Tell me more ×
Answers OnStartups is a question and answer site for entrepreneurs looking to start or run a new business. It's 100% free, no registration required.

I have an activity, which is very simple in conceptual sense:

  • Exploration of possible demand and extraction of interesting products.

    • Product is considered "interesting", if it seems to be (possibly) "buyable". Another condition is to check, if in some countries there are very nice prices (alibaba.com, for example, may be the source of data).
  • Walking to possible distributors and figuring out if they are interested in production.

  • If some of them are interested, concluding a contract about supply in few weeks or so.

Now I am planning to achieve automatization of the prediction process.

So, I would like to hear from experienced people if demand prediction is effective. I have heard about data mining, machine learning and many other ideas, but I am unsure, if it will work in my practice. Of course, extra huge decrease of "area of interest" (few thousands of products to dozens of really interesting ones) is acceptable too.

Thanks in advance!

share|improve this question
2  
Ask on JOS forum or look for stackoverflow.com/users/18858/joannes-vermorel He is in this business and have a lot of experience in demand prediction. – Ross Feb 20 '12 at 19:02

1 Answer

There's an anecdote worth remembering:

If you bear in mind that the past is not an indicator of the future, there can only be two types of prediction/forecast: Wrong forecasts, or Lucky forecasts.

Automisation of a process is one thing, "accurate prediction" is an entirely different holy grail altogether.

Narrowing your focus will definitely put you in a better position to make progress, but you really should talk to an AI/Machine learning person to understand the complexities involved with this type of work.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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