ezl

Subtext: I’m probably not cheating on my taxes, except for the fact that I haven’t done my taxes yet this year…

Benford’s Law is an observation that in many numerical data sets, the frequency distribution of the first digit of the individual data points follows a particular (power) distribution.  In particular, there are a lot more numbers that start with 1 than 2, more that start with 2 than 3, more that start with 3 than 4, etc. There tend to be ~30.1% 1′s, ~17.5% 2′s, and only 4.6% 9′s.

This is counterintuitive to many people who initially guess that the distribution of first digits might be nearly uniform.

This phenomenon doesn’t apply to all numerical data sets, but it does seem to apply reasonably well to financial data, and as a result, is used as one tool to aid in detecting tax fraud (people who make up fake data but don’t know about Benford’s Law often make up random numbers, but those are closer to uniform, or at least vary from Benford’s in detectable ways.

I thought Benford’s Law might not fit my financial data well since I have pretty atypical spending habits (or so I thought), and since track all my expenses, I decided to dump my data and check.  Turns out my spending habits are pretty damn close, over 1365 transactions from all of 2012 + Jan->May 2013.

2012 Expenses (Full Year)	Count	%	Benfords
1	267	28.7%	30.1%
2	167	18.0%	17.6%
3	143	15.4%	12.5%
4	86	9.3%	9.7%
5	67	7.2%	7.9%
6	80	8.6%	6.7%
7	46	5.0%	5.8%
8	45	4.8%	5.1%
9	28	3.0%	4.6%

2013 Expenses (Through May)	Count	%	Benfords
1	134	30.7%	30.1%
2	102	23.4%	17.6%
3	48	11.0%	12.5%
4	33	7.6%	9.7%
5	30	6.9%	7.9%
6	32	7.3%	6.7%
7	25	5.7%	5.8%
8	14	3.2%	5.1%
9	18	4.1%	4.6%

Another quick status update. I’ve been awful at weekly blogging, because I pretend to be busy, but I’m really watching The Mentalist and wishing I was still in Colombia, but despite that it’s been a great growth trajectory since I last wrote about my stats.

Keep in mind the last set of stats was during the slow season, when I expending zero effort on marketing efforts, and about to launch several campaigns for growth.

In the last month Rocket Lease has done:

Revenue: 4305
Profit: 1959.63
Credit Reports: 177
Customers: 65 (65 different landlords/property managers have processed paid applications through us)

Paid Applications: 5.7 applications/day (up from 1.7/day), 177/month (up from 51/month)
Revenue: $138/day (up from $47/day), $1410/month
Gross Profits: $1959.63/mo (up from $600/mo)

Total applications processed : 358 (we also offer free applications)

Biggest source of traffic that ends in paid customers to date is Google, and I’m looking for a way to expand that.  In the last 3 months I’ve experimented with

• Display advertising on real estate centric sites (bigger pockets, active rain) both mostly failures
• An alternate rental application approach to bring people into the funnel, marketed through landlord and property manager organizations (seemed promising, but sales cycle was long and its hard to catch people when they are in “buying mode” — seems more effective to find other ways to try to capture them when they are *actually* interested
• Ad retargeting (I used AdRoll, seems awesome. my CPC there is still pretty high, but it’s pretty inexpensive and I love it when people tell me they’re seeing my ads — admittedly not necessarily a positive ROI, but i’ll just run for the “it’s good branding” excuse that non-metrics oriented marketers seem to love).
• Traditional sales calls to property management lists (failed again)
• Content marketing strategies with blog crossposting (mixed results, I was hoping this would lead to a larger google footprint, but I’m honestly taking a very naive approach to SEO).
• Split testing various components of the site (honestly, this is still premature because my funnel isn’t even big enough, but it’s just so fun — one interesting thing was that removing my explainer video actually boosted my signup rate by 53%…)

I have been slowing down a bit, but a few things on the to do list going forward are:

• I built WAY too much complexity into RL. It’s capable of SOOO much, but people really want less. If i stripped out half the components, I’d serve 90% of the users still and be able to have way less documentation and support. I’m considering taking the codebase and building an alternate, simpler application process.
• Alternate landing pages. Let’s be serious, the landing page could use a LOT of work.  I’d like to design a better one, a long form sales one, and split them. I also want to try to start migrating people towards a simpler signup process, but that is a little more involved than a simple A/B test to determine the worth.
• Another push for SEO — Trying to loop a great SEO consultant into the operation to own that.

All in all, happy with progress, but I’d like to find a way to grow this about 4x more still.

Another positive note : I use customer.io to do behavioral emails based on where in the signup flow my leads are, and its great because I get feedback from customers on a consistent basis.  It’s always great to get notes from customers telling me how much they love Rocket Lease and how they’re so glad they found it.

Makes me all warm and tingly when people tell me they love something I was a part of.

We process tenant screening reports for property managers and landlords.

That include credit reports, criminal background checks, and court eviction records.

Generally, the most important of these is the credit report. The 2 most common patterns are:

1. Request ONLY a credit report ($15)
2. Request a package of credit + criminal + eviction ($35)

In the 53 days from 1/1/2013 to 2/23/2012, we have processed:

• 57 “packages” and 34 “credit only” requests for a total of 91 paid applications processed
• At least one request on 43 out of 53 days
• from 29 customers (15 old customers who joined the site prior to 1/1/2013, 14 new customers who joined the site after 1/1/2013)
• New customers found us: 9 via google, 2 via yahoo, 3 going directly to the site.
• Old customers generated approximately 2/3 of that volume (138 reports, 59 credit VS 67 reports, 32 crd)
• For a total of $2505 in gross revenue
• After the vendor cost of reports and stripe payment processing fees, that results in
• $1064.27 gross profits left to pay cost of operations (hosting, tools like ticket and repo hosting, chat tools, customer.io, analytics tools, advertising, staff, blog writing, office space)

That works out to:

• Applications: 1.7 applications/day, 51/month
• Revenue: $47/day, $1410/month
• Gross Profits: $20/day, $600/mo

To start breaking even, this needs to become ~8x bigger than it currently is.

That sounds a bit scary, but in our biggest month to date was Nov 2012, we did 209 applications, which is half way to the goal line. At the time, we were running a bunch of different approaches to acquire users, but most of them were naive and unprofitable.

Currently running a few new ideas for getting better sustainable traffic/growth, including re-implentations of 2012 failures. Excited about what’s coming up, though also getting tired of the frustration of seeing so many failed tests. I’m definitely still bullish, but it can become tiresome because it feels like it’s taking forever.

I unfairly get angry at the customer service representative whenever I have to enter information into the automated system and then the first thing the customer service rep does is ask me for the same information.

It’s really not the customer service rep’s fault. They’re just hourly wage workers and it’s really bank of america/verizon/tmobile/some shitty company’s fault that they don’t transfer the information they request.

It still irks me though and even though i know it’s not the fault of the person on the other end of the phone, my patience is always a little shorter.

Big companies: if you ask me for information, will you please just fucking transfer that information to the representative?

Here are some ideas I like, but am not pursuing.

These ideas are, of course, free for the taking, if anyone reads these and cares. I’ve thought about them more than the text indicates, so if anyone wants to execute, shoot me a note and I’m happy to chat further.

Put together family plans for people (cell plans, data plans)

Individuals pay as much as a 50% premium for having an individual plan instead of a family plan. No cell phone carrier requires any sort of restrictions for being a “family” other than your assertion that you are a family (doesn’t have to be same household, last name, or anything). You and I could be on a family plan.

Just a quick survey:

• http://www.att.com/shop/wireless/plans/familyplans.html vs
• http://www.att.com/shop/wireless/plans/individualplans.html

5 people is the max on a family plan. the unlimited plan divided 5 ways would be:

• (50 * 3+119) / 5 == 270/5 == 54 bucks
• the individual unlimited plan is 70 bucks.

Win.

• If you were to divide up the 2100 minute plan (420 minutes each),
• (10 * 3+110)/5 ==140/5 = 28/month
• Compare to the 450 minute individual plan for 40/mo.

Win.

Additionally realize that with any fixed, finite amount of minutes, most people don’t spend all of it. Cell phone providers make a lot of money because they’re effectively dealing in “overage”. If I sell you 450 minutes, you’ll almost certainly use some amount OTHER than 450. If you use 400 minutes, I pocket a free 50 minutes (not always, with “rollover minutes”, but whatever). If you use 500 minutes, I get to charge you for 50 minutes at some obscene overate rate. bundling packages helps people save on this front as well.

There’s a bit more complexity than that due to data and text plans, but you get the picture. There are also other advantages to grouping people into family plans (for example, no contract breakage fees – you could get on a family plan for 3 months, then bail)

this “business idea” is just a website that pairs people together to get into family plans with people they don’t know. you could be VERY involved (actually broker the transactions, take money, and have a ton of plans yourself), or decentralize it and just connect people who are looking to save money. Either way, you create a lot of value, so you can charge the end customer a bit.

Gym waiver forms

I do crossfit, which if you don’t know what that is, its just another gym. Crossfit gyms (“boxes”), dance studios, rock climbing places, karate dojos, etc all make new members sign a waiver of some sort. It’s usually a crappy piece of paper that has been photocopied many times over, is hard to read, low fidelity, stored in a manila envelope in the office, and then after its signed, the gym owner files it away in a cabinet. No big deal. But they suck.

Create a simple version of the online signing interface at http://signnow.com, specialize in just dropping a signature and date field into an HTML document for gym owners and charge them $.50 a signature. In return, they get electronic copies of their documents, you host/store all the documents for them (“in the cloud!” people love hearing that…). their docs are searchable and they can find whatever they want, whehever. They only pay for usage and its cheap. It automticaly emails copies to the owner and the new member. new members can just go to: gymwaiverform.com/mygymname on their phone to get the waiver, fill out a few html fields, drag their finger on the iphone as a signature, and boom.

the value proposition is cheap, searchable documents. always having them on hand. not having paper in your gym. email copies to you and the end consumers.

Build your own granola bars

Like this site: http://www.youbars.com/buildabar/ – create a sexy UI for building your own bar. This site should be a ONE PAGE app. use backbone or something with a list of ingredients in a checkbox, then do quick animations to make it look like items are being poured into a mold. Do batches of 12, ship them cheap. These are easy to make and cost like 20 cents a granola bar at most. It’s easy to hire people to do this kind of work, shipping is easy, spoilage is minimal. Target audience == office drones who are eating unhealthy crap during the day. Get them all on a weekly subscription plan.

The game here is to connect with office managers on linked in. ask them if you can send them a free sample. tell them its better tasting and better for their employees health than the cheetos they’re already buying. close weekly subscription plans for offices.

Connect dog walkers with dog owners

Tons of dog owners in cities like chicago, NYC, SF, LA pay people to walk their dogs while they’re at work. Create the 2 sided market for this service exclusively. build in reviews and rating systems for both sides (a la Uber).

This may sound ridiculous, but this market is surprisingly bigger than you’d think (I have seen a dog walking companies financials and I was flabbergasted). Charge 5 bucks to connect a pair. Or have them book and pay through the site and charge $1 per walk (dogs are getting walked every day…). Dog owners will probably like the convenience of online payment instead of always running to bank of america to get cash for their walker. also, if their walker is unavailable, they’ll be able to find another local walker that is well-reviewed.

This idea sort of exists. But nobody has traction. If dog owners/walkers don’t know about it, then it effectively doesn’t exist for them

Booze delivery

If you’ve ever been drunk and out of vodka, would you have paid a 5 dollar premium to have it delivered to your door in 30 minutes? Someone will.

Booze is cheap for liquor stores. If you can help them move product, someone will offer you a commission (call it an advertising fee, selling liquor requires abiding by laws, advertising doesn;t).

Drunk people pay a premium for delivery.

so that $25 bottle at the store, could be a $10 fee for the delivery business (5 from the store, 5 from the consumer).

the tech is minimal. a really clean and simple UI for picking between 4-5 choices (absolut, $15; grey goose $35; jack daniels $25, patron, $30; case of budweiser $20), an html form to take in name, address, and an image of your photo id, and a embedded stripe form for processing payments.

at the early stage, this just fires an email to the CEO.

The CEO’s job is to promote this website, get people buying, manage bike messengers/drivers to deliver the product.

After he’s delivered 100 orders, then talk about automation and scaling.

CSS Bounties.com

I suck at CSS. Lots of people are good at it.

Often, in 10 minutes, someone can solve the sort of problem that could take me 6 hours to solve.

Put bounties on CSS problems for immediate solving. Inspired by this reddit thread.

Let the internet solve your CSS problems, collaborate with the folks in #css on freenode.

The person with the problem posts a description, URL to the site, and the desired outcome on the site and an expiration date. They paypal the site with the bounty (minimum $5). They can cancel the bounty at any time before a solution is posted.

If the expiration date hits without a solution, it is returned to the poster.

People can post solutions to problems, if theirs is selected, the bounty is released to that person.

Last year, I taught a class at General Assembly
called ”How To Start A Startup If You’re Not A Developer”
where I focused on helping aspiring non-technical founders
get from the idea stage to an MVP.

Since then, I’ve taught several variations of that course. As a result,
I’ve had the privilege of connecting with many aspiring founders, both developers
and non-developers.

What follows is some unsolicited advice for first-time entrepreneurs.

“Bad” Ideas != Failures

Smart people often make the jump from “I wouldn’t use it” to “that won’t be economically viable”.

Meanwhile, that terrible restaurant down the street has been in business for 20 years and
people continue to pay $10 (or $200 dollars) a month for WordPress, which is open source.

There are customers other than you that will make the businesses you hate thrive.

“Good” Ideas != Profitable Ideas

It’s a mistake to assume that because an idea is sensible it will make money.

1. You’re not “everyone other than you”. You don’t know how they think. What is sensible to you might not be sensible to them.
2. People rarely adopt/switch technologies for economically rational reasons alone.

You Probably Need Less Tech Than You Think

Most ideas can be validated and even sold to customers before much product
development is ever done. Try to build less.

Getting Traction Is Harder Than You Think

Even if your idea is great, your product is great, and your customers love it;
expect the path to traction to be long, tedious, frustrating, and painful.

I love Dropbox. It’s awesome. My friends were giving rave reviews. It made sense
to me intellectually. I installed it at least 3 times and didn’t really start
using it for at least 12 months after I first installed it. I still don’t
pay for it.

Your product is probably less awesome than Dropbox.

Getting traction will require steely resolve, lots of hard work, and more time
than you want to give (or a ton of luck).

---

Most of these really boil down to this: You aren’t “everyone else”. It’s arrogant to
presume that you can predict what millions of other people
are going to like, think, and do.

One lie that I’m consantly told is “Early stage startups only need developers”.

It typically comes in the context of “I don’t know how to program, but I want
to work for a startup. Unfortunately, early stage startups only need coders.”

This is (a) false, and (b) an excuse.

What Do Startups Need?

Startups are a quest to make valuable things from ideas and sweat. Early stage
startups need people who create value.

While writing code is an easy way to feel like you’re creating value, it’s far from
the only way to create value. If you can’t code:

• Get customers / letters of intent
• Make cold calls
• Get lunch and coffee and dry cleaning for the developers so they can get the MVP out the door faster
• Optimize adwords campaigns
• Build buzz via social media
• Get bloggers to write about you / do PR
• Run focus groups
• Find free office space for a team by calling other startups in the area and asking if any of them will host you.

How To Get A Job At Any Startup

You can get a job at ANY early stage startup by emailing the founder and telling him you’ll create value for his/her team. Here’s a template:

Hi, my name is [name].

I want to work for your startup, for free. I’m not a developer. I am smart,
hardworking, motivated, and a quick learner.

I know you may not have a budget for a non-technical team member now, so I’d
like to intern at your company for no pay for 3 weeks. During that time, I
will do anything and everything to create value for your team.

This costs you nothing. If at any time, you can’t justify keeping me around,
we will both agree to amicably part ways and I will be grateful for the
experience. No expectations.

Could I show up on Tuesday of next week to get started? We can consider it an interview. If it seems like a fit, I’ll just get
started with seeing how I can add value to the team. If I don’t hear an explicit “no” from you,
I’ll see you on Tuesday!

Then you have 3 weeks to make yourself absolutely indispensible to that company.

Why This Will Work?

Here’s a secret about startup founders: They like money. So if you can join
their team for free and make them richer, they’ll be excited about paying you
to stick around. (Another hint: if they don’t want free value creation, they
probably suck or it’s not a fit)

If you CAN’T make yourself unquestionably indispensible in some way, then
you need to ask yourself whether you’re really adding enough value for the
startup to consider hiring you.

The next time someone tells you that code is the most important thing, consider this:

It is sometimes possible to outsource code. It is never possible to
outsource strategy or business development.

Non-technical people frequently approach me to build them an MVP. ”Why should
I work with you?”, I ask. ”Because I’m a great hustler”, they respond.

A lot of these people claim to have access to capital, connections,
tons of domain expertise, or whatever. Basically, there are a lot of great
hustlers in the making, but they don’t have a track record and can’t prove
that they are good hustlers.

I’d like to invert the typical “incubator” process. Applicants pay money,
buy into a company, get an MVP, and hustle.

First time hustlers, this is where I tell you to put your money where your
mouth is. Stop saying “if only i had product, i’d be an awesome hustler”

For $5,000, you get half of the company (an MVP), and operate as the CEO for
3 months. We will agree upon guidelines for success/failure metrics and if
you are the hustler you claim to be, you’ll win big.

If you fail, you’ll be forced to either recognize that you’re not the hustler
you thought you were or explain how/why you failed.

New hustlers wildly underestimate the amount of effort it takes to get
traction, how much unglamorous grunt work there is, and don’t have a good
idea of how to methodically test approaches to traction. Many new hustlers
are also often attracted to the glamour of startups but are pretty lazy when
it’s not just sitting around and talking ideas.

The hustler’s job is to take a mutually agreed upon idea (probably an MVP of my
choosing), and get traction.

The hustler pays $5k, gets a functioning MVP (that he/she agrees to work on),
options for 50% of the company, then hustles:

• write blog posts
• seo that sucker
• create landing pages
• cold call
• pass out flyers
• do logistics for our new food delivery service
• do dishes
• find an office space
• learn about a/b testing
• raise money

or any number of hustlery things.

The “fund” provides guidance on these things, points hustlers in the right direction,
but the responsibility is mostly on the hustler to move the needle.

The MVP is what you get, but obviously I’d like to protect my investment as well, so
if it’s clear that further development is helpful or required, we’ll do that as well.
The hustler, of course, will have to justify the development cost (You are the CEO
after all).

If you hit the agreed upon success metrics, then lets talk about building your idea.

The pot of gold at the end of the rainbow is that if you’re really half the hustler
you claim to be:

1. you’ll end up with a substantial equity stake in a business I think is profitable
2. you’ll have a track record
3. I’ll probably be pretty excited about building YOUR product too

Show me that you add value. It’s not hard, but it’s a lot of work.

Regarding a 5k buy in – A lot of people will be turned off by this. That’s ok. Most
of those “hustlers” will talk about ideas for the next year and get nothing done.

Any hustler that can’t raise $5000 isn’t much of a hustler at all.

I’ll lose some great hustlers with that filter, but I’m willing to give that up for the
better signal to noise ratio.

Before You Post Your Ad

Don’t be afraid to spend 25 bucks on a job post

Say what you will about Craigslist, it’s still where everyone goes to find jobs.
It costs $25 to post in the jobs section. You can post in the gigs section for free. If you’re not ready to invest $25 to get better candidates,
you’re not ready to invest in a salesperson.

Don’t make it too easy

Don’t give them all the information they need. Make them hunt through your website for
your contact info. Or linked in. Make your information available, but not obvious. This will weed out people who are just mass mailing
resumes.

Set up phone interviews

Set up a phone interview for some time in the future. Give them 2 options, and ask them to pick one that works for them. Have them call you
at the stated time. You are looking for: (a) a confirmation email “yes, i’ll call you at time X”, and (b) that they actually call on time.

These are important because they weed out (a) people who can’t communicate clearly about future events/expectations and (b) candidates who
are unreliable or can’t manage their calendar [managing future calls is important for followups to prospects]

The interview starts when they contact you first

Typos in emails, an inability to communicate effectively via email, unreliable, missed phone calls, are all reasons to end the process.
There are lots of fantastic salespeople out there, don’t settle for those with obvious deficits.

The Phone Interview

If you aren’t convinced by the interview, prospects won’t be convinced by their sales pitch

The phone interview is a sales call where the salesperson is selling himself to you.
Is he smooth on the phone, does he guide the conversation, does he ask good questions, is
he personable, do you like him?

• Do they ramble? If they do, let them go. See how long it goes. Sales isn’t about the salesperson, it’s about the prospect. Talking forever about themselves is a negative indicator.
• If the conversation stops, do NOT fill the space for them – see what they do with the awkward silence.
• Do they overuse filler words like “um” and “like”? You want someone that sounds smooth, assertive, and confident. If they don’t sound professional and convincing in an interview, they won’t when they sell your product.

Great candidates will ask you questions about the role, what you need, and tell you if it’s a good fit.

Awesomeness doesn’t happen in a vacuum. You don’t want a candidate who is just awesome alone –
he/she has to be awesome with your product.

Do they tell stories about themselves like they’re trying too hard to sell themselves? Great sales
is about finding FIT, not talking about how AWESOME your product is. They should be asking about
your NEEDS and seeing how they can fill them. They should be asking a lot of questions, not regaling
you with stories of how much revenue they closed in an unrelated industry.

There’s no perfect salesperson and there’s no perfect product.

The salesperson’s job is to find out the prospect’s needs/pain points, then show how your product
can solve that. If he’s not looking for fit between him and you, he won’t be looking for fit between
your product and your customers.

What sorts of things can’t you sell / What was the hardest thing you’ve ever had to sell?

Salespeople love to say they can sell anything. That’s never true. Some products
are too expensive, or don’t have product-market fit, or just plain suck. Ask what they struggled to sell or
what sorts of products they think they wouldn’t be able to sell. If the answer is “nothing”, it indicates
either delusion or an inability to reason.

Metrics

Ask them what sorts of metrics they track and how they track them. Calls, decision makers, demos, etc.
How do they typically gauge success?

Also, what system do they use for tracking it (assuming you don’t already
have this stuff in place). As a founder, your job isn’t to train people to sell. You want people who
already think about this the right way. You need metrics, and good salespeople will have answers for you.

Do they set a time to follow up?

This is important.

Wind down the interview in typical fashion. Thank them for their time. Don’t offer any future plans. Refrain from
saying “ok, if i’m interested, i’ll get back to you next week” or something like that. Just thank them.

A salesperson’s
job is to make sales. But many products require multiple touchpoints with busy people. Your salesperson needs to
be effective at getting to the next step.

Salespeople can waste a
lot of time chasing dead leads that aren’t interested. To be efficient, they need to be able to identify which leads
are dead and prune them. At the end of any sales call, they should be either:

• (a) realizing that the lead is no good, or
• (b) scheduling a specific follow up time

If they don’t either (a) try to eliminate themselves from the running, or (b) nail you down to a specific future time,
won’t do that with prospects.

Great questions:

• “What are you looking for in order to make a decision”
• “What could make you decide to either hire me or eliminate me as a candidate right now?”
• “In what timeframe do you intend to make a decision”
• “OK, so if I don’t hear from you by Friday, would it be ok for me to reach out to you on Monday to follow up?”

The result of this is that they’ll be forced to follow up in the future. See if they do.
If they can’t get their calendar together for a job, they’ll fail with prospects.

Other thoughts

Typing speed

This is usually a pretty good indicator of computer usage and that’s important to me:

• I do screensharing demos for Rocket Lease. A prospect gets bored watching you slowly type in a bunch of fields on a form.
• I communicate via email. If written text is important for you to communicate, you probably type reasonably fast as well.

Hotmail accounts

This is a bad filter, and I don’t recommend it for everyone, but if you’re writing me from a hotmail
account, wtf? Who uses that anymore?

I sell a tech product, and most people who live on the internet don’t use hotmail anymore.

Ask your interviewee what you should be looking for

You want candidates who are realistic, and can make decisions on behalf of your company. Ask the candidate
what he/she thinks you should be looking for in a hire. Basically, put him/her in the interviewer’s shoes.

Great candidates are comfortable stepping outside of the interviewee role and giving answers that aren’t
just “pick me, pick me!”

Try to talk them out of the job (thanks TG)

“This job isn’t for everyone… Well, we’re a startup, and so a lot of times that means it’s not very structured… some
people hate that… you may be able to make more money in the short term somewhere else…”

You’re not trying to scare them off, but this will help bring out some of the real issues with the job, their requirements,
and determine if it’s really a fit.

This might seem like you’re scaring away good candidates, but if you start down a path with someone where it’s not a fit,
you’ll be wasting your own time and resources, so you’d rather filter aggressively and be left with more high probability candidates.

Rocket Lease makes it easy for landlords to use an online application to screen prospective tenants. It automatically runs credit reports, background checks, and eviction histories, saving the property manager time and money. I believe so strongly in the product that I left my career as an overpaid options trader and to the point where I’ve invested money from my retirement funds into the venture.

I started working on Rocket Lease full time in Dec 2011 and publicly launched in May 2012. It has been my fulltime pursuit since. I am confident (perhaps delusional) that this is the best startup idea I’ve worked on to date and that my effort will ultimately be rewarded.

However, the journey hasn’t been without trials, so this is a short note chronicling the successes and failures to date.

Things That Worked

Talked to a lot of customers before I building

I worked with property managers and leasing agents to draw wireframes, describe the process and workflow. I did this at bars, coffee shops, and while working out of their offices alongside them. This was important for me to create the right product for my customers. One caveat is that I wasn’t aggressive enough about defining the truly MINIMUM viable product, so as a result of talking to many people, I ended up building in more features than necessary and spending more time than I had to on my MVP.

Spend money on the startup

It was easy for me in the past to try to build everything (there’s no better price than free!) and end up wasting a lot of time. On Rocket Lease, I took the opposite approach, where I almost always try to find some way to NOT build something in house. Same thing goes for services – instead of looking for free services or straining to say on a free plan in a Freemium or SAAS offering, if what I need costs money, just pay it. Some services that have made my life better: codebasehq.com, linode, mailchimp, sanebox, hipchat, idonethis, twilio, skype.

Spend money on me

In previous startups, I had taken the approach that I’m cash strapped, so I had to act cash strapped. Don’t go out, don’t see friends, work all the time. You probably already know this story ends in me burning out. While I still work on Rocket Lease 7 days a week and think about it 25 hours a day, I’m unapologetic about carving out personal time and enjoying sushi. I suspect, though can’t prove, that it’s made me more productive overall.

Outsource

My to do list is a mile long, and a lot of those tasks are outsourceable. I have found a fantastic blog writer (thanks Jake) who helps me manage content creation, I’ve outsourced SEO consulting and execution, VA tasks like compiling lists of property managers, sending emails to investors, making sales phone calls, picking up dry cleaning, creating twitter accounts (yep, tons of em) and sending boxes of chocolate to customers. My biggest single outsourcing expense is programming now. I initially wrote a lot of the code, the first version was built entirely by me, but I’ve since reduced my code contribution to < 20%. I still help architect, discuss and evaluate code, but I now focus almost entirely on how to get traction. Henrique isn’t just a hired gun, he’s a CTO for hire, a trusted and valued asset, and Rocket Lease wouldn’t be possible without him.

I’ve spent a lot on outsourcing, but my quality of life is definitely better as a result.

Interns

I experimented with interns this summer. What I will say about interns is that I’m glad I did it, and there were a lot of great things that came of it, but it was largely inefficient for me. My interns were all really awesome, highly intelligent, but it’s impossible to expect anyone to have the same level of fanaticism about your startup that you do. Also, this isn’t a knock on interns – it’s always hard to find good PEOPLE, whether interns, full-time, paid or unpaid. In that regard I was lucky that Mike found the highest quality people to work with. It was a good experience for me, I spent a lot of time, got more frustrated than I should have, but also learned a lot about managing people (and about my desire to manage people going forward).

Things that didn’t work

SEO

I suck at this. I spent an inordinate amount of effort early on trying to rank for keywords that I thought were valuable. Unfortuantely, the reality is that people don’t search for online apartments application solutions. They search for apartments. And these folks are tenants, who are NOT Rocket Lease’s primary customer.

Anyone who is already renting apartments (my target customers) already has A solution in place that works. The only time they start looking is when their existing solution breaks. And even then they might not turn to Google.

I don’t want to knock this entirely, because we’ve done very well with words that we wanted to rank for, and over 70% of our revenue last year came from customers who found us via Google. But they are relatively low volume keywords, so I think we’re tapped out on the total amount of traffic we can drive through the existing keywords we targeted. The next tier of keywords up that we’d LOVE to rank for seem way out of reach though, so I’m hesitant to expend more effort chasing HUGE keywords.

Adwords

I suck at this too. I wrote about my experience with Adwords here. Fundamentally, I think the real problem here is that not enough apartment owners and landlords are LOOKING for rental application solutions. Arguably I should just keep this going though since I think it can be cost effective, but optimizing the campaign was a lot of work, and at the time it was unclear whether or not it would pay off. Unfortunately also, since the search volume for high converting keywords is low, it seems like it would take a long time to optimize the effort to pay off. Also, the high converting keywords seem in striking range for organic search rankings.

Reddit

I ran ads on reddit. I wasn’t trying to recruit users though, I was trying to recruit evangelists.

I was very explicit and the ad copy was basically “I need your help, Reddit”. I briefly explained what Rocket Lease was, how I think it’s insane that 88% of rental applications are still processed on paper, and made it easy for redditors to send an email to their landlord.

I didn’t have much to offer to incentivize Redditors to help me in the cause (I could have done some sort of reward, but Redditors are smarter than that and see through stuff like that). Redditors have a high concentration of renters who would rather fill out electronic forms than paper. Part of me was hoping that would be enough of an appeal to help me kill the paper application. It wasn’t. But the campaign was cheap and fast so no regrets.

Contest

I created “win a free month of rent” contest (capped at $1500 bucks, sorry Manhattan). My target customer is the landlord, but my goal was to reach a huge number of renters and convince them to sell to their landlords. Fundamentally, it was just a lottery: “give me your email for an entry in the drawing”.

The bit of cleverness (if I do say so) was that the contest encouraged growth by making it incredibly easy for people to share on FB or twitter, then gave more entries into the contest if (a) they shared, (b) anyone signed up through their fb or twitter updates. I split tested: just twitter, just facebook, multiple fb options (messaging friends, status update, liking page), and a bunch of other combinations to see how I could maximize collection of emails.

Additionally, there was a customer.io lifecycle email campaign tied to the backend that would notify users whenever someone signed up through one of their links (to encourage them to come back) and mailchimp drip campaign tied to the backend of this that would let them know more about Rocket Lease and also give them tips on how else to share their links to get more conest entries.

There was one slightly nefarious component to this campaign: one of the options for twitter/facebook sharing was to authorize our twitter/facebook app so that we could tweet once a day on the users behalf. Each tweet/update was tagged with the outgoing message and the time of day. By A/B testing messages and publication time, we were able to get these links to drive approximately half of our signups.

Total signups 21,065. Total development time: 1 week. Total cost: $1500.

So I paid around $0.07 per email. To run the campaign for 6 weeks from mid October to the end of November. I guess I’m not sure if this was a failure. I think with my stated goal of email collection, I did quite good, however, the value of this list is questionable at best, and I can buy targeted lists of opt-in emails from confirmed real estate industry people for around a $.10/email, so its a failure compared to that.

My measured clickthroughs/signups/revenue as a result of this campaign weren’t that great. However, November was also our best month ever in terms of both traffic and revenue. As I write this, I feel like I should give this another look.

Guerilla flyers and stickers

I printed out flyers, then personally went out and put them on doors. The “riskiest” thing I did was print out stickers for Rocket Lease that were designed to fit in the corner of the Redeye (a chicago pop culture newsletter that people read on trains and buses on the way to work), then go around to the newspaper boxes and put one on each newspaper. It was intended to look as if it was actually sponsored by Redeye, but I was waking up at 4am to go from paper bin to paper bin putting stickers on newspaper covers.

Approximately 1% “clickthrough rate” on the URL from the stickers on the Redeyes.

Lots of waking up at 4am, $70 spent on stickers, and basically no results. Huge waste of time.

Purchased emails

I bought an email list from one vendor. Sent about 100k emails, and got a huge unsubscribe rate. In retrospect, the emails probably weren’t legitimately acquired, and I got more annoyed hate mail than I’d like to read. I still think email marketing is valuable, I will probably do it in 2013, but this time I’ll be smarter about it and get higher quality lists.

A few things I thought about but didn’t execute

Chalk Art

Draw rocketlease logos and information on sidewalks around Chicago, maybe or maybe not near leasing agencies…

Homeless people

Can I pay the guys that stand next to the freeway 20 bucks to hold a poster for me? What do these guys generally collect in a day? I wonder if they’d be game to hold up a big poster for me instead for the same amount of money?

What’s Next?

I have a few for Rocket Lease in 2013. My initial hope had been that using online tactics would work. I personally have a preference for the sorts of things that CODE can do to spread Rocket Lease because code is the sort of thing that can scale very quickly. I haven’t had the best results with that so now I’m currently experimenting with a more traditional sales approach.

Cold Calling

Cold calling leads. I’m not sure how this will turn out, but the lead list is bigger than I can handle. People in real estate are professional networkers and they all post their phone numbers in public forums and websites.

Door to Door

In person demos at leasing agencies and property management companies. I think this will be fruitful, because Rocket Lease is genuinely useful, easier to use, and better than most of the entrenched competition. The struggle will be identifying qualified candidates and scoring the appointments. Maybe someone really charismatic can just show up and charm their way into demos?

College evangelists

College kids just get Rocket Lease. They’re also tech savvy and great for doing demos. Another thought here is that if they’re connected to the school newspaper somehow, Rocket Lease might be able to score some .edu backlinks.

Email marketing, round 2

Email lists are cheaper to buy than I had originally thought. Even if response rates are low, the numbers are so big that this could work out.

Auxiliary real estate services

Leasing agencies want a bunch of things, and online rental applications isn’t particularly high on the list. I wonder if offering things like lead generation, posting things on craigslist on their behalf, helping them fix their websites – at cost, or at only a small premium, would help me get in the door to have a trusted relationship where they might start using Rocket Lease.

People in real estate often overpay for development/web design, so there might be a way to add value here. This seems really time consuming, so it’ll probably be a last resort.

Wish me luck

Hopefully 2013 will be the year I finally make it in startupland. Rocket Lease has been a phenomenal learning experience so far, and I think it has the makings of a great business. But until it gets real traction, I’m just another struggling entrepreneur.