“Don’t develop any software until you have a customer” – Interview of serial entrepreneur Anand Soman

As a part of the new, experimental PuneTech video series, we interviewed Anand Soman, CEO of Pune-based startup Infinishare, and serial entrepreneur. Infinishare provides software for a host of internet enabled devices, including digital photoframe devices, digital displays, home internet devices. They provide a full software stack for such devices, but their core IP is in device-to-cloud, and device-to-device communications.

Before Infinishare, Anand has had two successful exits as an entrepreneur – one for a bootstrapped startup, and one for a VC funded startup. See the PuneTech report of last year’s POCC meeting “How (and Why) to bootstrap your own startup,” for some of Anand’s thoughts on this topics.

In today’s video we asked him about his company, and more importantly about what advice he has for young entrepreneurs. (We are still experimenting with our video creation process, so the sound is still bad. We will had a fix for this soon, but in the meantime, please max the volume when you view this video):

Some interesting excerpts from this interview:

  • Don’t develop any software until you have a customer (so you’re sure there’s a market)
  • Focus on paying customers from the beginning
  • It is very difficult to get the freemium model to work
  • If you have users for your product who are not paying, don’t call them customers! Get real paying customers

See the video for these and other insights.

About Anand Soman

Anand is co-founder of Infinishare technologies.

Before Infinishare, Anand founded Intigma Inc, where they wrote AI-based engines for automated classification and extraction of content. Intigma was acquired by Emptoris Inc., after which Anand headed their India centre. Before Intigma, Anand founded Testchip Technologies, developing tools and cell libraries for Testchip design. Testchip was acquired by HPL Technologies Inc.

Anand received his B.Tech from IIT Bombay, and was the recipient of the Institute Medal for Electrical Engg. He received his M.S.E.E. and Ph.D degrees from California Institute of Technology, Pasadena, USA, specializing in DSP & Communications. He worked at AT&T, Murray Hill, before founding Testchip. He has published numerous research papers in International Journals & Conferences and has several years of project execution & management experience.

Reblog this post [with Zemanta]

15 thoughts on ““Don’t develop any software until you have a customer” – Interview of serial entrepreneur Anand Soman

  1. I think there are two sides to it. If you are going to offer a product that is in the well defined market serving a well defined need, you can focus on paying customers right from the beginning.

    At the end of the day, people play to their strengths. Some people are better at freemium and some at ads and some other at charging upfront. I doubt if there is a silver bullet or even a best-model kind of thing.

    1. @Manas, in response to your point, I should have added: as a techie, I personally am (actually, “was until recently”) strongly of the opinion that actually building it is a great way of defining the product, and once you have it, then it is easier to go around finding a customer/market for it.

      But based on my recent experiences, I am updating my view. Having a customer involved in defining the product and in defining the market, pretty much like what the agile guys recommend, is a great way of doing this. If there are major technology unknowns, or if some of the concepts are just too difficult to explain without a demo (as is the case in something like your own TrailOfView.com (which is looking really nice, by the way)) I think it makes sense to quickly prototype it, but just up to a point where the customer understands what you are talking about, and with customer input along the way.

  2. Agree with Manas.

    For a moment, let’s keep aside the “free” sites of the world like Google, Facebook and let’s say we want to build online office suite. How do I get the customers for that without anything up and running?

    At the very least, I need to build a proof of concept to show it to the potential customers. It’s hard to imagine customer will sign up to buy the software just based on, say, power point slides.

    PS: Sorry, I haven’t seen the video yet. So, I might be taking Anand’s words out of context.

  3. @Manas, I think Anand’s point is in fact targeted towards what you said: if your product is not in a well defined market serving a well defined need, then you should be focusing on defining it, rather than on developing anything at this stage.

    As for paid vs freemium vs ad-supported – don’t you think, especially in the Indian context, it is much easier to point to success stories in the paid model than in the other models?

    I think what Anand meant is that you need to have a potential customer who signs off on your idea at the powerpoint stage. Obviously, the customer is not expected to be paying at this stage, but he should say: “If you build this, I will pay. And I’ll help you figure out what to build.” If you’re finding it too difficult to get somebody to say that, then maybe you’re working on the wrong product/service.

    Fortunately, in our current startup, we’ve managed to do that, and I believe that it has made a huge impact on our feature set. Instead of the endless debates that happen between co-founders about whether to do X or Y, it was much easier to check with 2 or 3 of the potential customers and the decision was much easier. Also, the fact that these potential customers were thinking from the perspective of “I need to pay for this”, and “I need to make money from this software,” and that fact that the customer was visualizing using the software himself, was very crucial in ensuring that the decisions and suggestions were well thought out, and serious/sincere, and economically sound. (e.g. “Don’t do XX; that will piss off some of my customers”, or “Do YY, that will make it easier for me to justify the price to my customers.”) Over time, as the product got closer and closer to the real thing, we could see how the suggestions of the customer got more practical and specific.

    Having a potential customer involved from the beginning, and having it clear from the beginning that he was ultimately going to pay for this, was a very important factor for us, and I believe that is pretty much the advice Anand is giving.

  4. @navin There are two things here: 1) Users and 2) Customers

    And they are not one and the same thing. The track some companies take is – get-the-users-and-we’ll-find-customers. Some other companies say – users-are-customers. While some other companies say – customers-will-find-users

    All these are different models and there can be many more.

    Yes, it’s best to have a paying customer upfront but it may not apply to all the models. It’s a different debate whether it’s worth pursuing any other model in India or not but the baseline is that the models are different.

    Basically, the act of creating the value may be different from the act of monetizing it.

    But I do not intend to say that one should try to create value in a vacuum. That’s very hard to work unless someone gets very lucky.

    I fully agree beyond any doubt that the person for whom value is being created must be part of the value creation process. Otherwise, we are just playing darts in the dark.

    All that I wanted to say was that it was a big leap to say that focus on paying customers from day one. That pretty much excludes several other models from consideration. I do not disagree with what Anand says, I just wanted it be better qualified 🙂

    And by the way, I would be pinging you separately for Trailofview in a couple of days 🙂 The context is pretty much the same as this thread (don’t worry, I am not expecting you to pay anything) 🙂 🙂

  5. i think this is a negative thought. i would say develop as many softwares as you can in this recession period, because you are going to get developers at minimum cost, also a company can invest in R&D which can not be done in peak time(because we have to complete project), many times we compromise with some buggy code or non-optimized code when delivery date is closer.
    All these mistakes can be recovered during such recession time.
    Also if we have softwares ready for some domain in this recession, we can always get good price in boom, also we can offer the solution to client in less time than other rival companies..

    but all above calculation will depend upon the investing capacity of company & aggression in the management.

  6. This advice holds good for people who have industry experience & contacts. Since they have access to the market & right people, they can get their first few customers before building the product. They can get some audience based on their credibility before getting started.

    You asked a right question about how young founders without industry contacts get started. I still don’t understand how can they attract people with industry experience to join them when they are un-proven.

    As I understand, Product business is highly risky but the rewards are also high. If you have an idea & you believe that it works & if you have done a basic SWOT, then you should go ahead & build the product. It’s fine if you have access to a customer to listen to your idea but that should not be a mandatory step to get started.

    1. @Manas, Good point about the fact that the users might not be the same as customers. We’re in agreement that you definitely want to involve the users in the product creation process. And also we agree that there are business models where act of creating value is different from the act of monetizing it.

      My point is that somehow on the web, everybody automatically gravitates towards this solution – and I think it is much, much more difficult to build a successful business if your paying customer is different from your user. You first have to go ahead and build the user base, then you have to convert user base to paying customer base, and that might not happen in sufficient numbers. Or, you discover too late that you focused on the wrong kinds of users. Or that you built in the wrong features.

      In general, if you are focusing on paying customers from the beginning, it becomes much easier to judge whether you’re focusing on the right things or not – because you can simply follow the money trail.

      @nobody, A couple of points:

      • The customers don’t have to pay anything upfront. All that needs to be agreed is that if you deliver what you promised, they customer will ultimately pay; and also that the customer is willing to keep giving you feedback on the intermediate versions. That is much easier. What I have seen is that if you show the drive, passion and enthusiasm, and if you are solving a real pain point of the customer, they will actually push you into building the solution.
      • “If you have an idea and you believe it works” – I’ve seen far too many techies believe strongly in ideas that people don’t really want. Often, people will say, “Yes, that would be nice” to an idea, but back off when you mention money. One reliable way to figure out the worth of an idea is by how much people are willing to pay for it. Of course, you can also try to estimate the worth of an idea without this, but that is significantly more difficult. And young techies are precisely the wrong people to do this (this is based on a number of examples I’ve seen in the POCC).

        If you are not getting access to a customer to listen to your idea, then you are working on the wrong idea.

      • As for young founders attracting people with experience – What I have seen is that if you have the enthusiasm, drive, passion, presentation skills, and a decent idea, it is not difficult to get experienced people to show interest. The mistakes I’ve seen young founders making is this:
        • Being too secretive: You can’t get people interested in your idea if you don’t tell them the idea!!
        • Bad presentation skills: If you approach an experienced person, you get about 2 minutes before he decides whether he’s interested in you or not. Your passion and idea need to come through during that time. Work on improving your pitch. Google for resources on this aspect.
        • Not approaching people: Multiple times in the past on the POCC various people have put out feelers saying that mentors are available and willing to help out young founders. What I have seen is that nobody comes forward.
        • Not accepting that an idea has no future: This one is a tricky one. On the one hand, far too many ideas that young founders work on are almost pointless – typically solving a problem that very few people are interested in. And when the experienced person tells them this, they are not willing to accept the fact. But, of course, even if a young founder did come up with a brilliant idea, it is quite possible that some (or all(?)) advisers might tell them it is a bad idea; and sticking with their idea in the face of all this rejection is the primary trait of a founder. So I really don’t know the correct solution to this issue. But maybe it again comes down to the money – figure out how you’re going to make money from your idea, and then figure out how you can get early validation of this

      I think that is precisely what is being said.

  7. Navin,

    In reality, most of the time you don’t know whether your idea is a great idea or a stupid idea before executing. It’s hard for anyone to predict that. Even successful VC’s success ratio will be less than 25%. If it’s so easy, then there won’t be so many failures in this world. There is no short cut to avoid failures & it’s not worth trying to minimize that.

    Problem explaining someone about an idea is – different people perceive an idea differently based on their exposure. And there is difference between a wish & a need for a customer. Even with a real product, it’s hard for customers to say “yes” in 10 minutes. That’s why there are so many shelf-wares i.e. customers buy a product thinking that it’s useful but then realize that it’s not something they need.

    Take ideas like Twitter, Search engine idea of mid 90’s, Web mail idea of 90’s, Operating system in late 70’s, Social networking in early 2000.

    It’s hard for most people to accept these ideas. That’s why Yahoo under estimated google & never thought search engine to be so powerful.

    Just imagine your reaction if someone told you about micro-blogging ( posting messages with 140 chars ) like Twitter. I am sure no VC will be willing to bet their money in that idea before seeing the idea. I am sure even their founders wouldn’t have expected this type of usage / success when they started with that idea.

    Again, real time search itself evolved from the Twitter idea which is again not something which was thought when it was started.

    So waiting for someone to accept your idea & say “Yes” is not a good idea. The same idea would have become a stupid idea if that didn’t take off. But the founders would have learnt a lesson or two from that.

    Another example is iPod. No customers in this earth would have accepted that idea when it was explained in words. The first major review criticized it as LAME. But you know how customers felt when it was actually launched to them.

    Most of the time, it’s the execution that matters more than the idea. Most failures happen because of poor execution. Even a great idea with a poor execution will fail & it deserve that.


  8. @nobody, What you describe is indeed one way to succeed – and there have been some spectacular successes that way. However, too many young founders assume that this is the only way forward. And due to that, I’ve seen too many young founders work on ideas that are frankly pointless.

    I am just attempting to swing the pendulum a little bit in the other direction.

    Also note: that by the method you are proposing (which I think can be termed the “standard” method), you would need outside funding (angel or VC) to develop your product and then market it. And that brings its own host of issues. It’s a pain to get funding. And it is an even bigger pain after you get funded.

    Also, see this interesting post: Bootstrapping and the infinite runway.

  9. Navin,

    Thanks for the link. It was really a great article for those who bootstrap.

    I agree with you on the pain in raising money. VC’s also follow either credentials ( the one who built a successful startup ) or successful products ( in case of new founders ).

    I am not proposing that model. What I propose is to first focus on the first version of the product & then worry about getting the first customer or building the team.

    What I propose is the model of Balsamiq where the idea came from his own real need.


  10. I think this is bullshit. Programmers program because they love to do so. It would be great if i can make some money out of my personnel projects but making projects for money makes me a Pseudo coder. Programming is a passion not a profession. That is what we should teach. This is Science, This is Technology not Economics and Marketing. We write code for making things easier for other. And yes, every software has at least one user, “The Programmer Himself”.

    I don’t mean no disrespect. I understand that the author is much senior to me and deserves respect and I do respect him. I just wanted to show my opposition to the ideology he is trying to impart to future computer scientists.

    1. @Owais,
      If you read the post carefully, you will notice that the advice is for “young entrepreneurs” and not “programmers” or “future computer scientists”. Hence, your well-articulated rant is rather off-base.

      Yes, programming should be a passion. Yes, every programmer should write some, if not all the programs he writes for the love of programming. That might make you a great programmer. But, small startups cannot afford to do that. They need to make money on a regular basis or else go out of business. That’s where all the advice of this article comes in. How to make a startup that is not going to go out of business. That problem has very little to do with “future computer scientists.”

  11. We should look at anyone’s advice from the perspective of his/her background. First look at the company that anand founded and then it may justify the advice that he is suggesting.
    Also in case of facebook , tweeter etc customer where always there and will be there, they are advertiser but in order to acquire them you needed users. Any site that can collect interest will get advertiser as customers. So here there was no need of making any pitch or anything as it was universal truth that if you have users you have advertisers and nobody can predict what the masses wants and VC’s will also not been able to predict.

Leave a Reply

Your email address will not be published. Required fields are marked *