5 ms·

AirBnB: 'Let's send emails, teach [users] professional photography, and test them. "We said, 'Screw that.'" [4] Instead, they rented a $5,000 camera and went d

AirBnB: 'Let's send emails, teach [users] professional photography, and test them. "We said, 'Screw that.'" [4] Instead, they rented a $5,000 camera and went door to door, taking professional pictures of as many New York listings as possible. [1] https://growthhackers.com/growth-studies/airbnb https://growthhackers.com/growth-studies/airbnb

elvinyung 8y ago
The original blog post from PG talked about that: http://paulgraham.com/ds.html http://paulgraham.com/ds.html
echevil 8y ago
Interesting! I think it could be scalable even at much bigger scale -> Hiring more professional photographers to do that as the company grows doesn’t sound like a terrible idea. Just like Google having tons of people driving cars around cities to create content for street view
sonnyblarney 8y ago
Should be noted that the founders were effectively trained designers and had probably better than amateur photog skills. They knew what they were doing when staging/taking snaps.
- sctb 8y ago
  It's true, but I think this case is still counterintuitive. This was in the middle of the YC batch, like “Where are the AirBnBs this week? Oh, they're in New York taking pictures of apartments.” At the time it just wasn't obvious that was what they should be doing.
  - sonnyblarney 8y ago
    It's funny because I would see that as a spectacularly good use of their time. I don't even see why it's considered a hack. People are visual - we make decisions based on how things look. Their product is online, the pictures are basically everything. The snaps are 90% of what's being presented, so making sure those are good would be worth 1/2 of their time overall. The technology part of AriBnB at smaller scale is just not rocket science, it's almost commodity.
    - nostrademons 8y ago
      It's a good example of knowing your market. It's not universally true that more photos = better for your users. Google is a good counterexample. Google in its early days was famously not visual - the visual design was incredibly sparse, there were no pictures on the results, there were no pictures on the ads - and that was a huge contrast to most other sites, where you had animated punch-the-monkey graphics scrolling across the screen. The reason for this is that the primary utility for Google (in the early days) was to get you off the site and to your destination as quickly as possible. Anything that drew your attention - other than the result you wanted - was a distraction. Google's made attempts to put pictures on the results pages since - we had plenty of data showing that peoples' attention is instantly drawn to images, and when your department's name is "Search Features" it's awfully tempting to draw attention to the features you're developing - but every such attempt seems to cost Google in brand equity in the long run, and ends up getting reverted eventually. The key insight for AirBnB was that they are selling an experience, not information. When you're deciding where to book a hotel, you want a visceral sense of how it would feel to be on vacation at that place, and a picture is the best way to achieve that.
      - bsvalley 8y ago
        The reason for this back then was a slower internet connection. If Google had the bandwidth when they first started they would have inserted images and videos to their search results and ads.
    - philipodonnell 8y ago
      > I don't even see why it's considered a hack. It seems straightforward, but the alternative would have been to establish a network of local photographers with contracts and agreed rates, scheduling software, performance management, QA, etc... That would be more scalable, so if you just do it yourself instead of going for scale its a "hack".
    - cromulent 8y ago
      Agreed, but "Do Things that Don't Scale" is counter-intuitive enough that PG wrote an essay on it, mentioning AirBNB. http://paulgraham.com/ds.html http://paulgraham.com/ds.html

elvinyung 8y ago

The original blog post from PG talked about that: http://paulgraham.com/ds.html http://paulgraham.com/ds.html

echevil 8y ago

Interesting! I think it could be scalable even at much bigger scale -> Hiring more professional photographers to do that as the company grows doesn’t sound like a terrible idea. Just like Google having tons of people driving cars around cities to create content for street view

sonnyblarney 8y ago

Should be noted that the founders were effectively trained designers and had probably better than amateur photog skills. They knew what they were doing when staging/taking snaps.

sctb 8y ago
It's true, but I think this case is still counterintuitive. This was in the middle of the YC batch, like “Where are the AirBnBs this week? Oh, they're in New York taking pictures of apartments.” At the time it just wasn't obvious that was what they should be doing.
- sonnyblarney 8y ago
  It's funny because I would see that as a spectacularly good use of their time. I don't even see why it's considered a hack. People are visual - we make decisions based on how things look. Their product is online, the pictures are basically everything. The snaps are 90% of what's being presented, so making sure those are good would be worth 1/2 of their time overall. The technology part of AriBnB at smaller scale is just not rocket science, it's almost commodity.
  - nostrademons 8y ago
    It's a good example of knowing your market. It's not universally true that more photos = better for your users. Google is a good counterexample. Google in its early days was famously not visual - the visual design was incredibly sparse, there were no pictures on the results, there were no pictures on the ads - and that was a huge contrast to most other sites, where you had animated punch-the-monkey graphics scrolling across the screen. The reason for this is that the primary utility for Google (in the early days) was to get you off the site and to your destination as quickly as possible. Anything that drew your attention - other than the result you wanted - was a distraction. Google's made attempts to put pictures on the results pages since - we had plenty of data showing that peoples' attention is instantly drawn to images, and when your department's name is "Search Features" it's awfully tempting to draw attention to the features you're developing - but every such attempt seems to cost Google in brand equity in the long run, and ends up getting reverted eventually. The key insight for AirBnB was that they are selling an experience, not information. When you're deciding where to book a hotel, you want a visceral sense of how it would feel to be on vacation at that place, and a picture is the best way to achieve that.
    - bsvalley 8y ago
      The reason for this back then was a slower internet connection. If Google had the bandwidth when they first started they would have inserted images and videos to their search results and ads.
  - philipodonnell 8y ago
    > I don't even see why it's considered a hack. It seems straightforward, but the alternative would have been to establish a network of local photographers with contracts and agreed rates, scheduling software, performance management, QA, etc... That would be more scalable, so if you just do it yourself instead of going for scale its a "hack".
  - cromulent 8y ago
    Agreed, but "Do Things that Don't Scale" is counter-intuitive enough that PG wrote an essay on it, mentioning AirBNB. http://paulgraham.com/ds.html http://paulgraham.com/ds.html