A similar thing happened to me in college (compounded by the bank charging me another $35 every day until I had money again) and I was lucky I had a paid-for place to live and a paid-for meal plan to scrape through till payday.
Banks mostly stopped transaction reordering when it got a lot of attention post-financial crisis, but they're slowly going back to it, so some orgs and the CFPB are starting to shine a light on the problem again.
Usability testing is hard to get right. If you have 2 bad ideas, all it will show you is which is the least bad. They probably A/B tested this two-step layout with a text-only prompt and it did better.
The problem with the 'quantified self' is that more information doesn't necessarily lead to better outcomes. Even in cases where the stakes are high and information seems essential, like pregnancy and childbirth, more monitoring isn't better. That's true of 'routine' ultrasound monitoring during pregnancy (http://www.nejm.org/doi/full/10.1056/NEJM199309163291201) as well as heartbeat monitoring during birth (http://evidencebasedbirth.com/evidence-based-fetal-monitorin...).
What does work is identifying cases where there is genuine cause for concern - symptoms for healthcare patients, lack of ability to complete tasks for procrastinators - and collecting the data needed to determine if/when intervention is required.
The challenge for the quantified-selfers will be to sort the signal from the noise.
QS is a good idea, but I doubt the time has come for it yet.
It's a basic fact of life that you make better decisions with more information available, presuming you can think about that information sanely.
Humans generate lots of information, and computers can capture and record that information. It seems like there has to be some intersection, where we can write programs that help us make decisions.
That's not totally there yet, but it doesn't make monitoring bad. The worst case is that the data isn't useful right now.
You are right. I worked as BI consultant before. In our company the three main challenges we had were :
* collecting data
* having clean data
* extracting the right indicators and analytics from the data
I think the problems remain the sames with QS. Actual QS solutions lack of interoperability, do not really allow experimentations and do wrong about privacy. That's why Cozy could help: it allows to build data collectors in a few hours, a data browser is available to clean wrong data, and the best of all it allows anyone to hack new stuff on top of this new set of data without compromising privacy.
In particular, I think there are some dangerous side effects of collecting too much data for those of certain personality types or those prone to certain mental illness (OCD, compulsive hoarding, etc.) Besides the cost of collecting more data, there are costs related to its (negative) impact on the decision-making process.
Unsurprisingly, excellent advice phrased as succinctly as it could be for such an enormous topic.
I'm glad Paul Graham think that decks are on the way out, because they're a ludicrous (or at least inefficient) way of understanding what a startup does. If you have a product, show me that. If you have financials, show me those. Otherwise it becomes a competition to see which companies can dedicate their design resources to make the prettiest deck, and which investors can do the math on your '30% growth' number to figure out that you're growing from 3 to 4 users.
The advice about valuation is also great. I listened in on a conversation with very smart founders who are used to optimizing things, and they were super concerned about having a great pre-money valuation. It's tempting to focus on it because it's your only benchmark at a really stressful stage, but if things go badly it won't matter, and if things go really well it... won't matter either.
Another problem with emailing decks is that investors read them and decide, without much feedback to the founders. When founders can talk through the deck interactively with investors, they can learn which parts work and which don't, and what questions are unanswered.
I think showing the product alone works well only if the investor is in the target audience. I just met with an entrepreneur who has what I believe is a great product, but the investors he has talked with so far are just not going to be users of it.
In his case, I think having a couple of slides to help demonstrate the problem is very helpful. Otherwise there isn't a sufficient aha moment when he gets to the solution.
Certainly the business case is important to lay out, especially in the case you describe. Many decks don't do that though - they don't stand alone without a verbal pitch. They lack context and sometimes even topic headings.
I've seen many decks that have a page with a heading like "Opportunity" and a number (along with the obligatory graph going up and to the right). What is that? Competitor sales? Expected market share? Entire market size? Even if the investor is watching you pitch out loud, they may refer back to the deck and find they have forgotten what you said about that slide.
Pew has a nice little infographic depicting your exact situation: http://www.pewtrusts.org/our_work_report_detail.aspx?id=8589...
Banks mostly stopped transaction reordering when it got a lot of attention post-financial crisis, but they're slowly going back to it, so some orgs and the CFPB are starting to shine a light on the problem again.
http://www.forbes.com/sites/halahtouryalai/2012/02/22/are-ba...