App developers? Grocery Comparison App?

[Heather]

I did a quick Google and didn't see a phone app that would do what I'd like.
This:

- Take your phone to the grocery store. Photograph the price label on the shelf.
- App runs OCR, and optionally bar code scan. Determines what item is and how much it (or equivalent item) costs per unit  (kg, pound, ml etc).
- App gives you instant MMM-ish feedback:
  -- Warning! Do not buy (over 10% more expensive per unit than usual)
  -- Buy only as required (normal price)
  -- Buying emergency: Fill your shopping cart with this item!!
- Single keystroke operation. Point at item, hit button, get an answer.  You can press another button if you want details such as when and where this item cost less.   If you have to type in items or prices, app is not good enough.

Must not require a data plan.  I'm too cheap.  The app stores its database on the phone, and syncs up when you get home to your WIFI.
Compares prices only from the two or three stores that I select. (I don't care if a store across town has a sale).
Updates price database either just from my own scans, or better, from all scans of other members in my area.

I guess to be useful to me it has to run on my BB or perhaps Android.

Developer could monetize by allowing local stores to upload weekly price lists.

Or, I suppose I could just take a notebook to the grocery store, and start writing prices down... 
 

[jsloan]

I had actually had an idea similar to this and looked into it by calling our local grocery stores (I'm a developer).  The problem I had was that all the grocery stores I called were very protective of their inventory data and were not willing to share it even if I offered to expose the data over the web for free (ie code whatever was needed to connect to their inventory systems).  They were especially not willing to share their standard prices for non-sale items.

You would need to rely on user entered data and there are a lot of issues there as far as data validation goes.  I think the reason grocery apps have lagged behind other price comparison apps is because of the lack of good pricing data available.  You could always compare the prices to amazon.com, but that's not going to help you with ground turkey prices.   

       

[Heather]

I'd be happy if it just compared the prices with others that I had scanned myself over time.
Comparison with other users or stores data would be a bonus, but not required.

[Cecil]

I had this idea a while back. It's on the list of cool to-dos, I just need to decide how high a priority it is. :)

[MoneyMage]

It's not like you can keep a static database because prices can change daily, at any time during the day. The app would have to go out to multiple specific online retailer websites, search for the product, and crawl the page to record the price. That only works if they are not hiding the price by forcing you to add it to your shopping cart first. And you wouldn't be able to get prices from in-store, just online, which is likely to be cheaper than off-the-shelf. To get prices from items from in-store you'd probably have to crowd-source the data from other app users, and that's probably not going to be updated consistently or reliably, especially since a single store location could have different prices than the same store in the next town over.

I used to do QA for a big online retailer who had their own internally-developed pricing comparison application. So I've seen the amount of time, cost, and work that goes into building something like this, while maintaining any level of accuracy. Though they were trying to have a constant analysis across their entire product catalog, not just one item at a time.

If you want to just look up one item at a time that would be something reasonable to achieve, until you get blocked by all of the retailers. If someone's building this, I would assume they're building it for their own personal use (to fly undetected) and not for general consumption. :)

[NumberCruncher]

Sounds something like CamelCamelCamel, but for smart phones and user entered data.

It would take a while for the database to be useful, but you could give the option of sharing geographic data and getting input from other users. This would be most effective in larger urban areas, of course, and it would be a bit difficult to enter information for things like produce. User interface would be really important for this.