Colin Delany February 21, 2007

Five Must-Dos When You’re Planning A Web Project


Guest article! Tia Sumler, an old friend and the web guru at civilrights.org, has plumbed the depths and scaled the heights of her experiences with online projects and put together a list of things that you just might oughta do when you’re starting out on something new. This piece, like Heather Gardner-Madras’s recent article on choosing designers, can save you a ridiculous amount of time, money and grief (you have been warned).

Five Must-Dos When You’re Planning A Web Project

New online campaign? Site redesign? Transitioning to a CMS? Whatever online project you’re undertaking, here are five things that you must do. (There are a lot of other things you should do or could do, but I’m a busy lady. Five is all you get).

1. Assess your Needs

Get agreement on the goals and final product internally before you go shopping. There’s nothing worse than realizing a month into a project that you aren’t exactly sure what “success” looks like, or why you ever went down this road at all. Or that your consultant/development team has built you exactly what you wanted — but it isn’t at all what you need. One way to force yourself to be disciplined is to draw up a RFP that lays out everything in writing, even if Bob, your webmaster, is doing most of the work. It’s about process. The key elements you need to pin down are:

  • Purpose — Why are you doing this? What would success look like? No consultant or developer can tell you this. If you don’t know, you should not be launching the project.
  • Audience — Who are you trying to engage? How? Why?
  • Strategy — At least a basic plan to achieve the goals. Tactics and features can come later, but you need to have a concept.
  • Budget — How are you going to pay for this, and what are you willing to spend?

2. Decide on a Decider

Although some things can and should be done in a committee format, you always need one individual with the authority to make unilateral decisions and the responsibility to keep everything on schedule. Most contractors will require that you identify this person from the outset, but it’s better to know before you’re sitting in a room with them, looking dumb and pointing fingers at your colleagues. In some situations, the Decider will be obvious. If you’ve got a choice, though, you might want to consider some qualities of a good Decider:

  • Understands the goals of the project and can convey them verbally to others
  • Familiarity with lingo, technology, concepts of online campaigns. You don’t need a developer, but you also don’t want to need an interpreter for basic communication
  • Affable — Some people think that putting a dick in charge will keep the trains running on time. In fact, dicks will piss off everyone along the way, which usually results in all sorts of overages over petty bull crap. Pick someone you’d want to drink beer with.
  • Anal — Note, I’m differentiating “anal” from “dick.” A person can be detail-oriented and firm and still be nice.

3. Shop Around

Once you have a handle on what you want to do, start asking colleagues, friends, contacts, listserves, or even people you meet at a bar if they have any suggestions as to vendors for your project. Yes, you can also use google and the yellow pages, but personal recommendations are more reliable and you can ask about all the intangibles, like whether the company is trustworthy or whether the guy in charge of design smells funny. No matter what you hear or from who, make sure you talk to a few shops before settling on anything. This process, where the ideal meets the real, often results in revising your RFP, your expectations, your budget, etc.

4. Scope It Out

Regardless of who is actually contracted to do the work — a consultant, a vendor, your webmaster — you need to be absolutely clear on exactly what will be delivered. That’s what the scope is for. Clearly, you can’t always get everything you want for the cash you have to invest, and/or the technology is ultimately not worth the ROI. So, there will not be a 1:1 relationship between the RFP you laid out and the scope, although they will cover the same ground. The scope will have a lot more detail, including a specific timeline and cost estimate. Remember that the scope is what you are agreeing to when you sign the contract — not what a vendor said in a meeting, not what you included in the RFP . This is why you must go over the scope in excruciating detail. Even if it seems beyond obvious that something would be covered, if it isn’t in the scope, it ain’t guaranteed. Usually, this is not about a developer trying to screw you over. It’s about having different expectations/understandings of the project. The scope will probably change over the course of the project, but the better it is at the outset, the fewer changes will be necessary.

5. Expect Problems

Even if you plan for months, obsess over every detail, have daily check-ins with your contractors, etc. etc. there will be problems. Deadlines will be unavoidably missed, features will prove to be impossible (read: too expensive) to include, staff will leave one team or the other, you’ll become embroiled in a legal battle over image copyrights — all sorts of stuff that can’t be predicted at the outset. If you anticipate the problems — or at least the time and cash it will take to deal with them — they become much more manageable. How do you plan for the unexpected?

  • Include it in your contract — make sure there are provisions for what exactly happens if the project is behind schedule or you are not satisfied. Make sure that you have budgeted at least one round of feedback and changes in every step of the process.
  • Build in extra cash flow. If you don’t need it, sweet! But if you do, and you’re running on empty, it sucks.
  • Build in extra time for project completion. Never, never, never believe the date you see on the initial scope.
  • Build in extra staff time for planning, reviewing, and revising the project
  • Nip it in the bud — make sure that throughout the project there are regular meetings/updates on project status, and dedicate time to really reviewing what’s being developed step by step. This does not mean you should stalk your vendor, it just means you need to be involved throughout the process.

I can’t stalk the vendor? But I had these really neat night vision goggles all picked out! At least you’ll still let me pick one up in a bar….

Nice work, Tia — beers are on me next time.

cpd

Back Top