Views of a Modern World

A recent Springwise newsletter described Perkler, a website that helps people (primarily Australians, given that it’s an Aussie company) manage their perks, loyalty reward card offerings, credit card offers, and even some potentially redeemable rewards (such as cashing in airline miles for some sort of gift certificate).

It’s an interesting application of an idea I had many years ago. I thought that it would make sense to take all of this same loyalty information and encode it in/tie it to a single uber-reward card. Either a standalone card or tied to a suitable credit card, e.g., American Express, the card when swiped or scanned would trigger access to a database that would query for the best possible reward given both your available options and a set of preferences you have set. For example, some people would prefer to maximize dollar value of rewards while others prefer dealing only in one “currency,” such as airline miles.

A similar concept to the “one card to rule them all” is the all-in-one discount card DIY tool [Ed. Added link via lifehacker.com]. This allows you to create a single wallet-sized card with several sets of barcodes/numbers on it to save space in the wallet/checkbook. I think it certainly beats putting all those tags on a keychain like some people do.

In any event, the issues with implementing the OneRewardCard are tied to privacy, database construction, and vendor incentives.

On the privacy front, I’m one of those people who uses basically a single credit card for everything (bad long-term budgeting as I was leaving the Army and heading to law school taught me some credit lessons that boil down to “don’t spend what you haven’t got – ever, except for houses and education”). AmEx already knows where I shop and quite possibly what I buy (I don’t know how much line-item detail gets sent upstream through the authorization process or whether it varies according to vendor). I wouldn’t be particularly concerned about them getting extra information, and they could certainly enhance privacy as an offering for interested consumers.

Database construction is the issue of getting up-to-date rewards information into the database so people can get relevant offers and choose accordingly. The current example of what this problem looks like without vendor involvement is the plethora of deal/coupon websites that track many old/expired/inappropriate discount or coupon codes. At one point, I resolved to search for discount codes anytime I used a new ecommerce website, particularly one I hadn’t been receiving offers for. After searching for coupon codes, I got fed up pretty quickly with the inability of Google to parse those various sites (and the carbon copy/leech versions that abound as well) and the inherent difficulty for the sites themselves to have a comprehensive listing of up-to-date relevant offers. Human involvement or AI searching/screening are two answers, but bringing vendors inside the circle of trust is another.

Vendor involvement depends on the value proposition for a vendor in offering a discount. For example, a vendor seeking to reward customers with a special discount for past loyalty will not want to share that discount with new customers who haven’t “paid their dues.” Similarly, a vendor seeking to attract new customers would prefer not to give discounts to people who would buy anyways at the normal price. On top of that, the actual number of redemptions of an offer might rise dramatically, for which marketers/sales teams would have to adjust. My suspicion is that the initial thoughts of vendors will be that this program would weaken their connection to the customer since the element of knowledge of the origin of the discount is dramatically weaker — if I get a discount on flowers under the OneCard paradigm, I don’t necessarily automatically learn that the discount came from my USAA membership or my Platinum card or from my Allegheny College alumni status. I don’t believe that this is an insurmountable problem, just one that has to be thought out. (Quick idea: the OneCard could in fact be a smartphone app that would give proper credit to the right vendor or channel partner.) The real underlying goal of vendor involvement from the consumer perspective is getting the fullest set of offers available. It might well be that in exchange for better targeting data, and the possibility of point-of-search (a la perkler) or -sale impact, the net redeemed value of perks would rise.

There’s a business here that I just don’t have the contacts, time, energy, or money to pursue. Good luck to someone; I’d like to be on your board.

After reading this post on Springwise (a site/feed I enjoy very much), I realized that I needed to post another orphaned idea.

People have seen a number of “prepared food” offerings, where a company delivers a bunch of meals to you or your family that are designed (usually) to control portions and assist in weight-loss.

The story on leViv is different because the company takes a slightly different take, focusing on specific health “issues” and producing five separate programs, one for each.

At one spot, the company makes a reference to “evidence-based recommendations,” which almost makes me think that this idea is close to being adopted.

The idea that I had, building off of 23andme’s stated goal of continuously improving their product by reference to the evolving scientific literature connecting SNPs and other genetic sequences to specific diseases or other traits, was to develop a private chef company (what I called these prepared food delivery companies in the past) that would take a similar approach by integrating nutritional research into changing and updated recipes targeted to specific client attributes. So, a study that says 1/3 cup of blueberries a week reduces heart disease risk for 40 year-old male nonsmokers means that I would get a slightly different menu. The integration of the two businesses, genetic information about potential predilections for certain diseases along with nutrition guidelines related to those risks, would lead to perhaps not genomic medicine but almost certainly to genomic nutrition.

After all, there are very few nutrition studies that talk about taking in large amounts of dangerous things like fat and sugar, so it’s probably a riskless improvement to diet planning that can only help prevent possible medical problems.

Clearly, the more individualized the assessment and menu, the more complicated this would be. However, the true private chef could use this sort of information to create an optimized shopping list and menu for a family to serve as the palette on which to base his or her creations.

I’d be delighted to hear from the folks at leViv about their reaction to my idea, and even more delighted to read more about how they are actually constructing menus for their five plans.

Why don’t recommendation systems create, or allow for, a way to post recommendations that respects poster privacy? For example, I may want to post something about a doctor that I use but not want to reveal my true name or information that reveals too much about me — say, an indication that I have some particular health problem.

Yes, there is a tension between anonymity and usefulness/veracity of recommendations, but that can be dealt with separately; it explains why this could be an option for some posters.