Show HN: Survey Builder with Drag/Drop Logic Branching
visimojo.com
visimojo.com
I just have a couple thoughts:
1. Give the demo by pointing out things in the actual application. Psuedo UIs or presentations aren't nearly as effective.
2. Allow me to use the scroll function (in both directions, like on a trackpad or magic mouse) to move around the interface. At first I thought it was broken before I realized I needed to click and drag.
3. When dragging a question to reorder, change it so that you are dragging the actual question, not a copy of the question. Leave a dashed border around where the question was.
I often find it helpful to highlight the drop-able areas as well, but this may not make sense for this use case.
4. Make the Settings and Delete icons on the questions more clear. Since they are hidden in the bubbles it is hard to tell what the icon actually is. Be clear first, clever second. If one of those has to go, get rid of clever. (paraphrasing Jason Fried)
Very nicely done. Keep iterating and you'll have a great product.
Same kind of in-out flow but for marketing, sending emails, segmenting contacts, etc. The fun part was drawing the bezier curves using canvas between each node!
1 - Form Item labeling. I can't say "First Name" = 'firstName' in wufoo, instead it gives it an arbitrary id of form_1_154. This is BS and means I have to create a mapping table.
2 - Embed on site. Give the ability to embed the form as-is on an existing website without having to rely on an iframe or other such hacks.
3 - API. API. API. Allow the creation, deletion, edit, export, and import of form information.
4 - Internationalization. Allow me to translate my text in X languages.
5 - Reporting. This is where wufoo really has a good setup.
"Visimojo is developed by Martin Angelov, a 23 year old developer / designer from Bulgaria. What makes this project interesting is that it shows how a kid from the farthest corners of Europe can build a service that can teach the incumbents in a $7 billion industry a thing or two about interface design."
That solved my problem.
I do think that there should be a minimum resolution set as the application is unusable on small screens. Any suggestions on how to handle it gracefully would be greatly appreciated.
Also, the mere presence of such a 'feature' just makes me think that the design wasn't thought through well enough. I actually don't know, because again there's no effort made to convince me otherwise.
Q: You where asking permission to perform tasks when I was not using the app on my google account. Why is that? :-)
Tip: Id love to see you obscure the URL instead of using ID: http://visimojo.com/survey/take/114 as it would be extremely easy to ruin other peoples surveys. :-)
Q: I will definitely use this app when creating my surveys, and I would very well will be willing to pay for it! What would your pricing-structure be?
Would you consider charging for each Survey or each Response instead?
I'd like to keep pricing simple - pay a fixed price and use it as much as you like. This to me is both easier to implement and will encourage people to use it more.
But about the pricing; consider this: * You'd probably earn hellofalot more with individual pricing? I mean, I pay per sent SMS/Email/Whatever on other services. It's the usage that will cost you money in terms of data-transfers?
If I'd use your service with a newsletter of 100 000 subscribers, let's say 10% would answer my very quick survey, you'd have to host these 10 000 users throughout the survey with $26 in your pocket.
It might be sustainable in the long run if you have enough users that does not utilize the service that much - but with a couple of enterprise users you might be out of luck(?)...
I'd still consider thinking twice about your price-structure, you might want to charge 24/month, including X answers or something?
I'd certainly be willing to pay a fixed price per month (Maybe lower) and then $0.xx per answer to my survey?
Best of luck! :-)
All I did was move the box. The connections remain the same. The "No" connection should be distinguished from the "Yes" and default connection.
Other than that, looks awesome, and really responsive on my slow netbook which is great for an app like this.
I should handle the resolution better - I set 1366x768 as a base minimum for screen size, but a width of 1280 is perfectly acceptable, and I will update it in a moment.
Moreover, why shouldn’t I be able to use this at 1024x768?
Looks like a great product, but I think most of the people building these would be less technical. I saw "rick click to delete" while on mac/chrome, there's no mouse play built in.
Maybe consider simplifying the initial screen to just one block and a "do this next" message so that the user can try it? At least for lesser tech users, if you can define where they're coming from (or more often).
ex: split facbeook / twitter / HN traffic in and make assumptions as to what kind of users are coming from where.
1. It would nice to have more advanced logic. For example, if I had a matrix question ranking my like or dislike of an array of sports, it would be nice to allow control flow so that I could ask follow up questions on sports I didn't like. You could do that now with separate questions for each sport (Do you like basketball? Yes/ No -> Explain), but if you rated them in a matrix there is really no way to do that as far as I can tell.
2. Using previous answers as variables of following questions. (What did you order for lunch? Sub/Pizza) -> (Did your find your %answer% to your liking)? You could accomplish this by having the Sub answer follow one path and the Pizza answer follow another path. But that would create separate questions for reporting when really you just wanted to know if they enjoyed their order, but you'd like to personalize the question based on their own answers.
3. Another logic example might be that regardless of an answer on question A, you'll go to question B, but question B might have some of it's options disabled based on the answer to question A. Again, your system could handle this by branching each answer on question A to it's own path with a modified version of question B only showing the relevant options, but I would again assume that this would create separate question B's in reporting when you might not want that.
I do realize these are edge cases and someone needing this functionality might not be your target audience. Just thought I'd share.
Great, great job so far. As others have said, it looks really good and I found it really intuitive.
I've used Google Consumer Surveys and Survata (highly recommended :) http://survata.com/ ) recently, and I didn't feel constrained by the survey choices. Did you detect a market need for this?
I wonder if you could also let it serve as a Wizard for people basides just a questionnaire. So it would suggest results based on a user's choices. (That could include links, but I don't it should automatically re-direct to the page.)
Depending on how simple you want to keep it, it might make sense to add a more advanced logic than just checking one boolean. For example, maybe let a later box depend on which specific choice was chosen earlier.
I initially started with a DSL that would generate everything for me but I decided a UI would be nicer.
Great job. Ours will never be available for public consumption and has to be tightly integrated into our site so it'll never see any kind of third party usage, but it's nice to see someone else doing it too!
I'd prefer to click a few times to fix a mistake then have to click twice for every answer.
You could also allow users to select Text-to-speech or integrate eLance or http://www.voices.com/ and sell human voice-overs.
What I'm really looking for is a surveying solution that helps me conduct Relative Importance Testing. I haven't seen that anywhere out in the market.
I have a suggestion along the lines of what nathanbarry suggested, I thought using scroll to zoom in and zoom out, and having collapsible (tree structure like) questions will make it easy for users to create long polls.
All the best :)
Slightly off-topic; there have been a ton of great submissions "Show HN" posts lately. I find myself signing up for most of them just because I either think they are useful right now or will be very shortly. This is no exception. Keep up the great work all.
Missing scrolling with the mouse, now I need to drag to scroll... Scrolling with the trackpad on a Mac would be so neat!
Keep up the good work!
My employment for the past 6 months has been on a wide and deep dataset collection tool, one that takes in arbitrary questions that can have both validation rules and workflow rules applied.
Getting the tooling designs correct and usable seems to be the hardest part.