At Clockwork, I’m privileged to have been a part of many, many site launches—hundreds—each with their own share of excitement.  It’s a special thing to release something you and your colleagues have labored over for weeks or months (sometimes years!).  However, launching a new or replacement website is a ton of work up until that magical, happy moment I’m thinking about right now.  Timing is a simple but critical factor for a successful site launch.  After years of experience, I want to make one, simple recommendation: launch your website after lunch on a business day.

Whenever you think, “When should I launch my new/redesigned website?” remember:

Launch = After Lunch on a Business Day

5 Reasons to Launch After Lunch

1. The Launch Team Has Their A-Game

Web designers, front-end developers, programmers, project managers, information architects, and sysadmins are hardworking folks.  They are often asked to perform complex, technical tasks in the wee hours of the morning in a dimly-lit office with coffee or a can of Monster nearby.  Don’t ask this of your launch team.

Instead, choose a launch time when your team will have a good cross-section of colleagues available, sharp wits, and plenty of caffeine in their bloodstream.  A daytime launch means the technical experts will be operating at their peak.  Morning is too early and evening will come after a long day of (non-launch) work.  Launch after lunch.

2. Decision-makers are Available to Make Decisions

Who are the stakeholders of your website?  Can you get away with asking them to be up and available at 01:00 AM “just in case?”  Why not instead launch the site when the folks who need to say “yea” or “nay” are awake and reachable by normal means.  Your website should have been thoroughly tested prior to launch, using the same launch procedure, but we all know that the unexpected crops up when it’s time to go live.  Stakeholders in your company can also react to immediate feedback and traffic as it comes in, since they’re awake to do it.  It can even be fun!

Launch on a business day (no weekends or holidays) after lunch, when you can reach a broad group of stakeholders for instant decision-making.

3. More Thorough Testing With Less Stress

No project manager is asleep when their site is scheduled to launch.  I’m willing to bet that ad-hoc acceptance testing will be more thorough and careful at 1301 than 0001. Even better, you can send the link to a broader group of reviewers at your company since you don’t have to ask them to be awake at midnight!

4. Testing at Load Under Watchful Eyes

Web traffic ramps up quickly in the morning.  Midnight launches bring sites online during their slowest period of traffic.  This is a mistake if you have already throughly tested your launch process—you know there’s a 99.9% chance the site will go live, so there’s no reason to minimize the number of people who see it in its first hours.  Instead, launch when traffic is at its peak and your whole team is watching the site.  In general, a website is never watched as closely as it is just following launch; make use of that eagle-eyed watchfulness.

5. You Can Watch for and Respond to Feedback

Modern web publishing systems (and custom sites) are often enormously complex.  No matter how many times the team performs a test launch (and they should!) inevitably something that worked in testing and staging breaks in production.  Your users will very often find this before your testers can; launching after lunch will accelerate the discovery of any glitches provided you have good feedback systems in place (e.g. a contact form, automated error reporting). You won’t see much user feedback at a midnight launch, since your users aren’t awake.  If your launch team was up until 2 AM it’s likely that your users will be awake and finding problems before your team conquers its “launch hangover.”

Make it a Soft Launch

Launch the site before it’s known to the general public—perform a soft launch.  The client and website teams should agree on a soft launch time (after lunch!) to allow both teams to spend quality time re-reviewing the site (you tested before this, right?) in its live, launched form.  Small tweaks can be made and pushed into production with less stress and pressure than if the site was being hammered by new users right away.  This sense of soft-launch calm prevents “quick fixes” from turning into more defects.  If possible, choose the hard launch time to be person-friendly (avoid midnight!) for the same reasons discussed above.

The soft launched website can be protected with a simple username/password to prevent someone from stumbling across the new site before the announcements have been made.  This technique is useful for new sites with a firm hard launch date, when the public must not see the site beforehand.

Or, if the site is not new, launch to a stage URL for the soft launch and switch domain names only at the hard launch time.  It’s easier to confirm that the URL change was successful than testing an entire site.

What if Something Goes Horribly, Horribly Wrong?

Launching at lunch isn’t very risky, because if something horrible goes wrong, the site isn’t going to launch.  This same property is true of sites that launch at 1 PM and those launching at 1 AM.  There’s a difference, though: it’s less risky to ask your designers, engineers, and sysadmins to hack on the site for four hours at 1 PM and launch late than it is to do the same thing at 1 AM, when everyone’s dead tired.  If something is wrong, you want careful, considered actions from your team, not sleep-deprived hacking.

Consider that an aborted site launch at 1 AM likely means that work can’t resume until 9 AM; the site may not be ready until noon, a nearly twelve-hour delay.  The same problem at a 1 PM launch could mean the site is corrected and ready for an evening launch, without sleep deprivation—perhaps only a six-hour delay.  Launching after lunch reduces your risk in the event of a difficult, unanticipated problem.

Summary

  • Launch your website after lunch on a business day, because:
    • 1. The Launch Team Has Their A-Game
    • 2. Decision-makers are Available to Make Decisions
    • 3. More Thorough Testing With Less Stress
    • 4. Test at Load Under Watchful Eyes
    • 5. You Can Watch for and Respond to Feedback
  • Soft launch when you can, especially if the hard launch time is inflexible
    • Protect new sites with a simple username/password, to be removed at hard launch
    • Launch existing sites to a stage URL, rename URL at hard launch
  • Launching after lunch carries less risk and minimizes launch delays due to difficulties

Please, don’t decide to launch sites on January 1st or at the stroke of midnight (or both).  Take my advice: launch after lunch.  You will have a better launch experience and your teams will thank you for it.