Hacker News Reader

Comment by myth_buster | Hacker News Reader

I'm surprised by the reception this is getting.

I may be wrong here, but my approach to this would be completely counter to the one taken. Instead of overhauling POS systems (which are painful/tedious to do), I would try to build an OCR app that reads the items and does all that viz in the app, saving some paper and ink in the process.

In addition to above constraint with POS systems, I don't have one stop shop for getting all items in one category (say groceries).

View on HN

Your solution may be the better product, but info on receipt has 100% user adoption compared to the extreme minority that will put in the extra work to download the app, scan the receipt, etc. Distribution is the big win here.

If you overhaul the system to provide those data via an API (instead of OCR-ing receipts or printing these relatively useless visualizations), then not only could DIY folks like the parent and myself do the analyses ourselves, but any number of personal finance tools could also support the system to make it easy for their customers to visualize their own data.

> info on receipt has 100% user adoption

How do you define "user adoption"? If a customer never looks at the receipt, then has the customer adopted the practice in any meaningful sense?

Same. Not everything fits into a single category. Is dried fruit "fruit" or "snacks"?

Also, I don't find the bubbles easy to scan or read.

If I was going to redesign the receipt, it would have everything I bought in descending order by price. The biggest pain point is how did my bill get so high? Your high value items will explain that much faster than these arbitrary categories.

[Item 1]: $56.78

[Item 2]: $28.48

[Item 3]: $10.65

Etc.

If that wasn't enough, then maybe a small table at the bottom of items in price categories:

$0-$5: 15 items, $38.94

$5-$10: 1 item, $7.52

Etc.

Even without the categories ordering the items in descending value with the accompanying graphic representation make a huge difference in the ease of parsing where your money went. Those two changes alone would achieve much of her goal.

I would love to provide an email address and have the receipt sent to me in a machine readable format. It would allow me to automate tracking how much I spend on each item at the grocery store, for example.

Apple do this and I’ve shopped at some other places that also do it.

I don’t care about a printed receipt. Find a way for me to have proof of purchase without.

And the thermal paper used will fade out after a while, which can be a problem when you need it for warranty purposes. I scan the important ones, but it would be easy to have it already in electronic form at the point of sale.

This guy made it for New Seasons, a local high end supermarket.

https://github.com/dtbrad/grocer

You can get them to email receipts.

It's easy for techies to immediately jump to the "just use an app" solution, but the reality is that the vast majority of regular grocery store shoppers still use and rely on paper receipts, and that isn't going to change anytime soon. It's better to come up with an incremental solution that will be adopted by 70% of the population rather than a perfect one which will only reach 10%.

I wish we could skip that step altogether, and have that info gathered at the POS and sent to me by the method of my choice (could be through an email, etc) which I could parse myself.

Some retailers already give you the option to send an email receipt which I assume is tied to the card used.

Paper, sure, but this is thermal paper, so no ink is used. You could arfue that the chemical coating on the paper is equivalent in resource cost to ink.

Reply on news.ycombinator.com