I am toying with the idea of giving agile software development teams budgets that they can spend to improve velocity. A typical agile team has 5-9 people.
I want to get a sense of how they might spend the budget.
If you wanted to improve software development on your team (no matter what software process/framework you use) and you had $1000 to spend, how would you spend it?
Similar questions: How would you spend $150 on developer tools?
A second, 20+ inch LCD monitor for each person or, if that's already the case, better chairs.
I'd get the team:
In that order (unless there is a marked lack of memory, then it's first).
$1000 isn't a lot (eg. to spend on training or something similar). So i would invest it in the tooling and hardware.
Some Various and inexpensive software that I personally use:
I would definitely go and ask the team this question though, chances are they know the tools or hardware that they need that will make their jobs easier. After all, they will be the ones using whatever is purchased.
I'd spend the $1000 on a banner that says "$1000 is not a significant amount of money in software development!" Licensing pretty much any tools you'd want or need is going to be more than that, and while you can buy a low-end computer that way, if your teams need hardware that badly, you're already starving them to death.
It MIGHT be enough for a book budget for a short while for a small team. MIGHT.
This really depends on your team.
Depending on the size of the team, $1000 isn't much. Developer tools go a long way - ie: things like Resharper can do wonders to increase productivity, and are inexpensive.
That being said, if your culture supports it, $1000 can be a very effective motivator as a bonus.... I'd argue that it can bring more results per dollar than any other single thing, if handled property. This is a tricky thing to manage, though, so it really depends on your team.
As everyone else has noted $1000 isn't going to do a lot in terms of better software. So I'd concentrate of the morale of my developers.
Pizza, Mt. Dew and a nice Espresso machine.
EDIT
Actually I did think of something $1000 would buy. Some testing systems running some kind of VM software. The hardware would probably be around $250 a pop, and you can find CRTs for free. Since no one is going to be using them for prolonged periods this should be fine. Back when I was using Ardence Smart Clients each license went for $125 per system so you could probably get 3 testing systems with $1000.
"So I'd concentrate of the morale of my developers."
Me too - how much do tazers cost?
For agile teams, you might be interested in agile-oriented project management tools. From what I hear (not having used them myself), both Mingle and Scrum'd are pretty good. If you're not interested in that, ReSharper is an indespensible tool for a .NET developer. You might also look into setting up some sort of continuous integration system.
You could buy a dozen or more books with that. Has everyone on the team read The Pragmatic Programmer?
Some flowers and a air humidifier, and a pen if I didn't just break the budget.
If you have a sufficient number of developers, have a voluntary weekend coding contest to solve some appropriately sized problem relevant to your business. The $1K (or whatever is leftover after paying for pizza, etc.) goes to the winners (or maybe have 1st, 2nd, 3rd prizes). If you group people into teams for this, you will (hopefully) build some esprit de corps.
Remember, the fundamental, slowest and most expensive problem in software has nothing to do with tools or productivity - it is knowing that you are building the right thing. Everything else is easier than this.
As far as improving the velocity, you have to know that there are obstacles, then what are they, and how can you remove them, and can the $1000 be used to do this? If there aren't any obstacles, what makes you think the team can increase velocity?
A lot of things which cost almost nothing can improve the environment: reducing pointless meeting requirements, taking out phones, improving stakeholder involvement, etc. All the best practices have been covered in the literature: PeopleWare, Mythical Man-Month, Code Complete, Writing Solid Code, etc.
$1000 won't go far, but it will start a library, so I think books are your best bet. I would let the developers decide how to stock the library. I would also give each developer a marked part of the shelf where there own personal favorite books can be stored, just like a bookstore where the people recommend their favorites. There is also the possibility of Safari subscriptions or ACM or IEEE memberships which come with digital bookshelves.
If the team doesn't already have whiteboards in every workspace, both private and shared, change that.
Put a high-end SSD and 4GB RAM in every machine. With the $500 you'll have left, get everyone at least 2 20" flatscreens.
I'd take that money and create more incentives for your programmers. If they have some sort of bonus for making some level of measurable improvement in some aspect of your business, then that can motivate them in a unique way.
Oh come on now... no one has said this yet?
Without a doubt, get a copy of FogBugz up and running for your group.
Take the team out to lunch to a mid-priced restaurant (not pizza, out to lunch) once per week. When the $1000 runs out (that duration depends on your area), fight for more money.
The point being that morale has a huge impact on development speed. Knowing the people you work for/with care about you as a person and not just a resource means you care about them, and what you're doing for them, in the same way. Being motivated to do good work at a good pace can provide just as much speed/velocity/whatever as the best tools (and you can't afford the best commercial tools for $1000)
It's worth consulting the team to see what they prefer, if you have a couple options. If you can take them to lunch every week, or a baseball game twice, ask them what they think. The baseball game outing may sound extremely fun to you if you like baseball, but could be the most boring thing in the world to half the team. (speaking from experience on this one)
If your teams running on Windows Vista, pre-order Windows 7 and give them the relief that they are finally free.
Divvy the money up into a book budget for each team member and send them to this question for ideas. Particularly if they're willing to use AddAll to look for bargains, that'll go at least a little ways to supplementing libraries.
I would check out Axure, http://www.axure.com/, I haven't treid the free download yet, but I will be soon, and a prototyping tool for even a single person to go through and work out the requirements with the customer would probably help you a lot, and I know that a license is around 500 bucks. Proper design to fit the needs of the customer will probably give you the most bang for your buck.
At a thousand dollars, about the best you could do is invest in moving your developers into a dual monitor arrangement. It sounds silly to management, but even if you realize a 1% performance increase that is a pretty great rate of return on a developer's salary.
The interesting part of course, is that research has shown the figure is actually closer to 20-30%!
http://lifehacker.com/software/dual-monitor/dual-monitors-increase-productivity-168488.php
I personally have a 3 monitor arrangement in my home office (a 26" 1680x1050 Samsung that I use to read papers, a 30" 2560x1600 Dell that I use for code and a 42" 1920x1080 HDTV that I run a browser on and keep little bits of clutter on or use when watching presentations. By completely filling my peripheral vision, I don't get distracted nearly as easily. I can always have whatever I need at my finger tips and can cross reference and compare different sources without losing my place or dragging windows out of the way to see.
In the office I use a much less flashy dual monitor arrangement, but it is still a net win over just hunching over the laptop display.
$1k isn't enough to make a dent in some of the tools we would benefit by at the office (for example, altova's link text runs $1700 per person.
If all I had were $1k to spend, I'd spend it on books. Like the above folks, I'd ask the crew/team what they're interested in. All our developers have dual monitors already.
I would definitely invest in a big monitor, nice chairs and headphones...
Take each development team out for drinks on a Friday and use the $1000 to buy rounds. It will go a long way to cement your leadership and team morale.
Any team will follow a leader to the gates of hell if they know that he/she has their back!
$1000 seems to be enough to buy Visual Assist for 5-7 team members in two months. That's only for Visual Studio teams.
If I have $1000, i would like to invest to buy Ebody chair
That chair is so cool and good for entire body during the programming.
Tiger
Personal fridge.
Nothing says you care more than one refrigerator per developer. You'll gain another level of respect if you fill them up with beer first.
How would that improve development? A personal fridge has the same effect to a developer as a "manager" title to a sales person. It makes coming to work early and leave late that much easier.
It also makes developers care more about the code they write, after all, the programming desk is like a home away from home...with fridge and beer...throw in lots of t-shirts and pizzas, the entire dev team is practically your children.
You should buy a foosball table for the team. Nothing builds team morale better than letting them have fun once in a while. And let them play a game or two in business hours each day. It is very refreshing for both mind & body!
We have a table where I work and it is a pretty good reason for staying at the company. The team has much fun (and would you believe it, but we even won the town championship!).
PS - Put the table in a room where it does not disturb too much if possible.
Tell your team they can give out a $25 (gift card) bonus to anyone at any time for doing exceptional work. This might be for giving you a 30 min code review, helping you fix a nasty bug, or finishing a work assignment in above average time. With $1000, this is enough for 40 awards. Average about 0.5-1.0 award/person/week.
The time/date/reason should be recorded by single party to reduce the risk of abuse. I had this at my last job, and I was not aware of any problems.
Some people might think that using a bonus is a poor way to encourage morale, but $25 is really not that much, so its more of the recognition thing. When the awards come from co-workers instead of managers, it will usually have a larger impact to the receiver. I think its important that people are not left out, even the newest employees, so there should be a different scale for different workers.
I would try it out for a few weeks, if there are problems, then make an adjustment or try something else.