Why don’t you sell me stuff?

I really wanna buy it

We are living in this super-connected information age but we are not benefitting from it much in our shopping habits. Why is that? Information about stuff to buy reaches me from many different channels and I’m constantly filtering and sorting what I might want to buy and what I do not. But why are you not making an effort to understand me and sell stuff that I would like ?

This has bothered me for a while and now I would like to give you my ideas so you can fix this.

• I’m an iPhone user. A new phone comes out; sell it to me.
• The new KickStarter campaign with cool playing cards from Oatmeal comes out. I like Oatmeal and games. Sell that to me.
• My friend has a birthday coming up. Sell me his present and do it early enough so that I can order it in time.
• Mothers’ Day is approaching. Sell me a surprise for my mother.
• Recently I have been googling a pebble watch. This means I have an interest in it. Two days later I found out that you can get a used one for half price from Amazon. Why do I need to figure that out? Sell me that.
• I usually only remember to book a place for a New Year’s party sometime in December. But by that time there are no places left. Remind me about that in August, and sell me it.
• The list goes on and on…

We already have influencers to help/force us to make decisions. Advertisements, branding, communities, friends and networks are driving our buying habits every day. Amazon and the like have tried to build a suggestion model to show me potentially interesting stuff to buy. It’s a nice way forward ( 10 years ago? :) ) but still I feel that it’s often “buy similar stuff” or “others who have bought it also have bought that”. These linear models work with a limited range and do not use much of the information available from me. Groupon’s impulse-buy logic has at least managed to introduce me to new stuff I hadn’t thought of before, but neither Amazon nor Groupon does much to consider my habits in their models.

But we already had that amazing old story from Target whereby they used big data analysis to predict what people would like to buy and even managed to predict one girl’s pregnancy. Why do we not have access to these models and solutions? Wouldn’t it work way better if I gave you data myself? We should be able to make personalized shopping an order of magnitude better… What if we create a platform that aggregates all of my personal data and uses it to suggest when I should buy something and what I should buy?

You are thinking that it’s an invasion of your personal life? Get over it; the world is already working in this way. All of the examples above operate in our lives in one way or another. I’m reading about the new iPhone in the newspaper, about the Oatmeal card game and friends’ birthdays on Facebook and Reddit’s community reminds me about Mothers’ Day in advance, and so on. Also, it’s your choice if you want to use that service or not. 90% of internet users are already giving that information to Google, which uses it to analyze what kind of advertisements to push towards you. Why should shopping be different?

Let’s dig in and see what we could achieve with that approach:

#1 Event incentive

• I’m constantly forgetting birthdays and when I do remember them, it’s usually too late to order something from the web.
• Let’s build a simple solution that remembers my friends’ birthdays and analyzes their interests.
• Use that data to find my friend a birthday present (or maybe suggest the top 3 best picks for me)
• I can choose my favorite gift and order it with the click of a button
• Also, let’s have it notify me maybe two weeks in advance so I have time to order and receive it.
• Why not use that same logic for the holidays (Christmas, New Year, etc.) or for personal events (anniversaries, 1 year at work, etc.)?

#2 My personal purchases

• Let’s analyze my Facebook, Twitter, Google+, emails, Amazon, eBay and other online activities and create an aggregate personal profile for me.
• Now you can input new stuff that could be bought from your platform and if something matches with my personal profile then why not sell that to me?
• Let’s take KickStarter example again: Let’s say that I have been searching for a good messenger-style bag; I have googled it, gone through eBay/Amazon, and talked about it in my emails and on Facebook. But now there is a messenger bag in KickStarter that has raised 1M USD in four days. People really like it. Why not sell that to me also?

#3 Discounts

• My search history suggests that I have been looking for a pebble watch but haven’t yet bought it.
• But now there is a 50% discount on a refurbished Pebble at Amazon. Why not notify me and sell it to me?

#4 Track your purchases

• For example, if you are a runner then it’s a good habit to buy new shoes after every 1000 km.
• We know how much I’m running from my Runkeeper / Endomondo or similar app.
• When the time comes for me to get a new pair, then recommend me the best and coolest new running shoes available.
• Send me reviews, Youtube videos and comments about those shoes.
• Allow me to make the purchase with one click.
• Also, why not track dentist visits, or vacations for work addicts, or suggest date nights with your boyfriend?

#5 Not only for offline content

• It may be even simpler to build it for online content such as TV shows, movies, apps, articles and so on.
• We have an enormous amount of online personal data that is already readily available and which could be taken into account.
• For example, I like computers, sci-fi, and quality shows that have high IMDB ratings such as Black Mirror, Dexter, Breaking Bad, Fringe and Sherlock, etc. It shouldn’t be hard to determine that I would like a new series called Mr Robot. Sell it to me.

#6 The sky is the limit

• …

Playful data point gathering

For this model the quality of suggestion is kind of important, so you should always think about how to improve. Some ideas:

• Create a Tinder-like left/right fast interaction with your suggestions for an easy, fast and fun data gathering tool.
• Every time a user interacts with some element on the site, try to understand why it was clicked, what people did after the interaction and add a datapoint to your db (Aliexpress, Amazon and so on are doing something like that).
• Ask users what they like. You do not need to go wild with your analytics to analyze the customers; sometimes it’s as easy as simply asking what they would like and adding these datapoints to your db.

Final thoughts

Today I’m already doing all that has been described above but I’m relying on accidental subliminal info-gathering from my life. There are easy ways to organize/structure that information into an effective, fun and personal shopping experience. Notify me when and why I should buy something, tell me what I should buy, and make my life more awesome while also making a profit.

Invisible Payments

Embrace Weirdness