do (e.g. Wufoo was online forms which was totally disruptive to administrative areas but actually ended up doing a lot of tailoring to online surveys which (I'm guessing) is where the market actually spends money)

3. How are you going to get customers - not everything is word of mouth. Some things are simply about distribution and good sales (what makes you buy the tin foil you choose - word of mouth?!)

Developers always seem to hate the idea of actually promoting anything. There is a feeling that if you build it they will come. In general though they seldom do.

Try to sell it and you'll find out what people actually do want.

I wonder if developers hate promotion because it's hard. Whilst writing code is fun and engaging, cold calling to do some market research or to sell your product makes you feel uncomfortable and risks personal rejection.

Sales is super-important and every company you admire does it Apple markets the life out of all of its products, Google has massive Adwords sales teams and Virgin America might be the best domestic airline but you can be sure their growth would not been have as great without the massive marketing campaign that went with it.

I wish the many awesome software teams out there would stop spending so much time on product they think people want and start spending more time on selling it to people and discovering what they do want. The process isn't always fun but it is very, very informative.

In my first year in the valley I wasted a lot of time in vague meetings with "strategic partners" and "investors." We called it "who's pitching who?" I'd say these types of meetings account for 70% of valley lunches.

Interesting article! In hindsight everything is distorted and seems so natural, however, it doesn't teach you much about the _current_ reality -- how can you tell if your company is suffering from a product or marketing problem _right now_?

It's a good article, and the main point is an important one.

That being said: am I the only one who finds it quizzical that Joel ever became such a notable pundit in the first place, in light of all of the articles outlining his (past) clueless behavior?

I think he's popular more for his engaging writing style than anything else. I would consider the fact that he made mistakes in the past a non factor at the least, and at best, an indication that he's capable of learning from mistakes.

I enjoy his writing style. But I think people have been putting him on a pedestal for a while, in terms of the advice he dishes out. The fact that he's made mistakes in the past would be a non-factor, if it weren't for the fact that people are following his dicta-- and how do we know that the advice he gives now is not going to be the fodder for next year's "I-was-so-much-dumber-then" article?

It's very simple, actually. Anyone who blindly follows another man's advice, without checking everything and bringing their own judgement into the formula, very well deserves all that happens to them.

I actually think his "clueless behavior" is one of the reasons his stuff resonates with people. Everyone makes mistakes, some of us make them a lot. Lots of people in Joel's position are tempted to leave out the mistakes they made when telling their story. By detailing all these blunders, Joel connects with the average "Joe Programmer". Instead of making himself out to be a super-human, he comes off as a an informed peer. It makes his stories much more interesting.

From a very good screenwriting website comes this essay, "Impressive Failure":

http://www.wordplayer.com/columns/wp08.Impressive.Failure.ht...

"In plotting an action-adventure type film, how your hero fails is at least as important as how he succeeds. If the plot requires the hero to fail, try to figure out a way for him to fail as impressively as possible.

"Consider RAIDERS OF THE LOST ARK. Indiana Jones is perhaps the greatest action hero in the history of the movies. And in his debut film, he flat-out fails from beginning to end.

"He loses the golden idol. Marian is kidnapped and he's unable to rescue her. He finds the Ark, but it is immediately taken away. His bluff to destroy the Ark is called, and he gets recaptured. He can't even look upon the Ark when it is opened. And the government ends up with his long sought-after and much suffered-for prize. This guy's an action hero?

"Yup. Because he fails so damn impressively, from start to finish. Indy fails so well, in fact, the audience is impressed as hell, and hardly aware of the fact that he's failing. The defeats are just setbacks that create more opportunities for heroism. As an added benefit, Indy wins the audience's sympathy -- the poor guy's trying so hard, you can't help but root for him."

Readable version: http://www.inc.com/magazine/20090901/joel-spolsky-setting-th...

Printer-friendly is more human-friendly than the original...

It ain't funny

Is it really prudent for Joel (CityDesk) to compete with big companies (http://www.microsoft.com/frontpage/) and FOSS (http://net2.com/nvu/) products?

CityDesk has been practically abandoned for a long time now. Fog Creek now focuses exclusively on FogBugz and Copilot.

CityDesk has been abandonned for a long time now. That said having an established competition is not necessarily a good reason to avoid a product space altogether. After all, FogBugz, which seems immensely successful and has been Fog Creek's cash cow for a while, has plenty of very established competitors - foss and non-foss.

Reply on news.ycombinator.com