Monthly Archives: April 2011

Startups Keep Tweaking Your Site. Lessons From

I’ve written before about how over the last few years my wife and I have written several small sites (you can see some of them listed in the sidebar.) It’s been an interesting experiment in creating stuff tightly targeted for certain readers. Since all startups have to have an internet presence, and since having people come by to see what you’re doing is an important thing, I’m very interested in how to reach people with a targeted message.

As part of that, we have a site called It’s nothing special, just trivia about stuff you can find on your paycheck stub.

Over time, however, we’ve been getting more and more visitors — over 5 thousand per month now. With that many people, I’ve been very curious if I could help them in some way besides just displaying information and ads. What the heck do these people want?

So I popped up a list of buttons and hooked them all into Google Analytics. Do you want to print fake paychecks? Access our online payroll system? Verify that a paycheck is legitimate? Download hundreds of paycheck stub images? I brainstormed a dozen reasons and made each one into a button.

The button didn’t do anything, just reported to me how many times it was clicked.

I found that 2-3 percent of visitors want to make fake paychecks. For what reason, I don’t know, but I’m obviously not going to be able to help them. Another 2-3 percent wanted to access the “free online payroll system” — which, of course, didn’t exist. .5 percent wanted to know if a paycheck stub was legitimate, and the rest of the buttons didn’t get much pushing at all.

So what to do with that information? I did some research and found a vendor who offered free online payroll, of course — at least for the first month. This was nice because I already knew that this was something 2-3 percent of my visitors wanted.

I took down the buttons and put up a plain text link to the paychex site. It simply says “Make your own paychecks online (free first month)” No flash, no graphics, no sales copy. Just “Here is what you want. Click to go get it”

It’s nice to realize that once your audience is qualified enough — once you know what they want — “advertising” is just providing them with a simple text link.

Eight days later, and I’m $250 bucks richer. Woo hoo! Even now, as you read this, people wanting payroll services are finding my site, getting some information, then clicking over. I could be cynical and ugly and say I hear the sound of cash registers ringing, but what I really hear is the sound of people finding what they were looking for, which is even better. After all, it’s not like you have to trick somebody into getting a payroll system. These were folks who wanted a payroll system all along, perhaps they started out by looking for an example of a paycheck stub, then realized there’s a lot more to it than simply filling out a form, then realized what they really wanted was an online payroll system. I’m just helping them do that — for free.

For the other 95% of visitors, the site looks mostly the same. Time on site is still great — lots to see and read — and we get a fair chunk of return visitors. (I still think the site sucks, but that’s a story for another day.) And the Google AdSense ads are still getting as much action as they were before the site upgrade. It’s not like adding one thing decreased something else. In fact, since the click-through from AdSense stayed the same, adding the link increased the usefulness of the site. More folks who were looking for stuff found it.

A couple of good lessons from this. First, figuring out what your visitors want want — getting inside their heads — is really, really tough. Anybody who says they can do this based on gut feelings is probably lying to you. But if your traffic is high enough? Ask them! Make some buttons and let them click them. Visitors are really good at clicking buttons, and if nothing happens usually it doesn’t piss them off too much. This means that having a site that’s related to your startup that gets thousands of visits or more a month is a key asset — like having a gold mine in your backyard. This is why everybody says startups are supposed to have a blog.

Second, explore new partnerships! By reaching out and doing some poking around, I found a business that was very happy to provide free payroll for a month. This helps both the reader and the business. Very cool.

With ten or more people clicking over for free payroll each day, I’m actually in a pretty good spot to start an online payroll company, if that’s what I want to do (it isn’t.) Over a year, I’d get 4 thousand people coming by to check out my product. People I didn’t have to pay a penny to have visit.

And that’s why this stuff is both very cool and is a generic skill applicable to all kinds of startups.

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

Software: More Battlestar, Less Gunsmoke

When I was a kid growing up in the 1970s, we had a show on TV called “Gunsmoke”

Gunsmoke was a western, which meant folks dressed like cowboys. Our hero, Marshall Matt Dillon, struggled every week with capturing bad guys, cleaning the town up, and just generally doing cool cowboy stuff.

Gusmoke had been running on TV for 20 years by the 1970s. It went on to run another 10.

Even though I was just a kid, it didn’t take me long to figure out that Gunsmoke was mostly empty of content — every week it was the same old rehased plots, with guest stars showing up and always the same kind of ending to the show. Folks loved it, but there was nothing new happening. The guys at Gunsmoke were simply trying to see how many television shows they could make and sell before somebody shut them down.

This past week, major networks in the United States announced that they were shutting down a couple long-running soap operas. These things had been going for decades. “All my Children” ran for 41 years. “One Life to Live” ran for 43.

There’s a new type of show concept, however, used in “Babylon-5″ and “Battlestar Galactica” The show has individual episodes, but each episode is part of a long serialized story. The story has a beginning, middle, and finally, an end.

What a great thing it is to watch one of these shows! It’s something like a cross between a movie and a long novel. Over a period of years and dozens of hours, a long, intricate story plays out. When the end finally comes, it’s really neat.

As a small business owner, I used QuickBooks. I upgrade every few years, because QuickBooks makes me upgrade: they simply make their program break. If I want to continue using it, I have to pay. Doing business accounting hasn’t changed that much in the 15 years I’ve been a customer. After all, a double-entry bookkeeping system and chart of accounts are pretty standard stuff. But QuickBooks makes me pay anyway.

Likewise my tax program requires me to shell out money every year. While the tax code is always changing, completing taxes is basically running an expert system against the data I input. Such a system could easily be maintained in the public domain and would provide a useful service to everybody. Whatever I’m getting, by the 15th time I’ve purchased it, it’s doesn’t feel like it’s worth the money I’m paying.

Microsoft is making tens of billions of dollars from Windows and MS Office. Every couple of years, folks pay hundreds of dollars — for what? To continue to write letters? To have a file system with programs that run in a window when you click on them? Maybe really cool for the 1990s. Perhaps even a bit novel for the 2000s, But at some point enough already. The extra purchase isn’t worth any extra benefit.

I thought about this the other day as I listened to a friend of mine Tweeting about what a pain it was to install program X. He was convinced there was some great value to be had taking 3 or 4 hours of his time doing the upgrade. That may be true, but it got me to wondering: what if I tallied up all the time I’ve spent upgrading various pieces of software. Let’s forget the money for a second, assume I’m made of cash, what about the time involved?

Before I started the list, just thinking about it, it was pretty depressing. All those versions of SQL Server. All those versions of Windows. Of Office. Of various programming languages. Of various programming toolkits. Of various graphics programs and toolkits.

Wow! It could easily run into the hundreds of hours. Over a decade or two, perhaps thousands of hours. Each of those installs promised some new and shiny future: faster query times, better debugging, being able to mail merge a list of friends. These features sounded pretty cool when I read about them in a magazine or imagined how awesome they would be. But inevitably what I used them for was pretty much banal stuff: writing a letter, coding up a business tier, creating a button for a website. If I had a list of all the cool new features from each of those hundreds of installs — a list that would run into the tens of thousands of items — what, exactly, would I be using today to make somebody’s life better?

Somewhere there’s some guy who is able to create an application in 10 minutes and 3 lines of code with one arm tied behind his back using his favorite toolkit. This guy spent ten thousand dollars and three months learning how to do this. With this cool new awesome super-power, he will write exactly 1 app (if he’s lucky), and then next year it’ll be all new stuff. He will start over.

In between that time, there will be a lot of “noise”. Customers will have requirements. He’ll have to maintain existing systems. Perhaps he’ll be doing sales, or tweaking a website to bring in more contacts. New people might have to be trained. There will be seminars and business development assistance, customer relations, re-architecture discussions, and patches and upgrades to existing code.

I’m not the sharpest knife in the drawer, but it occurs to me that all of this “noise” is really what business is all about. It further occurs to me that in software, if you are making a solution for somebody, it pays to never make them completely happy. People are making a lot of money distracting us from the real purpose of what our technology tools are supposed to be doing. There is always be some magic bullet they’re brining out next quarter. All to do the one simple thing that we were basically happy with 20 years ago.

Everybody wants to do Gunsmoke and nobody wants to do Battlestar. They want to get on the gravy train, find the cash cow, and then milk the hell out of it. That would be a hoot, except I’m beginning to feel a lot like the cow.

Enough of this already. Write it up, make it configurable, publish the spec, and move on. End it. Do the right thing. Treat people as you would want to be treated.

More value, less flash. More substance, less glitter. More solutions, less fantasizing.

More Battlestar, Less Gunsmoke.

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

What the hell happened to Apple guys?

Picture from the Apple commercial: I'm a Mac. I'm a PC.

Remember how Apple guys used to be? Hip, smart, creative — perhaps a subtle hint of smugness and self satisfaction — but fiercely free thinkers?[1]

I loved those guys.

Back when Intel announced plans to add an identification number to all of their chips, so web traffic could be traced, they were up in arms: who did Intel think it was, anyway?

Back when Microsoft announced Paladium, the plan to lock down every pieces of the hardware and O/S in an encrypted lockbox, they were taunting: Dudes, we’re not participating in your evil empire thing. We’re on freaking unix for cripe’s sake. You windows guys are just little drones for Bill Gates.

They eat locally-grown organic food, they are cool artists, they are always ready to rail against government intrusion in our lives.

Now that stories are circulating that you can take a brand new iPad or iPhone out of the box, walk around town for a few days, then get a list of where you’ve been? It’s like all my favorite Apple people have become drones for Steve Jobs.

It’s not that bad, they say. Maybe it’s bad, but it’s for a good cause. Ok, it’s bad, but Apple is fixing it. I’ve even heard “It’s a feature, not a bug! Please ask Apple to keep this!”

It’s like I woke up in some bizzaro world. Next thing you know, they’ll be holding Republican campaign rallies and showing up at tractor pulls.

Dudes. Apple was keeping a list of the times you came within range of hotspots. Then they were uploading that data. And you didn’t know about it. End of story. Full stop.

When pressed, they say they are not tracking you and never will track you. They just keep a time-stamped list of places you’re near. It’s a big difference. Then they send themselves this list, slice it up, and send it back to you to help your phone work better. All this noise you see on the net? It’s all just sensationalist journalism.

Nothing to see here, please move along.

Apple has a good reason for doing this. Pay no attention to the billions of dollars it stands to make with the data, nope, this is all to make location-aware services run faster. In the future, they promise not to keep the data forever, and they say they don’t track you per se. They also plan on encrypting the data in the future, “adding a further layer of protection.”

Protection from whom? Me? Apple can sure as hell read the data. Cops can read it. People who steal it will figure out how to read it. The only thing that is being protected here is my ability to see just what the frack is going on with my own phone. After it’s encrypted, Apple can put whatever the hell they want in there. I’ll never know the difference, and I agreed to all this when I “signed” the four-thousand page chunk of lawyer vomit that passes for a user license.

And you are thanking them for it?

Well, what the hell happened? Since when did you guys become part of the system?

Here are a couple random news stories on this. One from the daily mail. Another from

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.