1. There are little delays in Index and Currency pricing. We are working to resolve it.
    Dismiss Notice

Please take these suggestions in the positive spirit intended

Discussion in 'Pre-Sale Questions' started by legspeed, Sep 12, 2017.

Please take these suggestions in the positive spirit intended

  1. legspeed

    legspeed New Member

    Joined:
    Aug 27, 2017
    Messages:
    4
    Likes Received:
    1
    As I write this, I've tested your list of approximately 9,000 stocks using the Last() function call. 20 minutes in, MarketXLS has consumed Excel almost entirely and returned only a few hundred prices. It's 10:15 pm U.S. Central time and I'm on a Gigibit fiber link with a fast PC (lucky me!). The bottleneck seems to be elsewhere.

    I really want (and hope) MarketXLS succeeds! Quandl is not for everyone, Yahoo is changing (and wasn't always good) and Thomson Reuters is scared to create a product like this because of its price point. You could win this market!

    Here are my suggestions.

    Your software needs to be much faster: You need to deliver basic info like Last() about 10x faster than you are. Maybe focus your efforts on that. You need to accept that MarketXLS is really, really slow before this will get better. [Try T-One on Excel sometime, that's a good standard to shoot for.] All the charting and other stuff is fancy, but probably much less in demand than good data delivery. I'm not a real-time quote guy; I'm not sure you are really equipped to play in that market. Wouldn't it be better to focus on what you're already pretty good at?

    You need to keep your symbol book updated: It's not acceptable to miss the change from DOW --> DWDP, for example, or miss "Alphabet" from the screener because it's not Google anymore. Those are both current examples. MarketXLS, or your data providers, need to sweep for those changes daily -- weeks and months can't go by. Looks bad, is bad.

    Tell us how the software calls work: Many of us could and would optimize our spreadsheets if we knew how your calls work. Your "My Symbols" documentation is appallingly sparse. Let us know the how pre-fetching and calls work in detail.

    Don't make us guess what data you have (!!!!!): I (and I bet others) have spent hours testing what data you have for what symbol. For instance, the Beta() function returns a value for only a small fraction of the S&P 500. Against a list MarketXLS provided, the still-running Last() function is returning "N/A" for ~ 15% of the list. Publish a comprehensive guide to what you have and don't have. You'll get a lot of understanding from your subscribers (and even prospects) if we don't have to waste a lot of time guessing. Trust that we'll trust that it'll get better.

    Don't make us guess if you are having problems with data providers or systems: Let us know the score. If you are having a bad-hair day, it will save us pulling out our hair wondering why things won't update.

    Commit to the S&P 500, go from there: You might consider making a corporate commitment to being perfect on all S&P 500 stocks first. Maybe next all Blackrock, Vanguard, State Street and Invesco (Powershares) ETFs. Then maybe take on the Russell 1000, the S&P 600, etc. Perfect would mean all the data, all the symbols correct and a FAST response for all data queries.

    [Metastock, presumably to go after Mut Funds? I dunno. We tried and failed -- reliability was poor. Talk to Morningstar or Lipper.]

    Thanks for all your hard work.
     
    marketxls likes this.
  2. marketxls

    marketxls Administrator
    Staff Member

    Joined:
    Dec 21, 2015
    Messages:
    1,005
    Likes Received:
    10
    Thank you very much for your feedback.

    We will work on these and keep you posted.

    Kind regards
    Ankur
     
  3. marketxls

    marketxls Administrator
    Staff Member

    Joined:
    Dec 21, 2015
    Messages:
    1,005
    Likes Received:
    10
    Hi Legspeed,

    Please see comments inline...


    I really want (and hope) MarketXLS succeeds! Quandl is not for everyone, Yahoo is changing (and wasn't always good) and Thomson Reuters is scared to create a product like this because of its price point. You could win this market!

    Thank You

    Here are my suggestions.

    Your software needs to be much faster: You need to deliver basic info like Last() about 10x faster than you are. Maybe focus your efforts on that. You need to accept that MarketXLS is really, really slow before this will get better. [Try T-One on Excel sometime, that's a good standard to shoot for.] All the charting and other stuff is fancy, but probably much less in demand than good data delivery. I'm not a real-time quote guy; I'm not sure you are really equipped to play in that market. Wouldn't it be better to focus on what you're already pretty good at?

    OK. Makes sense. I understand, we will work on it.

    You need to keep your symbol book updated: It's not acceptable to miss the change from DOW --> DWDP, for example, or miss "Alphabet" from the screener because it's not Google anymore. Those are both current examples. MarketXLS, or your data providers, need to sweep for those changes daily -- weeks and months can't go by. Looks bad, is bad.

    OK. Sure. I understand.

    Tell us how the software calls work: Many of us could and would optimize our spreadsheets if we knew how your calls work. Your "My Symbols" documentation is appallingly sparse. Let us know the how pre-fetching and calls work in detail.

    Yes, we are trying to build persistent caching. So, that will ignore the need of pre-fetching. So, I think that will solve it. If we continue with this pre-fetching. I will put a lot more details out.

    Don't make us guess what data you have (!!!!!): I (and I bet others) have spent hours testing what data you have for what symbol. For instance, the Beta() function returns a value for only a small fraction of the S&P 500. Against a list MarketXLS provided, the still-running Last() function is returning "N/A" for ~ 15% of the list. Publish a comprehensive guide to what you have and don't have. You'll get a lot of understanding from your subscribers (and even prospects) if we don't have to waste a lot of time guessing. Trust that we'll trust that it'll get better.

    Ok. Sure. Will do. Beta and some other key stats had an issue with a bug that came on the server. We worked and fixed it. I did post an announcement but looks like we need to manage this better. I just tested Beta for all of SNP and we seem to have all results. I noticed its a bit slow, so I have asked to throw in some more compute power on the server.

    Don't make us guess if you are having problems with data providers or systems: Let us know the score. If you are having a bad-hair day, it will save us pulling out our hair wondering why things won't update.

    Yes, Sure. We will put a "Service Status" icon in the App in coming release. I understand.

    Commit to the S&P 500, go from there: You might consider making a corporate commitment to being perfect on all S&P 500 stocks first. Maybe next all Blackrock, Vanguard, State Street and Invesco (Powershares) ETFs. Then maybe take on the Russell 1000, the S&P 600, etc. Perfect would mean all the data, all the symbols correct and a FAST response for all data queries.

    Yes, Sure. I understand.

    [Metastock, presumably to go after Mut Funds? I dunno. We tried and failed -- reliability was poor. Talk to Morningstar or Lipper.]

    Thanks for all your hard work.[/QUOTE]

    Thanks a lot for the consutructive feedback. Means a lot.
     
    #3 marketxls, Sep 12, 2017
    Last edited: Sep 12, 2017

Share This Page