I'm approaching the launch of a web app that I've been working on for about a year, and I'm looking for some feedback on it.

The app is a web-based investigation management/reporting/auditing system for retail loss prevention and security. My plan is to market it as SAAS. I realize that a lot of people won't have experience in that field, but I think feedback from an unfamiliar perspective may be very valuable.

In return, I'll
(1) gladly answer any questions you may have about any part of developing/marketing/whatever; and
(2) if there's any interest, provide updates about what's happening on the business side of things and what I learn along the way.

A quick overview:
There are 5 main parts to the system.

1. A case management system -- used to track and maintain ongoing investigations.

2. A case reporting system -- used for reporting/documenting incidents (usually after resolution, often will be to report a case managed in the case mgmt system).

3. An auditing tool -- A risk assessment tool. Audits can be customized for practically anything.

4. An alert broadcast system -- A system for quickly disseminating intelligence, based on geographic proximity. Users are notified by text message and/or email when an alert is broadcasted.

5. A mobile web app -- access critical information, approve cases, etc. from a smart phone.

I don't really have any particular areas of focus for feedback. Really, anything that you think of will be helpful in some way.

Here's how to get in:
Address: http://23.21.153.133
Username: twoplustwo
Password: twoplustwo

A few notes:
-The communications features (email, text messaging, etc.) have been disabled. Seemed like a bad idea.

-The app is set up on a server running for precisely this purpose. Feel free to play around with it as much as you like, the whole thing can be entirely reset in about 15 minutes. That's also why there's no domain name tied to it.

-Some will probably notice that SSL isn't setup. In actual use it would be. It didn't seem worth the effort to setup a secure connection when the entire world can just login and see what's there.

-The mobile version is web based, no need to download anything. Most phones will automatically redirect to it by going to the main site. If not, just go to: http://23.21.153.133/m/

-The twoplustwo account has full administrator access, so you'll be able to play around with the admin stuff, add new users, etc.

-Only the actual app is online -- what a current customer would use. None of this would be used for marketing/promotion, etc.

Thanks in advance.

Fooled around and set up a case. Random thoughts:

UI was intuitive enough for a guy like me who has no idea how LP works. But still the first thing I looked for and did not find was any sort of documentation or help files. I'm not saying you need a full genie to automate the input and admin of these panels, but a help menu plus useful little quick help hover buttons in the right places will go a long way. (easy for me to say - UX on my own websites is generally pretty awful currently).

I like the little "prohibited" circle+slash icon when I hover over fields in a case you set up that I cannot edit.

In the case reporter panels looking at sample cases I feel like I want to have some sort of synopsis of the case, maybe on the left under the H2 and above the nav bar. For example maybe status, $ amount, narrative, and case type. Otherwise if I don't already know the case, I have to poke around in all the panels to figure out what the case is about.

I used to sell telephony equipment to businesses maybe 10 years ago and a common complaint was that feature overload looks great on a proposal "but what's the use if none of my employees will use any of these advanced features?" So hopefully your target audience won't have an issue with a learning curve or you can otherwise address this.

Also it's late and I'm only a hack when it comes to code, but it looks like your js IE shims are commented out for some reason?

Thank you for the feedback.

I completely agree about documentation, and I'm glad you reminded me of that. Its funny, I remember thinking several months ago that it was needed, and I think I've just looked at it so much and become so familiar with it I forgot how important that is. So, nice catch.

As far as the IE shim, it's actually supposed to be that way. Those are called "conditional comments". They basically exist because Internet Explorer sucks so badly that there needed to be a way to hack up the page if the user is using IE.

Here's a link to the one most people us, if you ever happen to need it:
http://code.google.com/p/html5shim/

First, it's good design work but it's a passive system correct?


Will there be tie-in to cctv systems?

If there are enough customers, data in the system, any possibility that one can create a risk assessment algorithm service to alert stores of patterns of shop lifting gangs?

Could you also integrate facial recognition software to alert a store if a suspect walked into the store? (there might be privacy issues here).

Would there be a push notification system with pictures/descriptions of perpetrators sent to store managers when events are reported?

The security audit portion seems to be light.

Overall, it's a nice platform to consolidate this type of data but the hook to me for selling it would be as a advance monitoring system that would actively assist retail managers rather than having another info system that they have to monitor.

my 2 cents worth.

Thank you for checking it out. You have some great points. I am actually working on a feature to include video with cases and alerts. The difficulty is that there are so many different CCTV vendors, with some of even having their own proprietary file formats, that it's tough to create a turn key solution that is directly integrated.

Instead, the strategy I'll be taking is to allow users to attach video clips in all the common formats for sharing. It probably seems hard to believe given how common video sharing is online, but for many retailers, even with well funded LP departments, video sharing is a huge challenge. Even the process for disseminating photos is pretty informal and clunky -- most just attach them to email and send them out, so it isn't very streamlined.

Facial recognition is probably beyond the scope of the system's capability at this point. To my knowledge, there aren't any retail case management systems that currently offer this feature. I could be wrong, though.

As far as recognizing theft patterns, that's actually also something I'm working on. It focuses on both merchandise patterns and people (matching names within given radius, matching addresses, SS#s, license plates, a few others.) I decided against putting a demo in this version just because there needs to be a decent amount of activity intel present in the system for it to work.

The alert system can absolutely be setup to push notifications to managers. They can receive pictures/alert data in their email as well as text messages when an alert is sent. There is an option to select a mile radius from your zip code for what stores you receive alerts from.

I'm not entirely certain what you mean by the audit system being "light", do you mind clarifying a little? If you're talking about the actual questions, those are just filler that I added to demonstrate how it works. In application, those can be very in depth with many sections/questions.

Thank you again for the comments, very insightful.

very nice design.. you got skills

I haven't played around with it too much, but I like where you're starting from!

One thing that really bugs me though is highlighting hover links in a way that moves other html elements. In this case you're making them too big and it's pushing the trailing links right (at least in my firefox browser on windows).

I think there are plenty of other ways you can make the hover standout without that annoyance.

Fixed.

Good catch, thanks.