4 ms·

Launched https://www.easydatatransform.com https://www.easydatatransform.com earlier this year. It has sold some licenses and I wouldn't call it a failure yet,

Launched https://www.easydatatransform.com https://www.easydatatransform.com earlier this year. It has sold some licenses and I wouldn't call it a failure yet, but the visit to download ratio is low compared to my other 2 commercial products. Not sure why. I'm pretty sure the traffic is targeted, I am solving a real problem and the website explains what it does. Any feedback would be welcome!

3pt14159 7y ago
Hey I remember you from the early days of HN! Please take this feedback as coming from a good place with good intentions. 1. The website and product would benefit from having a designer involved. There is a design fashion that changes about every five years or so and you're a couple cycles behind. 2. Get someone that needs your software to talk through the copy. "Merge, split, clean, dedupe, reformat and more" into what? From where? Is this online or downloadable? Does this work for my specific workflow? 3. Screenshot looks like it is on a Mac. Is it Mac only? If not, put a Windows screenshot right next to it. I know after you scroll down it shows Windows support, but that can be easily missed. 4. Block quote testimonials: Bold or highlight the most relevant part. I'm not reading a paragraph, I'm scanning. 5. Fix the about page. Many Excel files have sensitive data and people click about to make sure that they can trust the company. You need better, more reassuring copy and a better picture. Asking for feedback is fine, but the current copy borders on "this is really new and may not be reliable" territory. 6. If your target demographic is Excel pros, explain why user defined functions or macros couldn't or shouldn't accomplish the same task that you help with. Hope this helps and best of luck!
- hermitcrab 7y ago
  Thanks for the feedback. 1. It looks pretty modern to me. I'm not selling to web designers, so I'm too bothered with keeping up with all the latest trends, I just need the web design to be 'good enough' to be credible. 2. I think that information is all on the front page, isn't it? I don't want to make the tag line too wordy. 3. The screenshot is Mac. But it says Windows and Mac further down. 4. I've kept the testimonials short, but bolding might be worth a try. 5. Pretty much the same 'about' page seems fine for my other 2 products. Also, no-one much goes to the 'about' page. One of the advantages of my product is that your data stays on your computer. 6. Fair comment, but I don't think they would be on my website if they we happy with just Excel!
  - frequentnapper 7y ago
    the design looks modern and clear to me as well. I don't think that's the issue.
  - oulu2006 7y ago
    Agree with the OP, the website needs a designers touch, it does look dated. The general standard of websites is so much higher these days, if it looks cramped and dated I don't linger. Testimonials are too long as well.
    - hermitcrab 7y ago
      What would you consider some good examples of modern websites for desktop software? BTW You would probably also consider the design of https://www.perfecttableplan.com https://www.perfecttableplan.com dated. But that converts visitors to downloads just fine.
      - JamesBarney 7y ago
        I disagree with the other users I think it looks fine. I think the biggest differences is perfect table plan focuses on a very concrete idea that is easy to communicate and uses a paradigm people are familiar with (looking at a seating chart). I was able figure out what it did within 5 seconds of looking at the website. But data transformations are a fairly abstract concept. Most users I imagine don't think "I need to transform my data", they probably think of the problem more concretely like "I need to deduplicate this giant excel spreadsheet". They will be glancing at the website asking themselves the question "Will this deduplicate my giant excel spreadsheet?" and the first picture they see will be a dataflow diagram. The user has probably never seen a dataflow diagram, and only rarely will they have done in gui based programming before. They will probably think "This might deduplicate my giant excel spreadsheet after I learn a bunch of hard stuff about how to wire these different boxes together". Before ever realizing how easy your app is to use. If I were you, I'd replace data transformations with something like "Excel power tasks", and then have a list of all the tasks you can do, like deduplicate, concatenate, merge columns, etc. all with 15 second videos showing how easy your application makes accomplishing these tasks. So users lands on site, sees a list of the tasks they might want to accomplish, clicks it, watches 15 second video showing them how ridiculously easy your app makes performing that task. Btw the apps look really awesome.
Edmond 7y ago
I am looking at this and thinking, how many times would a typical user need this? I occasionally find myself needing to manipulate excel documents but not enough to want to invest in dedicated software. Is this meant to support ETL workflows? Is it possible to create transform recipe packs as part of the product offering? Basically pre-built configuration instead of users doing this. Maybe provide the software for free and sell the recipes? Is this something you can package as a component that other ETL products could plug into? This can be either native plugins or some kind of web API front-end. Just spitballing :)
- hermitcrab 7y ago
  Thanks for the feedback. >I am looking at this and thinking, how many times would a typical user need this? I am trying to keep the product easy-to-use and the price low to reduce this barrier. But I guess a lot of people would rather spend several hours tweaking their data in Excel, rather than pay $99 and learn a new tool. >Is this meant to support ETL workflows? No. I imagine my target user as a marketing guy, scientist or business analyst who has a bunch of CSV and Excel files they need to munge together and clean-up to get some useful data. Maybe they build a process that they repeat once a month or with several different datasets. But they aren't data science people building an ETL pipeline. In fact they have probably never heard of 'ETL'. Data science people seem fairly wedded to R, Python+Pandas and expensive corporate ETL tools. >Is it possible to create transform recipe packs as part of the product offering? I think everyone is going to have different data sets and different use cases. I'm not sure that is possible to generalize the process beyond the individual transforms. >Is this something you can package as a component that other ETL products could plug into? Probably not.
grok22 7y ago
I think you need to experiment with different price points for the product. To me it seems like I would hesitate buying this for simple jobs due to the price itself. But maybe it it was $25-$40 I might buy it as a impulse buy for speeding my job at work. Or maybe make the free trial dependent on amount of use rather than making it time limited (transform a small number of rows for free for example).
- hermitcrab 7y ago
  It is already a lot cheaper than other similar tools (but less mature also). Any cheaper and it might start not to look credible. And I anticipate that some users might need quite a lot of support, so I can't afford to make it too cheap. If it is only worth $25 to you, then you aren't really my target market. I went for time limited, rather than use or feature limited, because it allows people to test the full capabilities. The 7 days use don't have to be contiguous.
atentaten 7y ago
I'm curios to know what are your other two commercial products since EasyDataTransform looks useful.
- hermitcrab 7y ago
  My other 2 commercial products are: https://www.hyperplan.com https://www.hyperplan.com https://www.perfecttableplan.com https://www.perfecttableplan.com
pragmatic 7y ago
In a recent job, I was a software engineer who worked with the data science people in an "analytics" group. A big PITA for us was the data we'd get from various sources (vendors/clients/whomever). Oh, there's a rando string in a "number" field in row 9,100,500, huh, that sucks. You can't always build the validation into your ETL pipeline for various reasons. Or it fails the whole file (b/c the data science folks built it that way really aren't that experienced at handling exceptions in the file loads, etc) Anyway that rambling is trying to lead up to me saying that there's a chasm between your supposed target market (marketers, BSAs, scientists) and the engineers/devs etc that could probably use some better tools for CSV/JSON munging. Excel is terrible at this job. What I often did was load the csv into sqlite to and then ran queries on the data to figure out how many records were horked. Or if it was a process we were testing, to verify that we had the right data or a good approximation. That works for csv, but with JSON, it's even harder. Anyway, I'm just not sure that your target market knows to look for a tool like this or would really know what to do with it. On the other hand, an engineer might look at this tool as too down level. - No cli options - No api? I think there's a market for tools to help "data plumbers", software devs, engineers, support personal, etc. There's all kinds of application consultants/support etc that are moving data between systems for migrations, upgrades, custom exports, etc.
- rusticpenn 7y ago
  I can tell you from experience that this is primarily a Business Transformation problem more than a technical one. A tool would of course be helpful but solves only 5% of the real problem.
  - hermitcrab 7y ago
    And what are the 95% of 'real problem'? Are you talking about soft/people issues? Is it even possible to address these things with a tool?
- hermitcrab 7y ago
  Real data is almost always dirty. But the filtering and replace transforms could help with that. >I'm just not sure that your target market knows to look for a tool like this Perhaps not. I am getting traffic for things such as "merge csv files" or "join excel sheets". But they mostly bounce without downloading the free trial. Even when I send them to a dedicated landing page. Most people really hate change. >That works for csv, but with JSON, it's even harder. Currently I am only exporting JSON (not importing). Importing JSON would involve potentially flattening a tree into a table, which is harder. >- No cli options - No api? Command line is high on the wish list. Given that I am aiming mostly at non-programmers, I'm not sure an API would be useful. Note that there is a 'Javascript' transform, which adds a lot of flexibility if you are able to do some scripting.
speedplane 7y ago
> the visit to download ratio is low compared to my other 2 commercial products. Not sure why. I'm pretty sure the traffic is targeted, I am solving a real problem and the website explains what it does. Any feedback would be welcome! Who is your target audience (programmers, analysts, accountants)? What industries can benefit most from it? Your homepage is all about features and not about users or their problems.
- hermitcrab 7y ago
  >Who is your target audience (programmers, analysts, accountants)? What industries can benefit most from it? I'm trying to find that out. ;0) I think it is probably a non-programmer business analyst, marketer or scientist with lots of Excel or CSV files to munge together. But experience with other products tells me that you often end up selling to different markets than the one you envisage. >Your homepage is all about features and not about users or their problems. I would like to have case studies and talk more about types of users. But I am still learning the market. Also I find those 'sell the sizzle, not the steak' websites quite tedious. Tell me what it does!
  - speedplane 7y ago
    >>Who is your target audience (programmers, analysts, accountants)? What industries can benefit most from it? > I'm trying to find that out. You can start with why you use it.

3pt14159 7y ago

Hey I remember you from the early days of HN! Please take this feedback as coming from a good place with good intentions. 1. The website and product would benefit from having a designer involved. There is a design fashion that changes about every five years or so and you're a couple cycles behind. 2. Get someone that needs your software to talk through the copy. "Merge, split, clean, dedupe, reformat and more" into what? From where? Is this online or downloadable? Does this work for my specific workflow? 3. Screenshot looks like it is on a Mac. Is it Mac only? If not, put a Windows screenshot right next to it. I know after you scroll down it shows Windows support, but that can be easily missed. 4. Block quote testimonials: Bold or highlight the most relevant part. I'm not reading a paragraph, I'm scanning. 5. Fix the about page. Many Excel files have sensitive data and people click about to make sure that they can trust the company. You need better, more reassuring copy and a better picture. Asking for feedback is fine, but the current copy borders on "this is really new and may not be reliable" territory. 6. If your target demographic is Excel pros, explain why user defined functions or macros couldn't or shouldn't accomplish the same task that you help with. Hope this helps and best of luck!

hermitcrab 7y ago
Thanks for the feedback. 1. It looks pretty modern to me. I'm not selling to web designers, so I'm too bothered with keeping up with all the latest trends, I just need the web design to be 'good enough' to be credible. 2. I think that information is all on the front page, isn't it? I don't want to make the tag line too wordy. 3. The screenshot is Mac. But it says Windows and Mac further down. 4. I've kept the testimonials short, but bolding might be worth a try. 5. Pretty much the same 'about' page seems fine for my other 2 products. Also, no-one much goes to the 'about' page. One of the advantages of my product is that your data stays on your computer. 6. Fair comment, but I don't think they would be on my website if they we happy with just Excel!
- frequentnapper 7y ago
  the design looks modern and clear to me as well. I don't think that's the issue.
- oulu2006 7y ago
  Agree with the OP, the website needs a designers touch, it does look dated. The general standard of websites is so much higher these days, if it looks cramped and dated I don't linger. Testimonials are too long as well.
  - hermitcrab 7y ago
    What would you consider some good examples of modern websites for desktop software? BTW You would probably also consider the design of https://www.perfecttableplan.com https://www.perfecttableplan.com dated. But that converts visitors to downloads just fine.
    - JamesBarney 7y ago
      I disagree with the other users I think it looks fine. I think the biggest differences is perfect table plan focuses on a very concrete idea that is easy to communicate and uses a paradigm people are familiar with (looking at a seating chart). I was able figure out what it did within 5 seconds of looking at the website. But data transformations are a fairly abstract concept. Most users I imagine don't think "I need to transform my data", they probably think of the problem more concretely like "I need to deduplicate this giant excel spreadsheet". They will be glancing at the website asking themselves the question "Will this deduplicate my giant excel spreadsheet?" and the first picture they see will be a dataflow diagram. The user has probably never seen a dataflow diagram, and only rarely will they have done in gui based programming before. They will probably think "This might deduplicate my giant excel spreadsheet after I learn a bunch of hard stuff about how to wire these different boxes together". Before ever realizing how easy your app is to use. If I were you, I'd replace data transformations with something like "Excel power tasks", and then have a list of all the tasks you can do, like deduplicate, concatenate, merge columns, etc. all with 15 second videos showing how easy your application makes accomplishing these tasks. So users lands on site, sees a list of the tasks they might want to accomplish, clicks it, watches 15 second video showing them how ridiculously easy your app makes performing that task. Btw the apps look really awesome.

Edmond 7y ago

I am looking at this and thinking, how many times would a typical user need this? I occasionally find myself needing to manipulate excel documents but not enough to want to invest in dedicated software. Is this meant to support ETL workflows? Is it possible to create transform recipe packs as part of the product offering? Basically pre-built configuration instead of users doing this. Maybe provide the software for free and sell the recipes? Is this something you can package as a component that other ETL products could plug into? This can be either native plugins or some kind of web API front-end. Just spitballing :)

hermitcrab 7y ago
Thanks for the feedback. >I am looking at this and thinking, how many times would a typical user need this? I am trying to keep the product easy-to-use and the price low to reduce this barrier. But I guess a lot of people would rather spend several hours tweaking their data in Excel, rather than pay $99 and learn a new tool. >Is this meant to support ETL workflows? No. I imagine my target user as a marketing guy, scientist or business analyst who has a bunch of CSV and Excel files they need to munge together and clean-up to get some useful data. Maybe they build a process that they repeat once a month or with several different datasets. But they aren't data science people building an ETL pipeline. In fact they have probably never heard of 'ETL'. Data science people seem fairly wedded to R, Python+Pandas and expensive corporate ETL tools. >Is it possible to create transform recipe packs as part of the product offering? I think everyone is going to have different data sets and different use cases. I'm not sure that is possible to generalize the process beyond the individual transforms. >Is this something you can package as a component that other ETL products could plug into? Probably not.

grok22 7y ago

I think you need to experiment with different price points for the product. To me it seems like I would hesitate buying this for simple jobs due to the price itself. But maybe it it was $25-$40 I might buy it as a impulse buy for speeding my job at work. Or maybe make the free trial dependent on amount of use rather than making it time limited (transform a small number of rows for free for example).

hermitcrab 7y ago
It is already a lot cheaper than other similar tools (but less mature also). Any cheaper and it might start not to look credible. And I anticipate that some users might need quite a lot of support, so I can't afford to make it too cheap. If it is only worth $25 to you, then you aren't really my target market. I went for time limited, rather than use or feature limited, because it allows people to test the full capabilities. The 7 days use don't have to be contiguous.

atentaten 7y ago

I'm curios to know what are your other two commercial products since EasyDataTransform looks useful.

hermitcrab 7y ago
My other 2 commercial products are: https://www.hyperplan.com https://www.hyperplan.com https://www.perfecttableplan.com https://www.perfecttableplan.com

pragmatic 7y ago

In a recent job, I was a software engineer who worked with the data science people in an "analytics" group. A big PITA for us was the data we'd get from various sources (vendors/clients/whomever). Oh, there's a rando string in a "number" field in row 9,100,500, huh, that sucks. You can't always build the validation into your ETL pipeline for various reasons. Or it fails the whole file (b/c the data science folks built it that way really aren't that experienced at handling exceptions in the file loads, etc) Anyway that rambling is trying to lead up to me saying that there's a chasm between your supposed target market (marketers, BSAs, scientists) and the engineers/devs etc that could probably use some better tools for CSV/JSON munging. Excel is terrible at this job. What I often did was load the csv into sqlite to and then ran queries on the data to figure out how many records were horked. Or if it was a process we were testing, to verify that we had the right data or a good approximation. That works for csv, but with JSON, it's even harder. Anyway, I'm just not sure that your target market knows to look for a tool like this or would really know what to do with it. On the other hand, an engineer might look at this tool as too down level. - No cli options - No api? I think there's a market for tools to help "data plumbers", software devs, engineers, support personal, etc. There's all kinds of application consultants/support etc that are moving data between systems for migrations, upgrades, custom exports, etc.

rusticpenn 7y ago
I can tell you from experience that this is primarily a Business Transformation problem more than a technical one. A tool would of course be helpful but solves only 5% of the real problem.
- hermitcrab 7y ago
  And what are the 95% of 'real problem'? Are you talking about soft/people issues? Is it even possible to address these things with a tool?
hermitcrab 7y ago
Real data is almost always dirty. But the filtering and replace transforms could help with that. >I'm just not sure that your target market knows to look for a tool like this Perhaps not. I am getting traffic for things such as "merge csv files" or "join excel sheets". But they mostly bounce without downloading the free trial. Even when I send them to a dedicated landing page. Most people really hate change. >That works for csv, but with JSON, it's even harder. Currently I am only exporting JSON (not importing). Importing JSON would involve potentially flattening a tree into a table, which is harder. >- No cli options - No api? Command line is high on the wish list. Given that I am aiming mostly at non-programmers, I'm not sure an API would be useful. Note that there is a 'Javascript' transform, which adds a lot of flexibility if you are able to do some scripting.

speedplane 7y ago

> the visit to download ratio is low compared to my other 2 commercial products. Not sure why. I'm pretty sure the traffic is targeted, I am solving a real problem and the website explains what it does. Any feedback would be welcome! Who is your target audience (programmers, analysts, accountants)? What industries can benefit most from it? Your homepage is all about features and not about users or their problems.

hermitcrab 7y ago
>Who is your target audience (programmers, analysts, accountants)? What industries can benefit most from it? I'm trying to find that out. ;0) I think it is probably a non-programmer business analyst, marketer or scientist with lots of Excel or CSV files to munge together. But experience with other products tells me that you often end up selling to different markets than the one you envisage. >Your homepage is all about features and not about users or their problems. I would like to have case studies and talk more about types of users. But I am still learning the market. Also I find those 'sell the sizzle, not the steak' websites quite tedious. Tell me what it does!
- speedplane 7y ago
  >>Who is your target audience (programmers, analysts, accountants)? What industries can benefit most from it? > I'm trying to find that out. You can start with why you use it.