You're not short of effort. You're short of someone running the machine while you drive the business.

The quick answer: most launches slip because nobody owns the plan end to end, not because the team's slow. Work falls between product, marketing and community, decisions wait on the founder, and dates move one dependency at a time. The fix is a single owner for the launch plan with authority to chase, an agreed set of dependencies, and a weekly rhythm that surfaces problems before they're urgent. At an early-stage company that owner is often a fractional COO or a strong operations lead.

What can founders learn from GTA 6's delays?

Grand Theft Auto VI was meant to launch in autumn 2025. Rockstar moved it to 26 May 2026, then to 19 November 2026, saying it needed time to finish the game with the polish players expect (TechRadar). In May, Take-Two's CEO Strauss Zelnick put the game at about 18 months behind its original date (VICE). At the time of writing, a fortnight before launch, 19 November is still holding.

Even a studio with Rockstar's resources slipped twice, by a year and a half. But it moved the date twice, not every week. Nobody avoids every slip. What you control is how you re-baseline. A workable rule: if a critical-path item slips by more than five working days, move the public date once, with the reason, rather than a week at a time. You can only do that if someone owns the plan well enough to see the slip coming.

Why do launches slip even with a good team?

Because every team member is doing their own job well, and nobody's doing the job between the jobs. The developer ships the feature. The marketer writes the campaign. The community lead answers questions. Nobody checks that the feature the campaign promises is the one that actually shipped, or that the community knows the date moved.

The founder ends up as the integration layer, and the founder is also the bottleneck on every decision. That combination is where most slips start.

What does the operations gap look like?

Symptom What's actually happening
Dates move a week at a time Dependencies aren't mapped, so each delay surprises everyone
The founder approves everything Nobody else has authority, so work queues behind one person
Marketing and product tell different stories There's no single source of truth for what's shipping and when
The community hears news last Updates depend on whoever remembers to post them
The same problem comes back every launch Nothing's written down, so nothing's learned

What does good launch operations look like?

It's not complicated, which is why it gets skipped:

  1. One owner for the launch plan, with authority to chase anyone, including the founder.
  2. A dependency map. What has to be true before each step can happen, and who owns it.
  3. A weekly rhythm. Thirty minutes, the same questions every week: what moved, what's blocked, what needs a decision.
  4. Decision rights. Which calls the founder makes, and which the team can make without waiting.
  5. A written runbook for launch day and the week after, so the second launch is easier than the first.

In a regulated sector, put the regulator on the dependency map too. A crypto firm planning to serve UK customers should have three FCA dates on its critical path: the authorisation gateway opened on 30 September 2026, applications for day-one authorisation are due by 28 February 2027, and the regime is expected to start on 25 October 2027.

What does a launch dependency map look like?

One row per thing that has to be true before launch. The slack column is the one that matters: anything with zero slack is on the critical path, and that's where the owner spends their week.

Item Owner Depends on Needed by Slack (days) Critical path?
Payments live in production Engineering lead Banking partner sign-off 3 Nov 0 Yes
Launch campaign creative Marketing Final pricing 10 Nov 4 No
App store approval Product Release build 12 Nov 2 Yes
Community announcement Community lead Confirmed date 14 Nov 5 No

How do you run the weekly launch review?

Thirty minutes, same time every week, same order:

  1. What moved (5 minutes). Critical-path items only, against last week.
  2. What's blocked (10 minutes). Each blocker gets an owner and a date before the meeting moves on.
  3. Decisions needed (10 minutes). Anything waiting on the founder gets decided here or given a deadline.
  4. Date and comms check (5 minutes). Does the launch date still hold? If a critical-path item has slipped more than five working days, agree the new date and who tells whom.

Who should decide what?

Write this down once. It removes most of the queue outside the founder's door.

Decision Founder decides Team decides, founder informed Team decides
Launch date and any public delay Yes
Pricing Yes
Channel budget within the agreed plan Yes
Campaign copy and creative Yes
Community replies and updates Yes

What does it cost the founder when nobody owns this?

Time, first. But the bigger cost is attention. A founder doing integration work isn't raising money, talking to customers or thinking about what's next. And the strain shows. Communities notice when updates are late and answers are rushed, and they read it as a project in trouble even when it isn't.

What does a fractional COO fix before a launch?

Everything above is the job: the plan, the dependency map, the weekly review, the decision rights and the runbook. A fractional COO owns all five for a defined period, with the authority to chase the founder as well as the team, and leaves them running when they go.

The test of the work is what's left after it. If the engagement ends and the founder is back to chasing everyone, it didn't work. At a trading-education business, the result was a documented business development and sales process the team runs itself, not one that runs through me, and lead generation up more than 25%.

If you're weighing the cost, what a fractional COO really costs sets out UK retainers, the full maths on a full-time hire and the point where full time works out cheaper.

Should you hire a project manager or a fractional COO?

A project manager tracks the plan. A COO owns how the business runs, including who decides what, how teams hand work over, and what gets fixed after each launch.

Go back to the gap table. A project manager can fix dates that move a week at a time. They can't fix a founder who approves everything, or marketing and product telling different stories, because those are questions of who decides what, and a project manager doesn't get to change that.

A fractional COO costs more than a project manager and less than a full-time COO. Pay for one when the problem is decision rights and hand-offs rather than task tracking, and make sure the process stays behind when they go.

So who's running your launch?

Ask yourself one question. If you were off sick for a week before launch, would the date hold? If the answer's no, the business is running on you, and that's the first thing to fix.