We just almost didn’t buy a widget that saved us at least a dozen hours of coding in the first week. Because it was too cheap ($139). If it had been open-source/free, we’d have tried it in a heartbeat - investing three days of engineering time that costs me $1,000/day.

So let’s net that back: it’s perfectly reasonable to spend $3,000 of time to try to accelerate a project but only if you don’t have to spend $139 to do so.

What. The. Frack.

When I say it has saved us a dozen hours, I should be more clear: this widget saved not only the three days (24 hours) we put into making sure it did what it needed to do, but it at the end of the fourth day (8 hours), we’d finished a task that was scheduled for 6 days (48 hours). I’d say it saved us two days (16 hours) but we had to write a bunch of new test plans (4 hours) to make sure we got good coverage.

So this $139/investment saved us $1,500 (1.5 days @ $1,000/day). In a week.

That, my friends, is a good investment.

Oh, and we own the widget now, so I’m sure when we next re-plan we can bring in a bunch of UI stuff, I betcha. Which saves even more money. And since the project will have more slack time, risk is lowered.

And we almost didn’t do this because we hesitated to spend $139.

Next team meeting maybe we’ll discuss giving everyone a budget, say $250, to buy software tools. Heck, you don’t have to find but one or two on a project. And since this one is clearly going to keep paying dividends, it’s probably “free” money.

Nota Bene: You may have noticed that I emphasized 8 hour days. That is because we believe that employees should go home unless the place is burning down. Nobody really puts in 8 hours, but if you push people towards that they may only work 10 and will be fresh when you really need a push.

I was at a concert tonight and, like all the dads, I was pretty happily ignoring it while waiting for my kid to come on stage. But the conductor kept intruding - and not because she was so great, or the orchestra so bad, but because I was having to observe a musical Tourettes.

We all know people who love to hear themselves talk - they sit impatiently waiting for us to shut up so they can give their observations, advice, humor (mea culpa, I’m afraid), or just talk.

And this conductor was so clearly relishing the role of being in charge of 90 or so talented musicians. Plus an audience of several hundred forced to watch her gyrations.

And I think that is what a lot of people think owning/running a small business is like.

Look, if you want to wave a wand and have people know exactly what you want and then do it you should immediately stop doing whatever it is you are doing and go get a job at GE. Or Boeing. Or NASA. Someplace with a thin veneer of ‘hip’ or ‘innovation’ spread over a nice juicy stack of REVEALED WORD OF GOD operations manuals.

You wanna be a conductor? Be the plant manager of a factory assembling jet engine fans.

If you’re more comfortable with a jam session, then small business might be alright for you. Get some talented people, agree on a goal, and then try to do your thing in a coherent way. Make a lot of mistakes. Let the drummer do the 15 minute solo sometimes.

In our company we have a blended set of tasks - some stuff is done one way and one way only (finance, planning) but most everything else, especially the technical and creative stuff is pretty ad-hoc and depends on focus and experience.

I like it that way - I’ve been a conductor. I kinda like running the sound and watching other people perform.

(Oh, and my kid was the star, from the back of the chorus. What are the odds?)

This is a bit of digression, and I hope nobody thinks I’m calling p0wned on them or anything, but has anyone else noticed how many blogging articles are about how to write blogging articles? More whitespace. Less whitespace. Use pictures. No, use bullet points.

Good geejebus.

You want to know what Anthony Trollope said about writing? “Just sit down every day and write something you plan to keep.” He’s one of my favorite authors - a more popular contemporary of Dickens (there is a Facebook lesson in there for MySPace) who also was post-master general of Ireland and invented the post-box. (Yes, it was invented.) He also wrote more than 40 long novels.

So my advice, if anyone besides my children were to ever ask my advice, is to not try to write the best post ever at any point, but to just write something you plan to publish.

My best post ever was on a non-work related blog about something very strange that happened to me on vacation. It got several thousand hits in the first day and now, more than a year later, a dozen or so people show up to read it. Who knew.

I often notice anxiety in our more junior guys - when we hire junior guys! - about the quality of their work. I think they somehow believe they should be Andresson’ing the next Linux or something. “Dudes,” I say, putting on my really bad fake Indian hipster accent, “that was totally a gnarly hack.”

It appalls them, in a grandfather-in-board-shorts kind of way, but sometimes they understand: mostly at work you are doing a pretty good job if you produce production code.

Last week one of our guys started dancing around like a drunk Scotsman on ruffies. Apparently he’d solved a performance problem with some Ajax grid/tree thing that was taking seconds to load for small data sets on a website. Now he can load a 1,500 item grid/tree thing in under a second. Weeks of poking a sorta-solved problem in his spare time.

Total home run. He’s happy. Things run better. We’re all smarter.

But he wasn’t trying to write the best code (or post) ever. Just to do something for production that was as excellent as he could make it.

So say we all: you choose the team pizza today. But no jalapeno’s!

Sorry about the Princess Bride reference, but there was this great post on Performancing about Text Deserts. He talks, very wisely, about how to make your posts more visually appealing, etc, etc. I’m ignoring his very good advice.

What?

The purpose of this blog is not to have a zillion billion readers. The stunning lack of ads would probably tell you that. The purpose of this blog is to communicate with potential customers about what kind of people we are. How our company runs. What is important to us.

And we’re not a retail company, we don’t pitch hard sell stuff. So we really want people who can read complex sentences and follow thoughts that are more Faulknerian than Hemingway. Oh, and if you read Dilbert, enjoyed movies like Office Space, and think it’d be fun to go down to pub and tell a war story or two, well, you’re our kind of customer.

And I think a lot of people like that are finding our blog. We’ve had steady increase in traffic, lots of great emails (strangely, no comments - that is kind of startling), some good feedback from other posters talking about our posts, and a bunch of people very very curious about our product, how they can be beta testers, etc. Which is also good.

We’re purposefully being a bit cagey about the details, but once we go live, we have some really interesting posts to do from this development process, some of the things we’ve learned, and how we had to give a persistent VC the bums rush.

Stick around!

I have no axe to grind - Trust Seals (BBB, HackerSafe, Trust-Guard) are a minuscule part of our budget and, frankly, we spend more time/money trying to figure out where to put them than on sites than we do deciding whether or not to buy them. Ditto with SSL and “site ownership” buttons. But we use them all, liberally.

Not liberally in the sense of using other people’s tax money. I mean it as ‘more-is-better’ - site seals, SSL badges, etc, etc. They’re all good, below the fold.

There is a great thread on Webmasters World right now where a guy did A/B tests and increased his conversions 97% with seals. That is a ton, and they’re paid for in a day or two in incremental business alone.

We use Trust Guard because their seals are pleasing looking and their procedures, while detailed, are un-intrusive. They also have a price advantage over their competitors - they’re about half as expensive. I believe the last set I bought from them for one project was $400 and the project before that we used HackerSafe/BBB and it was a grand.

I’d be shocked if consumers thought there was a difference.

We’re pretty old-skool and had been using Verisign SSL infrastructure since before it was Verisign. Last year one of my guys wandered up to me at lunch and suggested that our registrar, GoDaddy, was a better deal at $20/year than Verisign at $400/year. I felt pretty dumb. So we’ve switched over and we took everyone and their significant others out for a very nice dinner. (How nice? Cabs home for everyone!)

So we try to do this stuff as inexpensively, in terms of both money and time, as possible. It reassures some people, the rest ignore it, and almost nobody can tell you the difference between vendors!

The internet is a very very big place, full of strange and wonderful pages. And then you run into things like DisposableWebPages. And you shake your head: no advertising, no link love (the pages disappear after 90 days), and just clever enough that they should have been smart enough NOT to do it. (My page is here. The password is “matilda” - feel free to add your comments if you like.)

There are a lot of sites like that out there that are just shake your head and keep surfing or stumbling. But there are some where you can see that people will depend on them and, then, one day they will disappear.

For example, OnlyWire, a system to semi-automate a small piece of the social bookmarking process. No revenue stream from ads. Lot of google link love, so maybe they’re going to do something with that in the future. Sometimes it goes down, who knows why, but it comes back up.

So what, you say, it’s free and it (usually) works.

Yeah, but your time isn’t free. Changing your business processes to use the site isn’t free. Changing them again when the site goes ‘poof’ isn’t free.

Oh, and as it is free, there is no TOS, no Privacy statement, no contract. So if something bad does happen with all the logins you put in…. Sure, you’re smart enough not to put your primary Digg account in there, but you know how they are on Digg - if there is a problem and it starts rolling back it could impact your business. And there is a cost associated with risk.

There is another kind of resource out there - things like Wikipedia. I’m sure most people don’t think about it, but it is an unprofitable not-for-profit. It’s not like it is going to disappear, but a lot of people depend on it and it sucks the oxygen out of the room for people trying to sell information. And it could, of course, disappear someday. I put the risk as low, but still.

Look, I’m not trying to pick on Disposable, Wikipedia, or OnlyWire, but I was really struck tonight about how we sometimes don’t think about the cost of using free services.

At our company we use a LOT of open source software, but only when the technology to fix it is in-house and only when there is a thriving support community. We buy lots of software from smaller widget manufacturers (we just got an AWESOME grid library for our new project) but only after we formulate an alternative plan just in case it’s, well, krep. And we watch those vendors carefully so that we can anticipate if they’re going to go out of business.

Maybe, if I had to, I would divide the free/cheap software world (in other words, nothing from Microsoft or Oracle in this exercise) into hobby, institution, or niche. Hobby stuff (OnlyWire) can disappear at any time - there are no visible economic or social (fame) drivers. Institutions have a certain amount of momentum so if they disappear, it’ll be a fading away. And niche, which can prosper for years though it is vulnerable.

We try to avoid the hobbyists, use the institutions, and watch our niche suppliers.

In every project there comes a time when everything seems to slow down and not quite stop. Like taking a nap when you have half a hangover - you don’t really sleep and you don’t really heal. You just slow down and stretch out your misery.

I have never really had a good explanation for it. I’ve seen developers just run out of juice when under pressure, when ahead, when working on critical path stuff, when working on utilities. In other words, I’ve seen it across almost ever circumstance I can imagine.

Sometimes there are physical reasons and you can fix those.

Say build times are taking too long - hours or even overnight: buy a faster machine. Take a few days and re-arrange your code so it’s faster. Get people involved in fixing the problem.

Are your people tired? Hey, the excitement of the start is gone and the deadline isn’t close yet. Send ‘em home. Make people take a four day weekend. Go on a team hike.

Maybe your team has spent too much time together working on hard stuff. People can de-bond as easily as they bond. Maybe it’s skillset mismatch - Fred is stuck doing all the PHP and he wants to code Ajax for a while. Whatever. Your team leads should be paying attention to that stuff.

But sometimes it’s just not clear why it’s happening, or the things you try aren’t working. Well, you know I’m a fan of re-planning. Take a day or two and replan. It’s not like much work is getting done anyway!

We’ve been considering creating a new permanent position in one of our tech team for four or five months. We don’t hire willy-nilly because we have almost zero turnover and have a horror of layoffs. We have a couple of folks who have been doing full or part-time contract work for us, so, because we’re a buncha chickens that like to manage risk, that group was our focus:

1> Reduced risk due to known ‘isims’
2> Reduced cost of recruitment
3> Reduced time-to-productivity

My business partner spoke to the hiring manager, let’s call her Sharon (inside joke), and she went through the exercise of describing the perfect candidate, spec’ing out the associated costs (lab, hardware, etc), the ability to push projects out faster - all the standard stuff. Then she had a think and came back to us with her preferred candidate.

We always ask the same question: Are they an “A” player?

Let me explain: Top people hire top candidates because they’re not scared of being shown up by a direct report. Second rate people are scared they don’t deserve their jobs, so they hire people they can control: third rate people. We shorten that to the headline of this post.

Sharon said that the person “had the ability to be an A player.” If she hadn’t slapped her head at what she said, I’d have used the rarely used Veto Pen!

Because we all know that an annoying habit in a job candidate is only going to get worse when you can’t get rid of them by shaking their hand on the way to the lobby.

It’s even worse when you hire a contractor who is turning in “B” work. You’re rewarding them with a permanent job, health care, and some prestige for their behavior. Plus you’re taking away the threat of instant-out-the-door. So they’re going to change their behavior why, exactly?

Look, we don’t blame Sharon for *almost* making a mistake. That is why we do things in teams - you can’t catch everything. The important thing to me was that she ‘got’ it and moved right on to figure out the right answer. That’s why she’s an “A” player.

No, not the “crips vs. bloods” debate. Nor the “Arsenal vs. Real” one. The one at work where one guy wants a triangle that points right (feng shui!) and another one wants a circle that spirals inwards toward the first letter of your company name. And, oh yeah, someone has an Aunt who is “creative.” Help!

If you’re going to be selling retail-over-the-web in a super competitive arena like fashion then you should spend a lot of time on your look, the shapes of your icons, etc, etc.

Otherwise, I would argue, once you hit a minimal level of quality then you’re done. My prime example of this is Amazon. A sprawling pile of ugly UI. Next up is GoDaddy. Enough said.

Do I think that REI’s site is prettier than LL Bean’s? Yes. So what. I’m not sure it changes anyone’s buying habits.

Social sites? If my dog looked like reddit I’d shave his behind and take him for walks backward. 8M visitors a month? Obviously ugly isn’t important. I think propeller.com looks like it fell through a time warp from 2001. And they just put it up!

The best way to save your time and avoid the logo/color hassle is to first realize that if you were to stackrank what is important about your website, worlds-best-graphic-design would be at the bottom of the list. Or near there.

So, once you’ve decided to avoid spending a $10K and (what is worse) a thousand hours arguing with everyone about logo shape and size, I suggest that you get on over to google and type in “logo design” and pick someone who will give you something decent for $100. You don’t have to marry it: “ok” is ok. (Hint, if their site is ugly to you, don’t buy a logo from them.)

Then, once you’ve got something that takes your fancy, go back to google and type in “Joomla template design” (or whatever) and find someone who will take a basic template (free or paid - I see little quality difference, frankly) and match the colors and give you some basic navigation icons in similar shapes and colors. This should cost less than $200.

So, there you go, a week elapsed time and $300 and you have a decent looking website ready to hold your navigation, content, and product.

If you really want to save money, reverse the order: find a free template you like and tell the graphics designer to match the color. I think it works better the other way, but, then, $200 is not such a big deal to me. YMMV.

If your minions are really upset by this, try to get them to focus on navigation issues, or tone of contact, or something that is actually customer affecting.

We were having a philosophical discussion about posting content to social networking sites - the diggs, reddits, propellers, etc. But also to the newsvines, canadakas, and all the others.

One of the guys was trying to justify posting stories from his dog blog into foreign sites which are over-run with (generously speaking) English general purpose stories. For example, he found NuMarketing and was excited - the dutch love dogs, he wants visitors to his blog. Perfect.

Except he was getting called him all sorts of names for being a spammer. But, of course, it’s not really spam, it’s just off topic. Well, actually, it’s not off topic from what’s being left on the pligg site, it’s just off topic from the quote on the side. If the quote on the side about the site being Dutch Marketing is accurate.

So, we got into quite a discussion about evil, mini-evil (lots of Austin Powers fans at work), and the tragedy of the commons (yes, we have libertarian economist on staff - don’t you?).

I tend to come down on the side of more generously putting my stories around when it looks like that is accepted behavior. But I think if a webmaster tells you to beat it, then you beat it.

But the fun part was trying to come up with a story you could post *everywhere*. This is the best we could do:

My dog and cat, both vegans, are legally certified to be companion animals for the blind, the aged, and people suffering from depression and disease and they travel internationally for both that and to compete in animal-lmpics. Their former owner, a GLBTU adoptee, died in a tragic car, plane, rail, and cycle crash due to global warming and while backpacking around the globe. We found out because they twittered and blogged and kept facebook and myspace up to date with their GPS, smartphone, laptop, and wimax service, usually while listening to an ipod with itunes while surfing ron paul, hilary, fred, rudy, etc.

We think you could post that to any social networking site. And be on topic.