Show HN: Staffjoy Boss – Automated Employee Scheduling
staffjoy.com
staffjoy.com
Today we are excited to launch Staffjoy Boss, our automated application for employee scheduling. At the core, we have rewritten how scheduling works so that we schedule workers when they prefer to work. We have also added features like time off requests, clock-in/clock-out, and employment law compliance we make it easy to have Staffjoy run a business of hourly workers.
On the technical side, we've rewritten our algorithms to search beyond optimal staffing levels for canonical optimal solutions that make employees happy. So, if one person prefers morning and the other prefers afternoon, from an optimality perspective it doesn't matter who gets assigned to which shift - but we'll try to put workers in their preferred time slot so that they are happier and more productive.
Finally, today we are also announcing our first round of funding today with participation from Semil Shah, Ray Tonsing, David Ulevitch, and others. More details are available at https://blog.staffjoy.com
I see a lot to like (from what I can read from the docs, I've yet to register for the demo to play around).
I do find it interesting that your example for scheduling has such a short window (I realize this is just an example): schedule to be ready to view 3 days before week start:
https://help.staffjoy.com/docs/schedule-timing
As a colleague is fond of saying: rotas and schedules aren't for knowing when you work, but for knowing when you're off (so you can plan your free time).
Does staffjoy work well with most employees on a set rotation (eg: I know when I'm off in November 2016, barring any exceptional circumstance, just as if my off-hours where every Saturday and Sunday. This is especially important when working nights -- as having consecutive days off doesn't really help much to alleviate the extra strain of working nights, if the worker cannot plan their downtime (eg: have coffee with a friend, plan to take part in a group activity etc).
I do see that you're not optimized for 24/7 work yet - which is fine of course (especially since you actually document it):
https://help.staffjoy.com/docs/running-a-247-business-on-sta...
Now, the reason I bring up rotations and long-time planning, is that I strongly believe predictability correlates with reduced stress for employees (and probably managers as well -- but your "pick-up" system for eg. sick days seems to a lot to help managers already). In Norway the law for overtime is different for managers and people that are able to determine their own hours - the idea is that if you can actually decide when you work, it's a lot less of a problem to work, say, a 100 hour week, than if someone else decides when you work.
Finally about day/night shifts -- does your system make an effort to prefer scheduling shift changes early to late, rather than late first, then early? That is the final issue I've seen a lot at my work-place -- with seemingly random changes from late to early. Apart from being illegal (late shift ends at 2300, early starts at 0700 -- meaning that there's not enough down-time to comply with local labour law, never mind local agreements with the union) -- it's an example of a completely unnecessary extra strain on the workers due to -- IMNHO -- poor management (and/or poor tools for shift scheduling).
In the general case, I think one would find that scheduling eg: 3 early shifts in a row, then 3 late shifts would be preferable to most people, rather than say 1 late, 1 early, 2 late, 1 early, 4 late -- even for those that in general prefer working the late shift.
At any rate, looks like an interesting product. I'm afraid we'll never use it at my current work place, as all purchasing decisions are centralized (for a large, international chain). But I'll make sure to mention it to friends that manage smaller businesses (that also don't operate 24/7 - so a better fit overall :-).
We're working on a new feature called "fixed shifts" so that businesses can set up similar rotations week over week. We've heard that businesses with a mix of full-time and part-time workers prefer to give full-time workers the same shifts week over week as much as possible (thus establishing a a consistent base), while using part-time workers to fill expected busy times, time off, etc. We're working on it - stay tuned!
With regard to 24/7 - today's release fixes previous technical limitations on 24/7 scheduling. We're amid revising our help docs to reflect this.
We agree with giving predictability where possibility - by setting your preferred working hours, we build in a lot of these schedules. However, changes happen often - holidays, vacations, sick times, hiring's ebb and flow - so changes have to happen quite often.
Staffjoy enforces a configurable minimum time between shifts for roles so that workers have enough time to rest between shifts.
Some places require by law that there are 11 hours between shifts. To avoid jet lag, add the requirement that 11 hours of downtime happen at the same time every day, in UTC or otherwise ignoring daylight saving time. This applies even on days when employees do not work. For example, a Monday-to-Friday daytime worker needs the 11 hour downtime scheduled on Saturday night even though he works neither Saturday nor Sunday.
To move that 11-hour downtime to some other part of the day, there are two approaches. One approach is adjust by half an hour each day. The downtime becomes 11.5 hours (a half hour is added on one side or the other) and then back to 11 hours (a half hour is removed from the other side). The other approach is a brutal sudden shift, but giving several days off for recovery.
Managers can also set "working hours" for workers - which, for a 24/7 business, can turn the hours they are eligible to work into a hard constraint.
For example, suppose the worker gets 6-hour shifts. It's fine to start at 3 AM on some days and at 9 AM on other days. That leaves a consistent 3 PM to 3 AM that they get for sleeping. Some days they work right after they wake up, and other days they start later.
It's an interesting problem because there are so many ways to define optimality. Does it mean working with the same manager? Working at the same facility? Minimizing travel between facilities? Number of shift in a week? etc. The problem can be extended with reinforcement learning with employee feedback. If there are manager preferences as well, there is a two-player problem with two perhaps competing interests.
I would love to work on this problem with high school scheduling for building both a master schedule and individual student schedules. No clue where I would get data though.
It's a very cool problem.
http://www.aistartups.org/2016/02/interview-with-philip-thom...
Class scheduling for schools is a huge issue but selling to schools is a nightmare. That means that turning it into a business would be really hard.
1. Who are you targeting this to - small businesses or large managed groups?
2. How will you compete with time trackers as they get integrated with POS? Does yours integrate?
3. How are you going to handle training? In my experience, people in the restaurant business do not know technology and any type of software requires extensive training.
4. Do shift managers find value over using their own existing methods?
The pricing seems ok but it's a tough market. Small businesses with hourly workers do not have a ton of money, while larger groups already have solutions in place. Just my 2 cents.
1. Who are you targeting this to - small businesses or large managed groups?
Our target customers are in the range of 50 to 150 employees per location, and we support multiple locations (e.g. a chain). We've worked with teams as small as 8 and as large as 400.
2. How will you compete with time trackers as they get integrated with POS? Does yours integrate?
The main benefit of time trackers in Staffjoy is that we can compare them to scheduled shifts. Detecting problems (missed shifts, truancy, clocking in without a shift) becomes significantly more powerful, and over time we can begin to model the performance of workers.
We support integrations and are amid publishing our first. At least one app is looking to trigger clock-in/clock-out events from their courier app to our API using this. We have an open-source Python client for our API and will be publishing better API resources in the coming weeks. https://github.com/staffjoy/client_python
3. How are you going to handle training? In my experience, people in the restaurant business do not know technology and any type of software requires extensive training.
We're onboarding in new verticals very manually right now to learn the best way and develop more scalable training.
4. Do shift managers find value over using their own existing methods?
Yes, because we automate a lot of decisions. The software shines when a new shift needs to be covered - e.g. somebody calls in sick - and the manager can find a replacement without having to recalculate a variety of overtime scenarios.
Secondly, how would this work for an events based company that may have 8 events throughout the world during a given time period?
Boss is our product for employees -where workers are assigned shifts.
They're very similar products, but we can't tell contractors when to work - they pick - and we don't have compliance built in for contractors (mainly because they pick when they work).
For the events company - we have timezone support for each location. As you start to add locations with different timezones, Staffjoy begins to show more localization information to aid in scheduling.
We define a workday as midnight to midnight, and you control the workweek start day when you create the trial.
Other nice-to-have features include that we try to give consecutive days off (like a weekend) when possible and we allow you to control the minimum amount of time between shifts to prevent "clopens" (e.g. when a worker closes at midnight then gets assigned the opening shift at 7am).
If you have more detailed questions, I'm happy to answer them via email too (philip at staffjoy)
1) If you haven't been scheduled yet - a sick day or approved time off request blocks you from being scheduled
2) If you have been scheduled - Staffjoy removes you from the shift, and other workers can claim it. (When showing available shifts to workers, we maintain compliance so that only workers eligible to take the shift can do so).