Ask HN: How do you handle support during weekends and holidays?
A few months ago I started a startup creating social games all by myself as a tryout, without creating a company. I currently have 300.000+ users and development has halted since I do support all day. So I decided it's time to create a company and hire at least two full-time support employees.
I'm currently doing all the support myself and it's turning into 16-17 hour marathon each day. Luckily, most users are in the same timezone, so they sleep at the same time I do. I want to split this 16 hours into two shifts, so I can cover everything with 2 people. However, I cannot ask employees to work 7-day weeks, so I'm considering what is the best approach.
One idea is to have employees do the support Monday-Friday and I would still cover weekends. Another is to tell them to work each Saturday+Sunday and have some other two days free instead. Or I could hire a third support person and make some combination but then someone would not work full time.
I'm looking for advice from someone who had a similar experience. Say you have small team and need to do 24x7 support, how did you handle it? Should I still do any support myself, or focus on developing features and fixing bugs only? How to handle support work on weekends? What did you try and how did it work?
Thanks.
If you just hire two full-time employees, you're still going to be picking up the slack on weekends or when someone gets sick, goes on vacation, or resigns.
A vendor that specializes in customer service will be able to train multiple employees on your product and rotate them as necessary to give you 16/7 or 24/7 coverage if that is what you need.
Set the goal of having your customer service vendor handle 100% of first-line support inquiries. They should have a manual that tells them what to do in any conceivable situation that could arise.
Obviously, it's a lot of work to write this manual, or even to think up/research all of the possible problems a customer might have. Start by just identifying the top 4 or 5 issues and writing down what you currently do when someone has that issue.
Give this info to the CS vendor and tell them that it is their job to maintain the manual and to contact you whenever a situation arises that is not covered by the manual. When they call you, tell them what to do, and insist that they record this in the manual and handle all future situations of that type in the same way.
You should never do customer support unless either you like doing it, or because you are concerned about getting out of touch with your customers' wants, needs, and desires.
And if you really want your business to grow, you should find someone else to write code and fix bugs for you, too.
You're not a developer anymore. You're an entrepreneur. Yes, you can be both, but you can't simultaneously be excellent at both.
You've figured out how to get 300,000 users for your games, and you're apparently making enough money to pay for employees, which is phenomenal. Wouldn't all of that talent and skill that you have be better spent identifying new products to create, new ways to monetize your existing products, strategic partnerships you could form, new ways to get customers, etc.?
An excellent book on this subject is "The E-Myth Revisited" by Michael Gerber. He talks about how a business needs three personalities: a technician, a manager, and an entrepreneur. Most people who start businesses get stuck in the "technician" role all day and keep working in their business instead of on their business. A successful, sustainable business has employees who do most (or all) of the day-to-day work and supervision, with the founder/leader free to think and act strategically.
Give them premium content or even just cool badges that other users can see in exchange.
Make it personal. Email them and ask for their advice. Listen to it even if you don't implement it all. People love to help and they love to be heard. If you do it right they'll like your game even more.
But going from zero employees of any kind to two full-time CS employees so that you can save time is foolhardy, at least in the short term. If you think you have no time to code now, just wait until you have employees yet have no training plan/manual for them to follow. I'd suggest you find one "all around" person and have them do CS as part of their job. The other parts of their job are various admin duties - mail, filing receipts, etc. If you find that, after 3-6 months, you need another employee, then you have a built-in trainer for the new person.
[edit]
Also good practice to not do any code deploys on Fridays or day before a holdiay
I also believe that prompt support and solving customers problems within a short period of time would mean a lot to users. If you are in a middle of highly addictive game and you get stuck because of our error, you might cool down and not be in the same mood to play it next time.