jasonswett.net

This might surprise some people who have been following my progress with Snip, but after about 1.5 years of self-employment, I’ve decided to get a job. I’ll explain why.

I guess I’ll explain first why I was anti-job for so long. I plan to retire at a relatively young age, and the way I intend to do that is to grow a business to the point where I can either sell it for a lot of money or delegate daily operations to someone else. I’ve taken several stabs at building that business, and they’ve all flopped except my most recent try, Snip. I can tell that Snip has the potential to earn considerable revenue (right now it’s earning just a trickle), and I plan to stick with Snip for the long haul, since I see no reason why I can’t grow revenue from what it is now to something significantly greater.

For a long time, I considered working on Snip to be fairly incompatible with working at a regular job. Growing Snip required going out and meeting with stylists and salon owners so they could help me make sure I was on the right track. Meeting with salon owners was also necessary in order to make any sales. I actually did have a regular job for the first six months of Snip (January to June 2011), but I didn’t get much done except coding, and you can’t grow a business by just coding. When I got the ax in June and went freelance, I had a lot more availability to meet with my “domain experts” (stylists and salon owners) and to go on sales calls.

Up until now I’ve had just one way of selling: to drive around to salons, walk in, and try to sell Snip to the owner. It was time-consuming, but it worked, and it was the only thing I could think of. I knew that I would eventually have to come up with a faster way to sell, but I figured I’d rather stay with this slow-but-sure method than risk switching to some faster method that didn’t even work. I did this for a long time, like a year, and it probably wouldn’t have been possible if I had had a regular job at the same time.

So that’s the main reason I’ve been anti-job for so long: getting a job would have interfered with Snip so much that it would have been equivalent to giving up on Snip, which would have been equivalent to giving up on early retirement, which would have been equivalent to sitting at a desk every day until age 65, which is about as appealing to me as crawling into a coffin.

What’s different now? I’m at the point now where I think it no longer makes sense to stick with the slow canvassing method of selling. I have a decent number of customers (three) and I’m confident the product is solid and I don’t see any reason to wait longer before switching to a faster method of selling. Instead of sticking with the “high-touch” sales I’ve been doing, I think I need to figure out something “low-touch”, where salon owners just come to the website and sign up for an account on their own. Conveniently, I think the work involved on my part for that can be done 100% on the computer, on my own schedule. So I don’t have so much of a need anymore to have a very flexible schedule.

I also had a misconception when I first started freelancing that being self-employed would afford me more time to work on Snip than if I had a job. I naively thought I could work 20 hours per week on client work and 20 hours per week on Snip. This conception turned out to be comically flawed. It’s not really possible, at least in my experience, to control that finely the amount of client work you do in a week. And it takes a lot more than 20 hours per week of work, at least with my current level of consulting skill and earning power, to support a family. I’m finding that it’s actually just as hard and just as time-consuming, again, at least for me, to earn a decent living as a freelancer as it is to have a regular job. If it’s just six one way and half a dozen the other, what’s the point?

So I’m making a deliberate effort now to find a job. I have a few requirements, in no particular order:

1. It has to be a job with a products company, not a services company
2. It has to be a job working with Ruby or another language of that caliber, like Python or Lisp (not PHP)
3. It can’t be a job working with Microsoft technologies
4. It has to be a job where I can work from Grand Rapids, meaning it’s either remote or in GR (somewhere else within a few hours’ drive, like Chicago or Ann Arbor,  is not entirely out of the question, but GR is strongly preferred)
5. It has to be a certain salary

I’m going through the process with Snip of figuring out how to go from high-touch sales to low-touch sales. As you might have known or guessed, high-touch means a lot of personal interaction with each prospect and low-touch is where prospects just sign up online without any work required on the vendor’s part.

As of now I have pretty much no idea how to go from high-touch to low-touch, but I’m learning. I at least know how I could ensure that I would not get low-touch sales, and that would be to have no traffic coming into sniphq.com. So one important piece of the puzzle is to get some traffic slowing to the Snip website.

Again, I’m somewhat clueless here, but at least I’ve narrowed it down a bit. I read somewhere – I think on Patrick McKenzie’s blog, or maybe on SEOmoz – that it’s a good idea to develop some content that would be useful to the kind of people who might buy your product (or at least a set of people that includes the kind of people who might buy your product) and put that in a blog. It’s kind of a “duh” thing, I guess.

I’m imagining, ideally, I would write some blog post that’s so useful that tons of people would start linking to it all over the internet, and tons of people would therefore start coming to the Snip site. A percentage of those people might wonder what else the geniuses at Snip have come up with and take a look around the rest of the site. A percentage of those people might try the demo, and then a percentage of those might sign up for an account…and a percentage of those would become paying customers. I believe they call this a “funnel”. So maybe for every, say, 1,000 visitors my blog post attracts, one of them ultimately ends up becoming a paying customer.

The hard part of this is coming up with the valuable content. The content should be valuable to a large number of people, but not just any kind of people. If you imagine a Venn diagram of “people who would benefit from using Snip” and “people who would find this certain blog post interesting”, the intersection of those two circles has to be sufficiently large, whether that means writing something that’s really valuable to salon owners specifically, or whether it’s something valuable to, say, all businessowners, which includes salon owners.

In trying to think of interesting content, I can imagine a spectrum where at one end I have “things that are most certainly interesting to current Snip users” and at the other end “things that are interesting to people in the hair/beauty industry”. The former is easy but not very helpful; the latter is hard but probably really helpful. I think I’ll just dive in and start with things that are certainly interesting to current Snip users. Perhaps I’ll find ways to gradually move down to the other end of the spectrum as I go.

So, with that in mind, I wrote a blog post that’s probably about as far to the “left” end of that spectrum as can be: instructions on what to do if you forget your Snip password. That question is the #1 support issue I deal with, so I have no doubt that it’s a topic in which users are interested, at least on the occasion that they forget their passwords. You can view the blog post here. My next post ideas move a little further down the spectrum: how to see which services make you the most money, how to see how much money you made in a week (or any interval), and how to track your client retention rate.

I have a new goal for Snip: get one salon to sign up and become a paying customer without having any personal interaction with me. I’d like to do that by, say, February 1st, 2013.

(I had set a goal not long ago to get 10 paying customers by the end of 2012. I’m not going to reach that goal and I’m fine with that.)

So far I’ve been selling Snip by personally visiting salons. That’s been going pretty well, but it’s necessarily slow. If I’m going to start earning enough money with Snip to support my family before I’m a wrinkled old man, I have to start moving the needle faster, and the max speed for selling online is a lot faster than the max speed for selling in person.

I have a lot of ideas for how to get the people of the internet to buy Snip, but I don’t know for sure yet which ones are good ideas and which ones aren’t. Suggestions are welcome and appreciated.

If you want to have a successful web product business, I think there are certain steps you have to follow in a certain order. I presently conceive these steps to be:

1. Think of a product that a lot of people would want to buy
2. Build the simplest version of that product that people could actually use
3. Get some people to buy the product
4. Continually improve the product and the system for selling the product until you’re earning enough money to support yourself

You can’t start on #2 until you complete #1, you can’t start on #3 until you complete #2, and you can’t start on #4 until you’ve successfully completed the other three. Inconveniently, it seems that you sometimes have to get to step 3 where you try to sell the product before you know whether you were right about #1. I had at least four “startups” where I went through steps 1 and 2, and then realized when I got to step 3 that I had built something nobody really wanted.

Luckily, with Snip, I’ve been able to get all the way through step 3 successfully, and now I’m at step 4.

Step 4 is a fundamentally different kind of problem than step 1. Step 1 is binary; either I picked something people want, or I picked something nobody wants. Step 4 is more of an optimization problem. I’m already over the hump of conceiving and creating a product that people are sure to buy. Now I just have to figure out how to get more people to buy the product, faster. I’m currently making X sales per month. I have to figure out how to get it to 1.1X, 1.5X, 2X, 5X, 100X. I don’t have to make a big leap, I just have to move the needle. It’s a much nicer problem to have.

So far my method of selling has been to visit salons personally. I’ve had a surprisingly good experience with this, but it’s still pretty slow. I’ve visited over 80 salons and talked with over 100 stylists and salon owners but I only have 3 customers (not including one more prospective customer who’s in a trial period). If you know stuff about sales/marketing, you’ll recognize that that 3-4% conversion rate is actually not bad, but visiting salons in person is time-consuming. Each sale I’ve made so far has required at least three separate visits, and my 4 “customers” (including the trial user) are in four separate cities (Comstock Park, Kentwood, Newaygo and Fremont, Michigan). I’ve calculated that I need about 85 customers to live off of Snip, and if I continue at the rate I have been, it’s going to take like 20 years just to be able to make a living, which of course unacceptable.

I set a goal in early June to get 10 Snip customers by the end of 2012. At that time, all I had was one salon using Snip for free, and no paying customers. I had two salons in a 30-day trial but I didn’t know if they would become paying customers or not. Happily, both did. So I have two customers down and eight to go.

The first two salons were smaller salons (two stylists each) and Snip seemed to meet their needs just fine, but the third salon was bigger (about eight stylists) and their needs were a little more complex. Having this bigger salon on Snip highlighted some somewhat serious deficiencies (e.g. it was too slow for a fast-paced salon) and I decided it would be wise to put sales on hold for a time and focus on addressing these newly-surfaced issues.

I keep going back and forth with Snip between “sales mode” and “code mode”. I always have one of two problems: either the product has some important things missing or broken, or the product is fine but I don’t have enough customers. I spent January 2011 to October 2011 in code mode, then October 2011 to June 2012 in sales mode. I went back into code mode in June 2012 and I’m still there. I intend to go back into sales mode at the beginning of October 2012.

I would just go back into sales mode now, since I certainly need more customers, but I’m not sure that that would be wise. When I got my biggest customer, I found myself buried with work, and I’m not talking nice-to-have new features, I’m talking about just getting to the point where the customer wasn’t frustrated with the product. The next time I get inundated with bug reports and feature requests, as I’m sure I will, I want to be working from as solid and clean a codebase as possible. To use an analogy, if I’m going to cook dinner and then do all the dishes afterward, I’d rather clean the kitchen and do all the dirty dishes first, rather than cook dinner in a messy kitchen and then have a huge mess to clean up afterward. So I probably won’t add any new features in September, just fix some defects and clean up my code.

So that’s my plan: spend September in code mode, then switch in October to sales mode, for the foreseeable future.

In late December 2011, I set a goal to get 100 Snip customers by the end of 2012. This number isn’t looking very realistic anymore, so I think I’m going to adjust my goal to something I actually think I can achieve.

At the time I set my original goal in December 2011, I didn’t have any customers at all yet. I had been planning to give away 10 free accounts and then start charging after that. The giving away accounts thing didn’t go very well. I think it may have partly been because people were skeptical of such a good deal. Plus they may not have valued the product because it was free. Anyway, I was only able to give away one free account. A few months after I gave away that one free account, I decided to just start selling the product for real.

So far I think six salons have said yes to a 30-day trial, but then only two of those six ended up actually signing up for the trial. The two salons that did sign up did so pretty recently, and they’re still in their trial period as of this writing. So I have a total of three users right now: one free user and two salons in 30-day trials.

I don’t know for sure whether the salons who are in trial will actually become paying customers or not, but I feel pretty confident that I’ll continue to be able to get people to sign up. A few months ago I figured that I could visit 10 salons a day, one day a week, 4.3 weeks a month (that’s about how many weeks are in a month) for about 43 visits per month. Assuming a 5% conversion rate (which is just a total guess), that would be two new customers per month. That number is obviously based on some imaginary stuff, but it’s a lot more realistic than what I was figuring would get me to 100 customers in a year.

I think 10 customers by the end of 2012 is a more realistic goal. It feels a little bit lame to set such a wimpy goal, but based on what I’ve been able to do so far, 100 customers just seems like pure fantasy, and an unrealistic goal is about as motivating as no goal at all. And 10 customers might not sound like a lot to somebody else, but I would actually be thrilled to have 10 customers by the end of December. That’s $150-600 per month of almost free money.

At some point I’ll obviously have to speed things up to a rate better than two new customers per month. To be honest, I don’t really know how I’ll do that, but I’m okay with not knowing yet. Two customers per month is kind of painfully slow, but it’s not so slow that I can’t stand to wait until December to have 10 customers. And by the time I’ve gotten to that point, I’m sure I will have learned plenty more about business and I imagine I’ll have some idea about how to get more customers faster.

Today marks a big milestone for Snip: the first 30-day trial sign-up. I had had a few appointments in recent weeks with salon owners to get their 30-day trials started, but each one rescheduled or didn’t go through with it for one reason or another. Today was the first time someone actually entered her credit card info and hit submit on the sign-up page. So in 30 days, if this salon doesn’t cancel, they will be Snip’s first paying customer. Pretty stoked about it.

Let me briefly review the last several months. December 2011 is when I first started pounding the pavement, walking into salons, trying to find my first free user. In January 2012, the first free user started using Snip. Then in March, after a few frustrating months, I decided to stop trying to give Snip away and start trying to actually sell it for money. In April, I got five different salons to say “yes” to a 30-day trial, but when it came down to it, four out of the five backed out for whatever reason.* Then, today, May 18th, I finally had someone not cancel on me, which was of course a great feeling. I scheduled another meeting with her in a couple weeks to make sure things are going smoothly.

I’m looking forward to going out again next week and doing some more canvassing. Selling is actually easier and more fun than programming. I had to talk to about 100 people before I got my first 30-day trial sign-up, so it will be interesting to see how many it takes to get #2.

*Two said they were too swamped, one said she had to talk to her husband, and the other one said she still needed a computer monitor. (She has her monitor now and we have a new meeting scheduled.)