I’m helping a startup founder who is outsourcing the development of a project. Tensions are running high on both sides as the project slips. The CEO is getting about two updates a week, mostly just “it’s ready to test now, the bugs are 80% fixed” and there’s no regular meeting. During the course of the project, “we’ve changed some features along the way.” They continue, “we are 50% done with bugs, putting us at about a month late. I feel like I should, as the CEO, do something about it to ensure it’s not a month late, what should I do.”
Here is my reply:
I would ask for better updates, like if you have a list of work items, get a list of which were done. Are they using lighthouse or similar ticketing system to track work items on their side? Do you have access? The work item list is a key tool you need to track progress and make decisions from.
First, from the tone, i’d get on the phone and clear the air. Don’t let emotions get the best of either of you. If he reaches the point of “fuck it” then you are the one that is fucked. The best thing is to be reasonable, professional and do the best you can given where you are today.
Keep in mind, most software projects are late or go unexpected places during their schedule…especially when feature requests change along the way, which is the case here.
One important thing you can do now is to eliminate features. I’m sure all the features seem important, but they are not. Trust me, most of your features can wait. Cut them out. Ask the developer which ones, if cut, would most get the schedule back on track, then you be the one to pick a subset of those — eliminate anything that isn’t absolutely critical. You are going for minimally viable release, and that means a skinny, starving puppy, not Lassie.
Bugs too. Decide which need to be fixed and which don’t.
Also, don’t sweat the look and feel. User friendly means “is useful” not “is usable”. If something is possible to get the job done, but just hard to use, consider leaving it as is for now and shipping. You can fix later.
Cut. Cut. Cut. You can fix later.
How can there be no regular meeting set up? Remind him that “Hope is not a strategy”.
Joe, I can't agree with you more except when you say “Don't sweat the look and feel”. In my experience, a hard thing to do with the product is often considered a lack of understanding for the market which is worse than a minor user experience bug. Sure the colors, font, etc are all low priority items. So I would just like to say that User Experience is critical in existing markets and crucial to demonstrating the companies understanding of the market andor business. But that does not mean you cannot cut, cut, cut even in User Experience, just understand the costs and benefits of any cut and it sounds like the company needs to cut, cut, cut.
True shippers know this mantra well… I've never met a feature I couldn't cut. Ok, maybe a few. But you know you're *actually* releasing when your feeling the painful cuts – multiple times.
Another thing I'm scratching my head about — “The CEO is getting about two updates a week, mostly just 'it’s ready to test now, the bugs are 80% fixed'”
Do they actually use an Internet-hosted version control system, like Subversion (SVN)? If not, they should do so immediately. Regardless of where the development team is, both the development team and the leadership should be doing regular builds. Check-ins should be daily at least — and no check-in should break the build.
Sounds like your CEO is an idea guy. They should be a maker or a manager – which would include detailed meetings about the project at which they should know which questions to ask.
Joe–
Nice post…your thoughts remind me of something I've learned from Agile/Scrum.
Manage what you can control…you can't control how long some work will take, but you can control how much work there is to do.
Often, we get frustrated and try to manage how long something will take by establishing deadlines and telling our teams to “work harder”. Occasionally that works, but more often than not it just leads to missed deadlines and unhappy teams.
Scott
Yeah, even if its a quick daily phone call. Simple discipline and best practice.
Good point. I think (as you imply conversely) that for new products it's more important to nail the value prop than how pretty things look. The term “User Experience” includes a lot of things, and each of those things are more or less important depending on the goals. By the way, have you seen this interview with Marissa Meyer of Google? I think you might like this: http://www.goodexperience.com/2002/10/interview… — nice quote: “At Google, we make a *useful* tool, and then we put a *usable* interface on top of that”
Thanks, good point. I'll relay that to them. The problem is that this CEO is non-technical and doesn't have someone to lean on. I'm wondering in that case what to tell them. Perhaps at least getting direct access to the check-in logs and making sure commit comments tie back to the ticketing system, I suppose. Any ideas?
They are an idea person, great product thinking and excellent hustler / sales / deals person, which is very important in the business they are in. But weak in project management and technical depth to manage the dev process.
Worse, I've seen the “work harder” thing lead to pressuring the devs into making poor architecture tradeoffs that en up biting later…in the form of production instabilities and ultimately slower pace of new feature development down the road. I love that quote from the 37 signals book, “Build half a product, not a half-ass product.” Thanks for reminding me of your blog post. I remember reading that when you first posted it. Great stuff!
Well said. Always seems to be hardest to cut features the fresher they are. Psychology plays into cutting
Interesting — in a remote situation like this, it's vital that at least someone in the “local office” have build capabilities to build and test locally. There's just no substitute for looking at a live product, being able to tweak it a bit, and see how it evolves (daily).
I'd say that if the CEO is trying to run a tech-related company of any kind and cannot actually walk to a desk to see that product taking shape, they are going to fail.