Sean’s blog

One of my employees did something slightly awesome yesterday.  He made a good relationship with semi-popular blog and arranged to write some content for it that highlights the products that we sell.  It’s a great situation all-around - great content for the other site, great for us as a business, and interesting work for the employee.

I wanted to make sure that he know his ingenuity was appreciated, so I made sure to explain to everyone at the company (we are small) what the employee did and why it was smart.  I also stopped by the bank and got him $100 cash so that he could do something fun.

We occasionally reward our employees that way - with “surprise” bonuses - sometimes cash but more often a videogame or something a little more thoughtful.  I felt a little weird after giving him the cash, since he was already really excited about his work in a more intrinsic way and I didn’t want to diminish his intrinsic satisfaction.

Then I read this article “Why Incentive Plans Cannot Work” -in the Harvard Business Review way back in 1993.  It is highly salient to startups since, particularly those that are venture-backed, use incentives - stock options, bonuses, etc. to motivate behaviors.

Some highlights from the article:

“Research suggest, by and large, rewards succeed at securing one thing only: temporary compliance. When it comes to producing lasting change in attitudes and behavior, however, rewards, like punishments, are highly strikingly ineffective”

“In general, the more cognitive sophistication and open-thinking that was required, the worse people performed when working for a reward.”

“Jenkins tracked down 28 previously published studies that measured the impact of financial incentives on performance.  His analysis ‘financial incentives’, published in 1986, revealed that 16, of 57%, of the studies found a positive effect on performance”

“Rewards have a punitive effect because they, like outright punishment, are manipulative. ‘Do this and you will get that’ is not really different from ‘do this or heres what will happen to you’ … managers are creating a workplace in which people feel controlled, not an environment conducive to exploration, learning and progress”

“Rewards, like punishment, actually undermine the intrinsic motivation that results in optimal performance”.

The content of this article really “feels” right to me, as a manager.  It is also completely contrary to much of what is practiced in venture-backed startups.

Today hasn’t been a good day. First, my business partner and I got our taxes back from our accountant - not pretty. The problem with bootstrapping a “profitable” business that requires working capital is that increases in working capital are considered as profit by the IRS. That may well be true, but it really makes it hard to grow such a business. While our taxes are about 30% of profit, they are well more than that of our cashflow. Ok, whatever, I can live with that.

However, I was also informed by the government today that I must garnish the wages of one of my employees for $8000 of student loans that he didn’t pay. Now, he didn’t get a degree and it’s not like he benefited at all from the somewhat sketchy school that he was attending. My take: that school probably shouldn’t exist and certainly shouldn’t charge what it does, the bank shouldn’t have lent him the money, the bankruptcy laws shouldn’t be such that student loans cannot be dispatched, and I DIDN’T SIGN UP TO BE A BILL COLLECTOR. Hopefully Paychex knows what to do with this, because I have no idea. I’m not happy about it.

Things could be worse with our government, but they certainly could be better. Movie recommendation - Maxed Out.

I recently had the opportunity to dig around in the marketing effectiveness data of a friend’s company. The results were really interesting, so I figured I would share. This company is an ecommerce company that sells a physical product to consumers online. They sponsor several blogs, one forum, do a significant amount of offline advertising and spend on Google CPC ads. Conclusions:

1. Over the time period organic search on Yahoo and Google generated $100Ks of revenue and cost nothing

2. Google adwords returned $18 of revenue for each $1 spent on ads, around 18x the effectiveness of the offline advertising.

3. The rates on the blogs are not efficiently priced, there is wide variance in cost -effectiveness across blogs.

One issue with online advertising is that there is a limited audience. It is often argued that you reach an entirely different set of potential customers when advertising in a different medium and that you cannot reach the same volume of people online as you can offline. Therefore, while online advertising may have a higher ROI, offline advertising is still important. That may well be true for some very mass-market things that don’t get a lot of search attention, or for companies with a much larger marketing budget, but in this case the company is spending most of their ad money on offline media and getting the majority of their visitors from online marketing.

Someone just asked me how I came up with the offline advertising impact. I used a multiple regression model that I will write about tomorrow.

Another modification - it is important to note that I used the direct sales impact from the online ads and the indirect sales impact from the offline ads.  This isn’t really fair and it actually underrates the effectiveness of online ads (because they have an indirect component as well) - see this HBR article for interesting data on the indirect effect of online ads.

For a while last year I was going to Columbus, OH every week for business and I couldn’t help but notice the huge ambitions of local airline startup Skybus. I have never flown them, but they seem like they are doing interesting things. The most interesting, in my opinion, is flying to really unusual airports such as Chicago/ Gary Airport(GYY), Portsmouth NH (near Boston, Portland and Manchester), Wilmington DE (ILG) which is near Philly and obviously Wilmington, and Niagra Airport (near Toronto).

Airport fees are an overlooked but significant part of airline economics. For example, they were 5.7% of revenue for Southwest in 2007 and 6.3% of revenue for JetBlue in 2007. In many cases Skybus will be the first commercial airline serving these 3rd-tier airports and it seems as though they will get bargain-basement rates. For example, Niagra Airport advertises rates 20% those of JFK for a 737 or A320. Part of the business model of the first discounters (Southwest, Jetblue, Airtran) was flying out of 2nd-tier airports such as Midway (Chicago), Manchester and Providence (Boston), Love (Dallas), Hobby (Houston) and Oakland (San Francisco) where there were lower landing fees and less congestion. Those airports have grown a lot and are now facing some of the problems faced by larger airports.

I am a big fan of alternate airports here in Chicago and I prefer to use Midway over O’Hare since it is 30 minutes by EL vs. 60, 45 minutes in traffic vs.90 and 20 minutes in no traffic vs. 35 from where I live in downtown Chicago. It also has more on-time departures, better dining (potbelly!), and less cancellations (0.02% vs. 0.12% of flights in 2007). I am not sure that I would use Gary since it is quite a bit further than Midway, but if it had less delays and cheaper flights I might. It is also only 45 - 60 minutes away by the excellent South Shore train.

I found some interesting data comparing landing fees at many of the world’s airports. The data was in a PPT published by the Air Traffic Research Society.

Landing Fees for a 747 at North America Airports

Landing Fees for a 747 at Asian Airports

Landing Fees for a 747 at European Airports

Landing Fees for an Airbus 320 (smaller plane, like a 737) at Asian Airports

Landing Fees for A320 At European Airports

Landing Fees for A320 At North American Airports

Per-Passenger Terminal Fees for North America Airports

Everyone is talking about Chris Anderson’s new article about Free stuff - very interesting article.

To me there are two kinds of free:

1. The company invents a way of delivering a product or service at nearly zero marginal cost.  Google and Flickr are examples of this strategy  - these are the good guys.

2.  The company invents a way to bill the customer in a sneaky way for extra little crap, allowing them to offer the main product for free.  Anderson’s example of Gillette and razors is an example of this strategy - as are bank overdraft fees, change fees on airplane tickets, the “Free Credit Report” ads that you see on late night TV, the extra minutes that are billed at a much higher rate than the minutes included in your cellphone plan, Blockbuster’s inflated late fees (before they got sued).  The list of examples in this section is longer because established businesses with lots of market power are more comfortable extracting value than creating value. These are the evil guys.

It’s really simple, record industry - music is an emotional, impulse purchase.  ITunes is a winner because it lets me buy music and listen to it right away.  Ordering from Amazon doesn’t cut it because I need to wait a few days and deal with the unpleasantness of having UPS deliver (no doorman) and I can’t remember the last time that I stepped foot into a mall (do they still have musicland?) or a Wal-Mart or a Best Buy, it has been years.

Until two years ago I had gone 6 years without buying any new music.  I borrowed from friends in my dorm when I was in college and then listened to sirius.  Lately I’ve been buying at least a (virtual) record / week.

I am not sure why this is so difficult.  Maybe it’s because the record company folks have all their music delivered to them and never have to seek it out or actually deal with procuring it.

I’ve been telling my friends for quite a while that EBay is a good short idea. As the owner of a company that sells both on ebay (about 6% of sales) and off Ebay (the rest), the Ebay experience isn’t very good, except for selling very high-volume, just released electronics. Anything nich-y usually doesn’t sell. I think it’s because the navigation is rather difficult compared to amazon or googling to find an online shop. Hiking fees will be great for profitability in the short-term but doesn’t bode well for their long term prospects (craigslist and froogle are free).

Chicago has proposed increasing the transfer tax on sales of property, for example the buyer of a $300,000 home today pays $2,700 in transfer taxes. That amount would rise to $3,600 under the plan before the City Council. As someone who is in the market for a new place (as soon as the slide is complete), and who generally doesn’t like taxes, I am not a big fan of this.

However….it is to pay for the upkeep of the public transit system, which is in perpetual financial trouble.  I know from experience that the public transit system is not well run.  However, having workable public transit is essential to Chicago’s competitiveness - without it my employees cannot get to work.  Public transit is the only way that we can maintain the population density that contributes to making  us a world-class city.

If we cannot reform the CTA to make it more efficient we have no choice but to pay up to keep it running.

This is a fascinating article about how the Tipping Point theory might be wrong - Is the Tipping Point Toast?  This new theory feels like there is some truth in it to me, similar to the post I wrote a few weeks ago Why Viral Marketing Doesn’t Make Sense, however, I did find this article from Guy Kawasaki’s blog, which would seem to disprove the new theory.

Excerpt:

Why didn’t the Influentials wield more power? With 40 times the reach of a normal person, why couldn’t they kick-start a trend every time? Watts believes this is because a trend’s success depends not on the person who starts it, but on how susceptible the society is overall to the trend–not how persuasive the early adopter is, but whether everyone else is easily persuaded. And in fact, when Watts tweaked his model to increase everyone’s odds of being infected, the number of trends skyrocketed.

“If society is ready to embrace a trend, almost anyone can start one–and if it isn’t, then almost no one can,” Watts concludes. To succeed with a new product, it’s less a matter of finding the perfect hipster to infect and more a matter of gauging the public’s mood. Sure, there’ll always be a first mover in a trend. But since she generally stumbles into that role by chance, she is, in Watts’s terminology, an “accidental Influential.”

Perhaps the problem with viral marketing is that the disease metaphor is misleading. Watts thinks trends are more like forest fires: There are thousands a year, but only a few become roaring monsters. That’s because in those rare situations, the landscape was ripe: sparse rain, dry woods, badly equipped fire departments. If these conditions exist, any old match will do. “And nobody,” Watts says wryly, “will go around talking about the exceptional properties of the spark that started the fire.”

I co-founded an online retailer a few years ago - we have a pretty large customer list (approaching 100,000) and we like to send them emails from time-to-time (maybe 6 times / year) because to tell them about new products.

There are several services that help manage such mailings. The three that I am most familiar wth are Mailchimp, IContact and Constant Contact. My company chose IContact because, at the time, we found it easier to use than Constant Contact and had not heard of Mailchimp. The advantages of using such a service rather than just sending out the emails yourself are: 1. deliverability - their mailservers are on whitelists to ensure that more of your emails get through 2. tracking - they track clicks and opens 3. managing the usubscribes so that you don’t make your customers angry by sending them stuff they don’t want 4. templates for helping to create attractive mailings.

With a list as large as ours, this service is a considerable expense for us - $699 / month - almost as much as we are paying for our dedicated server, bandwith, etc. which seemed a bit out of whack to me. I wanted to see if we could create a system with comparable functionality ourselves and spend the savings on something else.

I felt pretty sure that we could replicate #2 (tracking) and #3 (managing unsubscribes), and even improve a bit on the tracking by seeing what sales result from the mailings (in addition to clicks and opens, which all of the software packages described here track). We don’t really need #4 (templates and formatting) since we already have a template that we like and we prefer using Dreamweaver over the javascript html editors that the services use.

However, #1 (deliverability) was a big question mark. If the emails don’t get through when I send them from our own servers then it could be worth the extra expense for one of the services. I decided to run a test, but I got a bit carried away. While I was at it, why not test the deliverability of all the services relative to each other, as well as relative to our own solution.

It cost me a few hundred bucks and some time to do this, so I am posting it in hopes that somebody else can benefit from it as well.

For each of the 4 treatments (homemade and the 3 vendors), I chose about 3000 customers randomly from our mailing list. It wasn’t exactly 3000 because of the way that I randomized the list. Then I asked my friend Simone, an economics PhD student at MIT how I should run the test. She replied:

An easy way would be to run a regression with response=”yes” on the left hand side and dummy variables equal to one to indicate whether than customer got treatment X (so if you had 4 different treatments, you’ll need 3 dummies). The coefficients on dummies will tell you if the relevant treatment is significantly different from the response rate for the omitted treatment.

Ok, that’s easy enough. I sent the same message to all four lists at the same (rough) time, midnight on Tuesday 1/22. Then I waited until today (3.5 days later), downloaded the results from each of the services and aggregated them manually in Excel. Here were the results:

These results show that the difference in deliverability for the four treatments was small, but it is not clear whether the difference is statistically significant, so I ran the regression that Simone recommended. These were the results:

Coefficients - the first bolded column, shows that each of the services achieved deliverability slightly worse than my own solution. P-value - the second bolded column shows us how certain we are of the coefficients. For example, the 0.40 p-value for constant contact shows that we are 60% certain that it actually performed worse than the control (my solution). In other words, the negative result for constant contact was not statistically significant. Actually, none of them were. IContact came the closest - we are almost 90% sure that it actually performed worse than my home-made solution.

Conclusion: I am going to start sending out my email myself and save the $700 / month that I was giving icontact.

Another wrinkle - I told my hosting company about this plan and they were not big fans of it:

8:48:25 AM Sean: can you imagine any issue with me using php tosend 75000 emails from my box?
8:48:44 AM gino: you bet your ass volume alone will get yout blocked
8:48:55 AM Sean: by whom?
8:49:13 AM gino: unless you have a great personal approved relationship with alot of the major isps, they will block you based on volume -
8:49:48 AM gino: they may let 500 in wihtin x time, but the rest will get blocked and eventually they’ll see it as abusive and blacklist your IP and the range even - whcich puts us all in jeapardy
8:50:07 AM gino: can you prove these 75k are clients or have subscribed or doe busienss with you?
8:50:14 AM Sean: of course.
8:50:17 AM gino: and is the email being sent can-spam compliant?
9:01:09 AM gino: they have an email relationship specialist on staff - and several load-balanced clusters each with its own distinct IP range and they slowly dribble out the emails over time -
9:06:50 AM Sean: but if i think about it differently and instead of sending out 75,000 at a time, but instead send out
something to each customer 1 month after they have shopped with us, that could work
9:07:03 AM Sean : because i would be only sending like 200 / day
9:08:16 AM gino: yeah much more likely to work

So the new plan is to re-arrange my mailing strategy, so that we no longer send out clumps of emails, but rather gradually tell customers about new products. In some cases, that won’t work (when something really hot and cutting-edge comes out). In that case I will use Mailchimp, which has per-message pricing. I think I can cut these clumpy mailings to 2 / year, at their pricing of $0.005 / email that’s only $1000 / year, must less than the ($699 * 12 = $8,388 / year that I was paying previously).

I am working on putting some analytical rigor around many aspects of this ecommerce business. If anyone else has done testing of this sort and is willing to share, I would really like to speak with you, maybe we could collaborate and save ourselves some work.