81

I was asked to sell the source code (along with existing users) of small utility app I created years ago. I've investigated how to put a price on the source code but so far haven't come up with a good solution.

I've searched the net, but haven't found anything useful. Then I came across a few others who also sold their source code with users, but their prices seem unrealistically high. For example, one person calculated price per user at about $200. He had 80 users and ended up selling the source with users for $30k. How did he come up with this price?

Can I find a good price with this formula:

(number of users x app price) + (app price x num of new users in one year)

?

If this is a good formula, how do you price source that doesn't yet have users?

7
  • 5
    Warning. Some integration and heckloads of support will be needed, overcharging might be the right price. A percentage on sales might al sodo. How "core" would be your code to their application sale potential? This metric might help set a reasonable price.
    – ZJR
    Commented Aug 1, 2013 at 11:33
  • No future percentage of sale for me. The potential is a matter of improving the app. The current potential is number of sold units in 1 year, I guess.
    – deviDave
    Commented Aug 1, 2013 at 11:51
  • 8
    You're not taking into consideration one thing. That programmer probably knew the budget of the company that paid 30k, and he probably also knew how many users they'd need. Do the math and you come up with a price per user which is tailored for that company so that you know they'll pay you for it. I can almost guarantee he didn't pull those numbers from the sky like that.
    – Neil
    Commented Aug 1, 2013 at 12:52
  • 1
    @Neil Not really since he sold the sources to the startup company who needed his app for their startup. You may say that he caught them in hurry, but they were not some rich company. And I am also selling my app to a startup company. Like I would not know to set the price if the large company asked me to sell them the app :).
    – deviDave
    Commented Aug 1, 2013 at 14:42
  • 1
    @MathewFoscarini - Answering your comment about CW is better done in The Whiteboard. Short answer though is no, not yet.
    – user53019
    Commented Aug 1, 2013 at 20:13

5 Answers 5

79

Selling the source code for an app is very much like selling a business.

The standard formula is price = revenue * 3 + assets.

The multiplication of 3 is a factor of supply and demand. The more buyers a business has the higher the multiplier. When we hear about a business being purchased by ABC Corp in the news, it's often for a large figure. Those businesses can have a multiplier of 5 or higher.

For businesses that don't have a revenue history, then they depend upon an evaluation. The evaluation is an estimate of projected revenue and the multiplier is applied to that.

So we can calculate the multiplier for your example;

1.875 = 30000 / 16000 = 30000 / (200 * 80)

Assuming he sold all his licenses in 1 year, he (your example) would have a multiplier of 1.875 with no additional assets. That's not a very good deal for the programmer. Especially when you factor in future upgrades from those users adding to revenue.

Why is it not a good deal?

The buyer can recover his costs in less than 2 years. Most people take longer to pay off a car loan.

When we speak to the buyer in terms of setting a price, we discuss how long the buyer would like to recover his investment and start profiting from his purchase.

You are saying I'm giving up this source code, and its future revenue to you. The price is set based upon an estimate of what the future would be.

If you have not received any revenue from your source code, then you will have to argue with the buyer what the evaluation of its future revenue will be.

7
  • 4
    @deviDave the multiplier of 3 is based upon the average IPO price that businesses set for the first stock price. Investors often look for a good multiplier, but some IPOs are over priced. (i.e. Facebook had a multiplier of 6, but some investors argued their evaluation was also to high).
    – Reactgular
    Commented Aug 1, 2013 at 12:04
  • 1
    Thanks man! The multiplier thing is really helping me. I know you're not a business expert, but can you pass me a good link on things you mentioned here (multiplier, IPO, etc.)?
    – deviDave
    Commented Aug 1, 2013 at 14:46
  • 1
    @deviDave CNN Money has a business value calculator. It's multiple is defined by the industry type. cgi.money.cnn.com/tools/bizworth
    – Reactgular
    Commented Aug 1, 2013 at 19:52
  • 2
    @deviDave here is a free PDF on selling a business. Discusses the multiplier in more depth. evancarmichael.com/Buying-A-Business/890/…
    – Reactgular
    Commented Aug 1, 2013 at 19:54
  • 2
    @deviDave here are a few alternative formulas (including multiplier) by Forbes magazine. forbes.com/2009/09/23/…
    – Reactgular
    Commented Aug 1, 2013 at 19:55
63

The hardest part of doing this sort of thing for the first time is really psychological - there is a very strong tendency to think of what it cost you in man hours, which is usually wildly inaccurate when done retrospectively and ignores the "I wasn't sitting at a desk but I was thinking about that algorithm all day..." and other overhead details, etc.

So I'd like to invite you to change your frame of perspective using an analogy: you no longer have an app, you have a steel widget. You put things into it, and things come out the other side, and what it does to the things that go in have caused various people to become accustomed to having your widget around. To date, you've just been giving your widget away for free because someone gave you the steel for free so it doesn't cost you anything.

The Background Concepts

Now someone wants to buy the whole concept and user base of your widget from you.

First of all, why do they want to buy it? If it's a business, the answer is "to make money." Either they are improving an existing product they have and wish to increase loyalty and offer a benefit they think could sell more copies, or they want to avoid solving a problem they have and thus reduce costs or focus their own effort on other things. They might also want your users as potential "hot leads" who they know could be interested in their product and could be unusually likely to buy stuff from them.

The relevant equation:

Price Paid = (Buyer's Perceived Value - Seller's Cost) * Negotiation

So if it costs you nothing (you already did the work without expectation of pay), and it's worth $100k to them, do they pay you $1? Or $99k? $50k? It's all about negotiation - trying to determine where the final price is between the maximum they'll pay and the minimum you'll accept.

Sometimes negotiation is so weird people pay too much, and sometimes people sell for less than cost. These are edge cases, and so we ignore them - but yes, they do exist. I want to hire Instagram's negotiators for everything I do ;)

So first, what's it worth to them? This is by far the hardest thing to know, and one tactic is to flat out ask them. I know, crazy right?

Super Secret Negotiation Tactic

"I'm a reasonable man - what's this worth to you?" or "What's your budget for this sort of acquisition?" You'd be amazed at how often people just flat out tell you. They might not want to haggle, and if they just want to do their job and buy something from you and go on with their day, they could just tell you, "We've got about $50k in the budget for acquisitions like yours, and yours is relatively small compared to some of the other things we are buying, so we figured 5-10k would be reasonable given the straight-purchase we are requesting." Or "We figured it would cost us about $4000 in expenses to make this ourselves, so that's the most we'd pay under any circumstances," or simply "We're looking to seal this deal at around $3000."

And then you get to decide if that's ok with you and if you want to push it or take it. How hard was that? In negotiation it's almost always very important that you must not be the first one to name a price - so if they volunteer a price then you have a baseline you can accept outright or argue up from. But they might not name a price, and we have to see if that price is reasonable anyway.

There are a few accounting systems to determine the value of something, and this is what a rational business will use to determine a budget for buying your little 'entity':

  1. Cost

    The value is what it cost, perhaps with yearly depreciation. This is the most common form of accounting in the world, and it literally says "the value is whatever it cost to buy it, decreasing over time." Seriously - it costs what it costs. Not very helpful to us here, but its true.

    This is what people try to do by determining man hours, but I'll give you the bottom-line: this is meaningless in software. You can work 40 years on a million lines of code with a going rate of $50 an hour and the result is worth $0. You aren't freelancing or accepting a contract to build something at an hourly rate, nor did you make it "on spec" with the hope of selling it to recoup your expenses. This is psychologically pleasing, but utterly meaningless in the context of buying and selling.

  2. Replacement Cost

    The value of something is what it would cost to replace it. This can be easy with commodities, like asking "what is the value of a new Ford Focus?" But this is not so easy in software, because it can be like math - a single one line formula could take a century to discover if you don't already know it. Or what took you 10 hours could take someone else 100 - or maybe it would take them only an hour.

    So this would look at trying to estimate what it would cost to have a replacement for your widget built that simultaneously doesn't violate any of your rights as an inventor. Looking at the lines of code/complexity/difficulty of your app would produce a range of anywhere from "maybe a month for a prototype that is low on bugs if one person who knows what they are doing works on it" to... who knows. It must not be trivial or they wouldn't offer you money at all.

    If they have their own development team maybe their estimate of doing it themselves is very reasonable. But they don't want to - they have more important things to do with their time. They'd have to wait for months to get started, or they would have to hire someone - and who knows if they can deliver or if it'll just be a waste of time and money? There's so much risk!

    You've got the goods Right NOW, and this has a special value. Take advantage of this.

  3. Comps (short for "comparisons")

    This is what other things are going for. For example, if this business is used to buying apps/users/source code, they can say, "Well, this widget is easier to make than the SuperWidget we bought last month for $10k, but the output isn't as marketable as our DeluxeWidget we bought last year which we paid only $5k for." So perhaps they figure a comparative value is somewhere between $5k and $10k, and it doesn't matter if you have a million lines of code or 10, they don't have to know or care.

    This is how most non-commodities are sold (like real estate). It's a great system, and it's what you were trying to research, but in this market (software) there is very little public data, so you are at a disadvantage of being in the dark on this. Understand, though, they probably have more data on this than you do, and it is probably part of how they figure what they want to pay you.

  4. Income Multiple (Projected Sales) System

    As Mathew Foscarini pointed out, this is a system used to value business and commercial real estate properties.

    The idea is that you have an asset that generates an income. For instance, an apartment building takes in $50k per year in rent. Then there is a Multiple applied, which is based on the comp system (mentioned above), say 10. So the market value of this apartment building is how much rent can be collected, based on current occupancy and rent rates, over 10 years = $500k. Of course if you raise rents and improve occupancy next year to get an extra 10k a year in rent, suddenly your property is worth an extra $100k - and thus why so many rich people (and bankrupt people, too, of course) are involved in real estate.

    This system can be applied to software, but if your app isn't individually commercially viable it's hard to do this. With your example of 80 users paying $200 each, that means that if a company can convince those same people to buy a new version (which is way easier than selling to strangers), or convince a small portion of their larger client base to buy 80 copies, that's a quick $16k for maybe sending out an email blast and sending a memo to your sales staff.

    Good established companies have estimated lifetime values of users, and if this number is high (like, say, Adobe's buyers of Creative Suite), then paying $30k to pick up even 1 new user or retain an existing customer was a great idea.

What You Should Do

The first step is "talk to them." Learn about their needs, why they are interested, what need does this fill for them, just learn as much about them and what they want as possible. This is Being A Good Salesman (not a sleazebag salesman) - get to know your customer.

Maybe they are actually buying to resell. I've had people offer to buy code of mine because they had a contract where they were supposed to make something that did what my code already was doing. If their total contract was $500, obviously the most they were going to pay me was "less than that." I asked and they pretty much flat out told me just like that. Sometimes I wasn't interested (it wasn't worth the hassle to me for that price, or I was too busy), sometimes I just gave them the code for free, and sometimes I took them up on their offer to make a little extra money on code I already wrote and could still keep using.

Maybe they want to reskin/repurpose the app and sell it as their own product. Maybe they want to add it to a menu of their existing software. Maybe they don't care about the app much but want the users and the app to be a free bonus given to buyers of their next version. Maybe it will be compiled into their own source code and the existing app will be 'discontinued' but the feature will be available in their app now... etc, etc, etc. I could make stuff up all day long, but the only way to have even a vague idea is Just Ask. Even if they lie, who cares, you learned something!

Sometimes these are job interviews of a sort and they'll want to buy your services in the future, maybe they just want a widget to save them the trouble.

Final Caution

The devil is in the details, and they matter. Do you retain ANY rights to the code? Do they even want you to stop using/delete all copies of the code and the app of your own? Do they just want a license to use your stuff and 'transfer' the name and users to them and they could care less what you do after that? Do they want ongoing support, consultation, and if so what is appropriate to contact you about and when?

If they are to make future demands of your time and effort, that is a good time to offer something like "x hours of support in the transition/interpretation, then I'm available at $Y per hour after that." Be interested, be professional, be supportive - don't give yourself, your time, and your work away because you forgot to clarify and put things in writing.

3
  • 11
    +1 and welcome to Programmers, wish every fresh face wrote such quality content. Hope you'll stay a while! Commented Aug 1, 2013 at 19:17
  • 1
    Great post. This should be turned into an article.
    – Reactgular
    Commented Aug 1, 2013 at 20:00
  • Sometimes, you cannot avoid being the first to talk price - maybe they, too, know the tactic of never being the first to name a price point. In that case, you don't want to say a fair price and then be negotiated down from there to something you wouldn't be happy with, instead start at a ridiculously high sum and get negotiated down to a merely outstanding deal. (A rule of thumb I've been using is "Whatever I think is almost unfair, times five.")
    – shieldfoss
    Commented Sep 28, 2013 at 22:06
22

In theory, the price would be determined by the effort which was required to create the application. If a single person worked for two months non-stop (i.e. including weekends) to create the app and the average price per hour of a freelancer of this skill (minus all the taxes) is $200/hour, 200 × 8 × 30 × 2 = 96 000. This means that one can sell this app for $96 000.

In practice, the price is determined by the demand. If nobody wants the source code of your app, you can give it for free, it won't change anything. On the other hand, if a company with lots of money is hugely interested in your app, the price will grow consequently.

If you're face to face with a potential buyer who's strong at negotiation and has leverage on you (for example the buyer is a company which has enough potential to rewrite your app from scratch; you, on the other hand, urgently need money), then the price can drop to the extremes.

The fact that the price is determined by the demand also means that it involves a broad range of parameters, some being completely unrelated to the source code:

  • Is the code high or low quality?

  • Is there a good documentation?

  • Is it written using standards and internationally recognized practices?

  • Is it using popular third-party libraries?

  • Is there a support?

  • Is the code easy to implement in other applications?

  • Is the infrastructure (version control, bug tracking system, deployment in one step, etc.) set correctly and can be easily reused?

  • Is the application tested enough?

  • Is it reviewed by pairs?

  • Is it marketed enough?

  • Is it presented well? A dedicated website done by professional designers presenting many aspects of the source code as a product will inevitably help.

  • etc.

Another important point is to know exactly what's sold. There is a certain gap between selling the right to capitalize on the application by selling licenses to the customers, and selling the right to do whatever the buyer likes to do with the source code. Depending on the precise formulation of the contract, the price can vary a lot, because in one case, the buyer won't be able to do nearly anything, while in other cases, he will be able to create huge profits. Again, this is pure negotiation, similar to the negotiation in any other domain.

9
  • 1
    I tried with calculating work hours involved in the process, but that's not so easy. Firstly, most work hours have been undocumented. Then multiple types of work were involved such as coding, testing, marketing, support. But you reply is very good since it confirmed my own thought and my fears (selling source code = selling potatoes).
    – deviDave
    Commented Aug 1, 2013 at 11:59
  • 1
    @deviDave the above is very subjective. Sadly, the price of these kinds of sales is often not related to how much effort went into creating the source, or how well it's made. You don't really think the people at instagram worked a billion dollars worth of hours?
    – Reactgular
    Commented Aug 1, 2013 at 12:02
  • 3
    lol, I caution those who are upvoting not to think like programmers. What you perceive to be value is not what a business person is looking for when they make an investment. It's nice to believe our efforts are adding value, but the world is full of source code that yields no return in the form of $.
    – Reactgular
    Commented Aug 1, 2013 at 12:20
  • 1
    @MathewFoscarini: that's exactly what I try to explain in the second part of my answer. The quote: "it involves a broad range of parameters, some being completely unrelated to the source code" is also relevant. Commented Aug 1, 2013 at 12:42
  • 1
    @MathewFoscarini: "I'm just jealous you're getting more upvotes than me. lol": this is about to change. Here, an upvote from me (since I find your answer valuable). Commented Aug 1, 2013 at 14:38
7

As other answers have pointed out, the demand for the product is the essential criterion, but that is very difficult to establish. An easier way to work out what the product is worth is to calculate what it would cost to redevelop. If yours is the only example of its type, this is what you would be competing against: the cost to the seller to develop the program itself (or pay for it to be developed). I wouldn't necessarily attempt to sell it at this price, but it might be considered as a lower bound.

There have been some studies on the total cost it would take to develop a codebase from scratch, mostly involving lines of code. See for example: http://en.wikipedia.org/wiki/COCOMO, or here for a concrete example.

1
3

The value of the source code is not necessarily related to the cost to produce it.

The real value is a percentage of the value to the entity which intends to purchase the source code.

If the source code produces $75,000 in value for a business, the cost could be a percentage of $75,000. Maybe $15,000 to $30,000+ depending on your negotiating skills.

Not the answer you're looking for? Browse other questions tagged or ask your own question.