6 Ways Tinder ‘Smart Photos’ Gets It Wrong

Posted on

Today, there’s a new photo testing tool in town that’s setting Reddit on fire: Tinder‘s Smart Photos. How does Tinder Smart photos work? DOES it actually work?

Basically it rearranges your photos based on which photo is likely to get a right swipe.

The biggest pro of Smart Photos is that it’s in-app, so you don’t have to leave Tinder to use it. The biggest con of Smart Photos is that it’s not accurate. Here’s why:

1. Only gets data when someone swipes right

Chances are, you’re not drowning in right-swipe data.

Especially if you’re a male, your profile gets left swipes the vast majority of the time. This means that the rare right-swipe is the only time an algorithm can learn about what works. But since you’re probably not generating a heavy stream of right-swipe data for it to work with, it’s making statistically questionable decisions based on tiny data samples.

On the other hand, Photofeeler voters provide data with every vote, not just when they swipe right.

Photofeeler is a better alternative to Tinder Smart Photos photo ranking testing

2. Data is polluted by other variables

Did they swipe right because of your primary photo, or because of photo #3? Or because of your profile text? Testing your whole profile at once pollutes your test with numerous other variables.

Photofeeler gives you independent feedback on each photo.

3. One-dimensional feedback

Okay, so women don’t like your new photo, but why? Maybe you look hot, but untrustworthy. Maybe it was just cropped strangely, or there’s something in the background that’s a red flag.

Photofeeler tells you exactly how you’re coming across in photos, and users leave notes to help identify specific problems.

Photofeeler is like Tinder Smart Photos photo ranking but better

4. Experiments on your matches

As Tinder’s Smart Photos experiments with your profile, it is by nature showing inferior versions to dozens of people who might have been great matches!

Don’t waste these match opportunities! Test your photos BEFORE you use them live.

5. Not adjusted for voting style

One of your photos got a right swipe, but maybe that user swipes right 30% of the time. An approval from a 2%-right-swiper should be much more valuable.

Tinder explicitly says they’re just counting right swipes and not taking into account voter style or anything else.

Only Photofeeler‘s machine learning algorithms adjust to the specific behavior of voters to value the votes appropriately.

6. Doesn’t tell you how you compare to the competition

Smart Photos compares your own photos to each other, but Photofeeler tells you how your photos stack up against the competition.

Photofeeler is an alternative to Tinder Smart Photos -- like Tinder Smart Photos but better

Then there’s the most important thing of all:

Tinder says Smart Photos led to a 12% rise in matches.

Photofeeler users report 3-5x (200-400%) increases in matches!

While Tinder Smart Photos might be useful for reorganizing your photos for different user preferences, for choosing which photos to use in the first place, use

Go to now and give it a try!