Algorithms for Finding the Perfect Gift

Dec 20, 2016 · 4 min read

It’s that time of year again when, as one Reddit user put it, Christmas feels more like a deadline than a holiday. Gift giving is hard, and there’s a whole cottage industry that has emerged around the problem. There’s an element of a turing test to the act of gift giving — you’re modeling the person’s preferences, and thus testing how well you know the person.

I struggle to generate ideas for gifts for friends and family. What has helped is generating many ‘algorithms’ to optimize how I search for gift ideas, in contrast to my default approach of “think of the person, see what ideas spontaneously arise”. Below are several that I found useful, taking the outputs and alchemizing them into good gifts.

Clustering: Demographics, Interests, and Types

In general we’re not as special as we think. We can all be placed in broad categories of people that have similar qualities and similar preferences. One way to generate ideas for a gift is to take information related to general information about that person and use that to find things that similar people have used/enjoyed, identifying clusters of gifts. Basically how Amazon product recommendations work. Things like Age, Interests, and Personality Type can all be used as hints for things a person might like.

The algorithm visualized in stunning detail.

Do it:

List five things you know about the person, compare to a list in your head of similar people, and then reach out to them for gifts they’ve received that they like and/or what they want for the holidays (obvious caveat that you might set up an expectation of a gift to that person). Good question(s):

  • What’s one thing you’ve received in the past year that you love?
  • What’s the favorite thing you’ve bought for under $50?

Linking: Gifts and Emotions

The best gifts evoke positive emotional responses. What emotions are you trying to evoke with the gift. Comfort? Mirth? Maybe a solution is to work backwards on the problem — think of times you’ve received a gift that has evoked that emotion, and use that as inspiration.

Do it:

  • Think of desired emotional responses from the other person, and times they’ve shown that response? What was happening? What were they doing?
  • Pick three gifts you’ve received in the past year and visualize that moment you received them. What did you feel?

Search: Shared Secrets

Giving a good gift is showing that you know something about the other person. By remembering the small exchanges, intimate moments, and personal details you know about the other person, you can find inspiration.

Do it:

  • List moments you’ve shared with that person. In the last month? Last year? What feels particularly relevant?
  • Search through calendar for events you’ve attended with them, review correspondences (text, email) for references to small details that you could reference in the gift.

Random Walk: Experimentation opportunity

One of the benefits of receiving a gift is that you didn’t have to explicitly make the choice to buy the item — thoughts like “is this appropriate?” “does this really represent me?” “is this too much money for something frivolous?” no longer enter into the equation, and you can instead just enjoy the thing. This suggests another algorithm, namely buying something a person wouldn’t buy for themselves.

  • Luxury Products: Beauty products, jewelry, massages. Stuff that could be described as “treat yo self”.
  • Weird things: Niche products that have a 1/20 chance of being super helpful to the person. Anything from ShutUpAndTakeMyMoney.

Do it:

Imagine the person. If they were twice as bold as they are now, or conversely gave half the fucks they do now, what would be something they would own. Buy that for them.

Bonus algorithm: How much to spend?

I’d propose a step function, where different levels of relationship proximity bring different cash levels. For example, $50 — $200 for close family / best friends, $20 — $50 for friends, < $20 for work acquaintance. Multiply by a scalar depending on your personal wealth and/or grinch level.

Why do this?

There are tons more search routines/questions you can ask. Breaking the problem down into several framed questions helps turn brainstorming from a very stressful activity into a much more manageable task. I think that stress is the emotional response of the brain floundering around, trying to search through everything to find a potential solution to the problem, without actually knowing what success looks like.

your brain while trying to come up with ideas

There’s a general strategy here that seems useful to me:

  • When faced with stressful unknowns go meta and create frameworks.
  • Don’t just run the default search, experiment with other questions and patterns.

Setting up mental frameworks when faced with the task of buying a gift or for that matter any idea generation problem, turns it from a frustrating experience into a, dare I say, merry one? #forcedchristmasreference

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade