Phil's blawg

Lots of people have written lately about how Pinterest was not an overnight success, I had forgotten that my own startup, BudgetSimple, one of the first online personal budget solutions (launched a few months before Mint) had a slow start itself (in my head I remember it always being steady), but then I graphed user growth over time…

Yes, there is user growth in that small line that almost looks like zero, but it’s all under 100 per month (sorry I don’t provide the Y-Axis).  So it essentially took two years to accomplish significant user growth.  If you are wondering what happened around November 2008, it involved a redesign/rethinking of the site (not quite a pivot) and for whatever reason it started spreading semi virally (the huge spike was the result of a feature in a major magazine).

MicroISV or (µISV) was a term coined by Eric Sink in the mid-2000s to describe the concept of “one man band” software companies (which didn’t need to literally have one person).  These companies were similar to what we now call lean startups, but took no funding, were profitable, and who had no greater aspiration, at least initially, to do more then pay the founder’s salary.  In other words, they created a traditional, profitable business.

It was quite a popular concept at the time, with people like Ian Landsman and Andy Brice blogging about their companies, and Joel Spolsky and 37signals as the proof points of the community.  I created a mISV around that time (ChimSoft and to some extent BudgetSimple).

Lately though, you don’t hear about these types of companies anymore.  Sure they are around and doing as good as ever, but why don’t we hear about them anymore, and why is the concept no longer glorified?  

I think the answer is pretty simple.  mISVs were the answer to the .com crash of 2000.  In those tough economic types for tech companies, you couldn’t raise money to mow your lawn, let alone start a company.  Add to that the number of people who were disillusioned with corporate life when their stock options were worthless, and you have the perfect formula for mISV to become a trend.

Now, we’re back in the late 90s again.  Raising money is easy, and even Joel Spolsky’s latest venture is VC powered.  In case history repeats itself though, it’s worth noting that in my quick research, most of the mISVs I remembered are still in business, while some VC funded companies have already flamed out in less then 3 years.

I first heard Josh talking about this concept at SXSW, and it really inspired me to start using their app instead of FourSquare.  Checking in is tedious, but reliving experiences is something I could see myself looking back on, and making it worth the time to whip out my phone when traveling.  

I just tried the new version, and it’s very pretty.  New users will not associate this as a check-in app, in fact the UI makes it pretty non-intuitive to check-in to your local cafe.  Instead the first thing you’ll want to click is the big guide to Pittsburgh.  I think this is great, but I really do miss the ease of checking in to non-popular places, and REALLY miss the checkboxes, as I wanted to check things off my list of Pittsburgh.  I’ll be playing with it more in the next week to get feedback, but if you haven’t used Gowalla in a while it might be worth checking out again.

gowalla:

Nearly three years ago I sat on a beach at Lake Tahoe and watched as a large low-flying cloud hung above the lake, creating the backdrop for a beautiful weekend of solitude. As much as I enjoyed the time alone to clear my thoughts, I was overwhelmed with the desire to share this experience with…

Someone should make a bumper sticker that says something along the lines of:

“I’d love to hear more about why you think my technology stack sucks, but I’d rather be launching”

As I mentioned in my last post, people are usually wrong about what is correct to do financially.  For example, “common knowledge” is that a Savings account is the safest place for your money, and the few people who do have savings in our country probably have it sitting in one of these accounts.  But look at the Interest Rates.  Right now ING, one of the higher paying banks, pays 1%.  My local bank, PNC pays .02%.  Yes you read that right.  Interest rates are so bad, I did a little math, and it turns out that if you had invested your savings in Forever postage stamps when they were first issued in 2007, and sold them today for the current price of a stamp, you would have came out ahead of using a savings account!

How is this possible? Well first, the current rate of inflation in the USA is around 3%. So you should know that any investment getting less then 3% is essentially losing you money. The price of stamps, like everything else has had to increase because of inflation. Something that is interesting though is that the Government started issuing Forever stamps in 2007.   You buy these for X and they are usable forever as valid first class postage.  From that point on, these essentially became an asset that gains value instead of a fixed priced asset.

Just for kicks I did the math and figured out that if you bought 10,000 Forever stamps at a cost of $3900 in 2007, you would be able to resell those today for $4400. On the other hand if you decided to put $3900 into an ING Savings account, you would only have $4314 now. And ING pays considerably more then most local banks do!  Note that in 2007 ING paid 4% interest.  If it had paid 1% all along, Stamps would have been a much better investment.

My math is below if you are interested. Do not construe this as financial advice, talk to your financial advisor about Postal Stamps before investing.

I was reading a book the other day that was giving tips on how to have sound finances, not be in debt etc… it had a complicated series of steps, and I’ve read many books like this, but I realized it’s much simpler.  Essentially all you need to do is this:  The exact opposite of what “most people” do.  

In other words, “most people” are in debt, most people have no savings, therefore if you look at what most people are doing, it’s a pretty good guide of what a bad idea is.  So most people buy new cars because most people believe a car payment is a fact of life.  Most people buy a house heavily mortgaged, and then use any equity they get to put more money back into the house.  Most people have pets, most people buy new clothes monthly, most people have cable, etc… If you are like me, you’ve wondered how most people can afford all of these things when they are working for the man for less money then we in IT make.  Well, the fact is they can’t.  Which is why when the economy turns slightly sour everyone starts going bankrupt and getting foreclosed.  All of the things I just mentioned which define the American dream are liabilities that cause money to fly out the window. Most people have a negative or close to zero net worth.

If the common wisdom is you should get into Real Estate and an ARM is great because property can’t go down, don’t get into Real Estate.  If the common wisdom is that the stock market is a great place for your money, don’t buy stocks.  On the other hand, when the economy completely tanked in 2008, I opened a Mutual Fund.  If most people are doing something, it’s probably too late and it’s time to get out.  Most people make very poor financial decisions, and most people live in fear and uncertainty.

The converse isn’t guaranteed to make you rich though.  Just because most people are not buying corn futures, or Greek bonds doesn’t guarantee those will be successful. However because it’s more risky, if you do make money, you’ll most likely make more money then a safe bet.  Most people won’t start their own business, because they believe working for the man is safer, but because of that fear, most people also won’t be rich.

If everyone is questioning a decision you’re making because it goes against common sense, just remember that most people, by taking what they think are safe, common decisions, are actually making very poor financial choices.  The rich get richer for a reason, they are doing things the rest of us are not.

Lately I’ve heard some Internet famous people like Rand Fishkin and Joel Spolsky complaining about issues with not being able to be logged into multiple Google Apps accounts and/or Gmail at the same time.  My response to this is: “You’re doing it wrong”.  

It seems painful to me that people actually login to all these different accounts just to check email.  You can have a single Gmail box work for everything, and all of the drawbacks that existed in the past are gone these days.

Here’s how I roll:  First, login to all Apps accounts.  Go to Settings->Forwarding and change the account to forward all of your email to your primary Gmail, while still keeping a copy in your inbox (just in case!)

Now, you’re done with the silly Google apps account, and you can essentially never login to it again.  Ok, well that’s fine you say, but now I’ll have all sorts of crap from my 50 different accounts going into a single Gmail account, how do I keep it organized?

First, go to your primary Gmail account where you’ve forwarded everything.  Go to the settings area, then “Filters”.  Create a new filter, and make the filter for all emails where the TO address is the one you’ve forwarded from Google Apps.  Hit next, and apply a meaningful Label to all emails matching that filter.  Personally I have a label for each company/domain.

Now, whenever anything comes in, you have a colorful little flag telling you what business it relates to!

Ok, but that’s not all, what about when you want to reply to an email?  First, to be able to send email “AS” another address other then your Gmail address, you need to set it up in Settings.  So go to Settings->Accounts and Import, then click “Send mail from another address”, type in the email address you want to send mail as and click Next.

Now it used to be that if you set Gmail up like this, your primary Gmail address would be exposed to Outlook users with a “On Behalf Of” notice.  Gmail finally fixed this, but you have to configure things a little different.  Check the radio button saying you are sending email from your own domains, SMTP servers, and click Next step.  Configure it as seen below (but with your email and username/password obviously!)

Now, when you reply to any emails sent to your Google for Apps accounts, Gmail will automatically pick this as your from address!  

Hope that helps take some of the pain out of using multiple accounts with Gmail.  I have something like 6-7 Google Apps addresses I use, and this system works very well for me.

Yesterday a story blew up on Hacker News when Dave Gooden reported that he had proof AirBnB was using black hat spam techniques to grow their user base.  I won’t comment on that situation as I’m sure there will be many other arguments for and against the techniques AirBnB used, but there is one thing I think all startups can take away from what happened:

Targeted communication works extremely well.

Sometime in the past ten years spam has gotten bad.  I mean REAL bad.  I don’t mean the problem of spam has gotten worse, because I almost never see any spam in my inbox (thanks Gmail!), but the quality of spam has gotten downright atrocious to the point that almost no one is enticed by spam and they must rely on sheer odds.

Let’s look at one or two examples currently caught in my Spam folder and hope Google doesn’t blacklist my blog for this content!

From: emai1l@gmail.com

Subject: NAoKQAFgml

 comment4, <a href=”http://adipex.mijnbb.be/”>adipex physician</a>,  363,<a href=”http://551513.xobor.com/t6f8-valium-online-information.html”>buy valium suppositories</a>,  =OOO, <a href=”http://551513.xobor.com/t1f3-soma-online-information-online.html”>pills called soma</a>,  70942, <a href=”http://551513.xobor.com/t10f12-zithromax-online-information-online.html”>calcium blockers and zithromax</a>

Really?  Spam is essentially a sales letter, but how would this sale possibly convert!  Besides the fact that it’s utter gibberish, it’s completely untargeted!  I don’t buy any medication online, and am not in the market for calcium blockers or valium suppositories!  I can only think this converts by sheer random chance, IE if you send it to the 2 billion people that have emails, the odds are in your favor that some very small portion of those 2 billion will accidentally click the link, be interested, and be stupid enough to give their credit cards to some russian site.

For some reason the other half of my Spam folder is always Russian spam.  For example:

Установка кондиционера по выгодным ценам.

Гарантии. Монтаж в любое удобное время для вас.

Стандартный монтаж кондиционера 05-07 до 1,8 квт - 9200 руб.

According to Google Translate this has something to do with selling me air conditioning.  Well, at least it’s legible I guess.  But the proportion of people who speak Russian is very small, so this email is also very very poorly targeted.

I’ve always wondered how well Spammers would do if they just resorted to the old techniques we saw in 1996, which was… just send a well worded sales email!  It looks like AirBnB has done just that to great effect.  They were probably not picked up by any spam filters because:

1. They used a real email address 
2. They had extremely targeted content 
3. They had a very high conversion rate.  

Now if they just did a blanket spam to anyone with an email address, which seems to be the current spammer technique, I suspect they would get blacklisted so fast, the people for whom the content was relevant would never see it.  The way AirBnB did it, you could argue that it was hardly Spam at all because the receivers seemed to welcome the email!

So I wonder, why don’t spammers just put a little more effort into their sales pitch?  Maybe do some research and only contact people who are likely to WANT Viagra.  If the pitch is in English, send it only to ENGLISH speaking people.  Don’t just include a bunch of fake keywords, tell people WHY they should buy your crappy watches or illegal drugs!

We should all take a lesson from this and remember that quality, is much better then quantity.  Focus on converting quality leads, not dumping junk on everyone.

Amazon web services east coast has been down basically all day today.  It’s an amazing chance to see just how many companies rely on this single point of failure to run their businesses.  Even more interesting to me is to see how the effect cascades:

AWS down -> Heroku Down -> Your startup using heroku down -> Third party apps that use your API down

Some sites like HootSuite and Reddit have been kind enough to at least let their customers know that they have not fallen off the face of the earth, but it’s quite surprising to me just how many Heroku hosted companies are just….down today, like its a snow day.  

If you go to a site like this, you’ll see some latency then *POW*, rails error (or sometimes nothing comes back at all).  First, if your site is still just down 10 hours into this AWS failure, you clearly haven’t read my article on multi ISP redundancy, but that’s ok.  It’s not too late to do something about this!

Even though all of your friends on Reddit know that half the Internet is down today, it doesn’t mean your customers (or more importantly, potential customers) know Amazon is having a problem.  It’s crucial to have a plan B in place.

The minimum plan B you should have is a “Fail whale” type page letting people who come to your site know that your host is having problems but you’ll be back soon.  You can accomplish this by having a low DNS TTL (30 min or less), so when you have downtime, quickly bring up a light host with a status message on another provider like Slicehost or even WordPress, and switch your DNS as soon as it’s clear the problem won’t be resolved in an hour.

If you are really savvy, you’ve been doing backups of your site and database nightly to a different provider, and can at least bring up a “read-only” version of the site (as sites like Reddit have done now).

We’re getting to the point now where Amazon will need to enact their 10% refund policy to many people, but that most likely won’t cover the lost business and good will of you being down all day!