Thoughts on Magic+

A number of people have messaged me and asked for my thoughts on the launch of Magic+, an always on personal assistant, that they charge $100/hr for.

(I ran a personal assistant company called Superhuman, that hired real people as assistants, a couple of years ago and ended up shutting the service down because it didn’t work as well as I wanted at scale.)

I still believe there is a lot of mundane stuff us humans need to do on a daily basis that becomes a distraction from creating and doing what we want to spend our time on. I’m always excited to see when people try to make Life easier.

The following is a brain dump of ways that Magic+ may or may not work.

Ways it may work

1. By focusing on customers who understand their value they don’t need to spend time educating customers on how to use the service.
2. They are now able to hire professional assistants not people looking for a job on Craigslist. Any assistant who used to work for a fortune 1000 CEO would love a calmer lifestyle, working from home and not have to be always to be “on”. Just like Uber has for drivers, this opportunity could allow millions of people to work hours that most fit their lifestyle.
3. The same goes for the ability to hire expert concierges
4. Afford to train people to be experts. One of the hardest parts for Superhuman was training and educating people to be experts at their new job, which proved to be very costly. A higher price would make this easier.
5. People who have full-time assistants but don’t want to delegate tasks that are too personal or too menial will use this service.
6. Customer gets service for only what they request. People want to get things done and they don’t want to worry about keeping their assistant busy to get their money’s worth. As a result, there are lots of assistants/secretaries that spend a majority of their day on Facebook, and get paid a lot to surf the Web.
7. People don’t want to have to remember their assistants birthday.
8. The patterns they learn from customers they can use to help educate potential customers how to delegate better.
9. Can start offering monthly plans once they know costs for cohorts of users.
10. Technology may be invented in house or via a 3rd party that can automate lots of processes and you can still charge same amount. Harder to raise prices.
11. Building a better process helps everyone in the network. In other words, if one assistant learns how to do something better, quicker or cheaper, all of the assistants potentially learn that trick.
12. There are lots of rich people in the world who spends lots of money on lots of things and this helps them spend it to do whatever the fuck they want.

Ways it may not work

1. If they can’t build a consistent experience causing customers to lose trust in the service.
2. Hard to build habit with extravagant use cases.
3. It takes too much time for an assistant/system to learn preferences. Often times the user doesn’t even know their preference.
4. Personal assistants need to be fully integrated and need complete access to your life.
5. Lack of expertise leading customers to churn
6. Assistants aren’t happy with their jobs and quit.
7. People who don’t know how to use it properly will talk shit about the service.
8. Doing the same tasks once doesn’t mean the task will be the same the next time you try and do it for another customer.
9. Spend time automating the wrong things. Things that only make the user experience 1% better.
10. People don’t know how long tasks take and get upset about costs
11. People might churn and just hire someone full time.
12. It’s easier to trust someone that you see in person.
13. Some tasks require your personal information that you might be wary to give to a 3rd party.
14. Might get too popular and task completion time will slow down, chipping away at trust.
15. Which leads me to the last challenge: focusing on fulfillment and user experience can distract from building scalable technology.