Scot Wingo is co-founder and CEO of ChannelAdvisor, a public company in Morrisville, N.C. that helps e-commerce retailers optimize sales across channels. Wingo and team took the company public in May 2013.
This is a regular series, and you can read the kick-off post here for background. Ask him a question by commenting or emailing firstname.lastname@example.org.Part Two: Evaluating your idea.
This is part two of a three-part series that seeks to answer the question "How do you come up with an idea for a software startup company
?" asked by Aaron, a local web developer
In Part One
, I introduced the "startup idea" pattern. The key highlights from that pattern were:
* Business-oriented ideas are more numerous and easier to execute on than consumer ideas.
* The best 'seeds' for ideas for business-oriented solutions are add-ons, industry-oriented and solving a problem you have personally experienced (necessity is the mother of invention).
In Part Two, we'll look at the next step: once you think you have an idea or several candidates, how do you evaluate those ideas?
While Aaron didn't specifically ask about it, the next step, and in many ways the same step, is evaluating your idea. In my experience, one of the challenges is too many ideas. Hopefully the first post has stimulated some ideas and now you have a pool of ideas to work with. The problem is you can end up with an 'abundance of choice'. Now you have too many ideas and it can be hard to evaluate them, so you are probably asking, should I focus on A, B, or C?
Now most of what you read on this topic is written by venture capitalists (VCs) and they take the opposite approach I'm going to recommend here. It makes sense because they are evaluating ideas for investment and have different criteria than you probably will. I bring this up, because if one of your goals is to raise VC, you will need to 'think like a VC' and make sure your idea meets their criteria as well as your own. Even if you don't have plans to raise VC, it doesn't hurt to run your ideas through some of their filters just to help think through the priority.What do prospective customers think?
If you read a lot of biographies of the biggest innovators (Jobs
, Gates, etc.) you will read this quote by Henry Ford
: "If I asked people what they wanted, they would have said a faster horse."
When you say it like that, sure. But what people really wanted was a better mode of transportation. If you had asked anyone if they wanted a less expensive, faster way to move from point A to point B, you would have gotten interesting feedback.
My point is, when you read some of the top entrepreneurial books, it feels like you should put your head down for a year or more, develop your idea and unveil it to the world as the 'next big thing'.
My experience is the exact opposite. The sooner and faster you can talk to a prospect and get input, the better. If you are developing something for doctors, get a doctor or two to look at what you are building. There are several reasons for this:
*The world is changing at a dramatically rapid pace -
you can waste 12-18 months of time aiming for the wrong target. This is time you won't get back and in the world of startups, things tend to happen in an amplified fashion (like dog years). So not only will you lose a year, but it will be as if you lost five to seven.
*Business buyers are increasingly complex
and it can be very hard to predict their needs until you talk to them. Get them using your product and really solve their problems.
*Most businesses don't have the luxury of venture capital,
so you should have a 'plan A' that is 'customer capital'. If you want to start a business, you should plan on having customers fund it rather than VCs. If VCs become part of the picture, it will be because of your success with customers 99% of the time, not your great slides or idea.
*I have found that the Henry Ford quote is wrong for a software business and that listening to your customers is right.
To summarize, I like to say that I have a pretty short startup playbook three pages really:
1. Sign up some customers and make them happy.
2. Listen to their needs and ideas.
3. Rinse and repeat.
With that in mind, take your ideas and pitch them to potential customers. Listen to their feedback, record it and give it the most serious of weight as you evaluate your ideas. What's really exciting about this approach is it can be like pulling a string. You meet with a prospective customer, pitch some ideas, the team likes one, suggests another and then you refine what you are pitching, get more feedback, do a prototype, get feedback, iterate and suddenly you have a product, a customer and are off to the races.What do you want to build?
Another dimension you should evaluate your ideas against is where your passion lies. Expect that starting a software company is a big multi-year commitment. If you want to be successful, you will also be working 80-100 hour weeks for the foreseeable future. There will be big highs and big lows. During the big lows, you need to draw on a spring of energy and in my experience the source of that spring is best summarized by the word, passion.
Passion is one of those things that is hard for me to describe, but you'll know it when you see (or feel) it.
When you've talked to prospective customers, do you get into this zone where you are bouncing ideas back and forth like a high-speed ping pong match?
Do you stay up till 2 a.m. reading everything you can about a certain topic? Being passionate about startups isn't enough, I'm talking about a certain subject matter area here - HR software, medical software, e-commerce, finance, software development, mobile apps, gaming, etc.
Take the ideas you are evaluating and next to the column for 'prospect input', put a passion meter. If something doesn't score very high on this list, you should knock it off. If that eliminates everything from your list, start over and come at it from the passion angle first this time.
Said another way passion is an important filter for your ideas. If you aren't passionate about the idea, you don't want to be 18 months into it with 30 customers and bored with what you are doing or dreading the next two years. In my experience, there is a 100% correlation between successful businesses and very passionate founders. Unfortunately, there's not a 100% correlation between passionate founders and success, but that's something we'll cover in part three.What is the TAM?
The final evaluation step, or exercise, that I think every startup founder (regardless of whether you're developing software or not) is to evaluate your Total Addressable Market - TAM in VC-slang.
TAM is figured when you take all the companies that are potential targets and consider what your software, service or product would look like if they all used it. Are there multiple ways to think about this? What are they? Do you come out with the same answer or different?
Evaluating your TAM is a "three birds with one stone" exercise:
1. This will cause you to look deep into yourself and answer the question, "How big do I want this to be?"
2. If you ever do want/need to raise VC, this is going to be one of the first things you'll need to tackle, so it can be helpful to have thought about it, if not done the research.
3. Finally, when you spend a fair amount of time running your ideas through the TAM exercise, a nice work-product or artifact, as we say in software, is the start of a roadmap of how to attack this market. I will show an example in part three.
The first one is really important and is as much a reflective part of the evaluation process as it is an evaluation criteria. Before starting this TAM exercise, you should think deeply about (with your co-founders if that's the situation) about how big you want this enterprise to possibly be.
I feel it's important to jump in here and mention that I've built three different businesses each with different TAMs:
*Our first business had a small TAM ($20M)
and is what most VCs would call a lifestyle business. We were able to sell the business, but that can be quite rare for a TAM in this region.
*Our second business was impossible to calculate a TAM -
it was consumer-oriented and based on e-commerce which was only 18-months-old at the time. So how do you calculate that? It ended up being a great acquisition type business, and that's what we did.
*The third business, ChannelAdvisor, was our 'moon shot'.
When we started in 2001, we did the TAM exercise and felt that if things went our way (ecommerce adoption grew, etc.), we could build a business with $100M to $300M in revenue. It would potentially have a shot at an IPO and a $1B valuation over a long period of execution.
Which do you want? A smaller firm with lower risk and complexity, but also a probable smaller outcome? Or do you want to moonshot? Or, somewhere in between?
I know at this point, you may be frustrated and looking for an example and sources of data. Let me walk you through an example so you can see how this works.
Our first business, Stingray Software
, was an add-on (see part one) business. Calculating the TAM for this type of business is relatively easy.
Our solution was a Visual C++ add-on and at the time, Microsoft actually reported its number of users at 30,000. We priced our solution at $500/user. So the TAM roughly was 30,000 x 500 = $15,000,000.
When we started, that number seemed great and it was. Two years in, we were at an $8 million run rate. But we started to worry - were we running out of addressable market? We did the math and realized we were actually getting $1,200 per customer vs. the original $500 - we had added several products and over 3,000 customers. We were doing much better on an average revenue per unit (ARPU) and when we looked at the math again, 30,000 x $1,200 = $36,000,000.
This exercise also showed us we could turn the dial
on ARPU when we wanted to and dramatically increase our TAM.
Generally, a bottom-up approach to TAM will take the total number of potential customers, multiply it by a reasonable ARPU. A top-down approach is where you take the size of an industry, multiply it by an adoption factor (we're going to get X% of the industry) and then perhaps a revenue factor (and we'll make Z% of that part in revenue).Sources of information for the TAM exercise
I've done this TAM exercise hundreds of times over the years and have found several great sources of information to help you:
*Company research -
If you are adding on to or are in the same industry as a public company, one of the best sources of research is the reports published by stock analysts. Simply go to the company's investor relations site and you should find a list of analysts at firms like Goldman Sachs, Morgan Stanley and JP Morgan who report regularly on the company. These reports are chock full of great information.
*Industry research -
IDC, Forrester, Gartner and many other research firms publish paid-for and free research around the size, growth rate and trends inside a variety of industries. My favorite trick - many times one of your competitors may have paid big bucks for one of these firms to do research in your industry. Sign up for the white paper and BOOM, you get it for free. Thanks competitor!
Many VCs publish great information. For example, Mary Meeker
at Kleiner Perkins publishes a regular gigantic internet report. BVP publishes regular papers about the SaaS industry, key metrics and valuation details. Finally, be sure to check Slideshare. I've found there is generally information available on pretty much any industry I research.The headcount@TAM pattern.
Once you understand the TAM of your idea (a general range is ok), one final math exercise you should go through is this algebra formula to calculate your employee count if you were to achieve your TAM. It's a three-step process that I call the headcount@TAM pattern:
1. Annual revenue x operating margin = annual profit
2. Annual revenue - annual profit = operating costs
3. Headcount@TAM = operating costs / $100,000
The $100,000 in my experience is a good rule of thumb for the Triangle area. If you were to do this for the Bay area, you would need to make that number $250,000.
Let's work this through an example. One of your ideas has a TAM of $20M at the midpoint (maybe it's $10-30M). This style of software business frequently has 20% or better margins and at-scale grows between 5-10% a year.
Here is the calculation:
*TAM of $20M/yr in revenue (lifestyle business)
*$20M x 20% operating margin = $4M in profit
*$20M-$4M = $16M in operating costs
*$16M/$100,000 = 160 headcount at TAM
Are you prepared to run a business with 160 employees? This is definitely a swag and you can do it with less, or maybe it will take more. But as a rule of thumb this is what you are looking at for a B2C software business. This will give you a gut check and help you think through what the next three to five years of your life are going to look like.
Let's do the same math for a high-growth, lower margin $100M idea:
*TAM of $100M/year in revenue
*$100M x 10% operating margin = $10M in profit
*$100M -$10M = $90M in operating costs
*$90M/$100,000 = 900 headcount at TAM
Wow, 900 people. That seems crazy when you are zero, but that's generally what it will take to achieve that. Can you envision yourself as a part of a 900-person company?
If you liked this exercise, a fun next step would be to take the headcount @TAM and imagine what the 160 or 900 people will be doing. How many sales people do you think you'll need? How big of an administrative (legal, HR, finance) staff? How many software people? How about support? Any other groups you think you will need?
Now build a spreadsheet to chart the growth over three to five years from 0 to your TAM. How does that feel? Before long, you'll look up from this exercise and surprise, you have built a business plan!Conclusion
In part one, we talked about how to come up with some ideas and discussed some of the lower-risk areas to pursue (B2B vs. B2C for example). In this post, we looked at three different ways to evaluate your ideas to help rank them:
*The customer evaluation -
Does the idea get future customers
Does this idea get YOU fired up?
*TAM test -
Does the idea match with your desired business size?
It's important to note that these three evaluation criteria will probably all swirl together for you. Do you stay up late and have 10 spreadsheets calculating the TAM? Do you have passionate discussions with your co-founder(s) about the TAM? Do you wish you could have spent more time with that prospect to ask them that one more thing? Do you consider emailing the prospect once an hour with new thoughts?
If the answer to most of the above is yes, then go for it.
If the answer is no to all of the above, or you are freaked out by the TAM exercise or the prospect meetings go poorly, then it is probably time to look in the mirror and do a gut check on the whole startup thing.
It could be that you are passionate about software and the startup, but just don't like the direction you're going. That's fine and a signal that maybe you should get a founder on board to help you think through these things. Without help, you will be drifting around and not making headway. Maybe the CEO role isn't for you and you can be a sales or partnership leader, or a CTO/VP of Engineering.
By the end of this exercise you have hopefully narrowed down a list of five to 10 possible startup ideas to one or two.
The last thing to think about - and what can be used as the final criteria to evaluate an idea - is what I call execution. We'll cover that in part three.