Scout7 logo

Scout7

Back

research_paper

You Said Friday. It Went Live Next Wednesday: Marketing Project Planning

September 26, 2026 · 8 min read · Scout7

You promise a date, and the job lands days later. Looking back at last time is the usual fix. Here is the detail it leaves out.

You Said Friday. It Went Live Next Wednesday: Marketing Project Planning

You said Friday. It went live next Wednesday.

Here is the most common planning tip. Look back at how long the job took last time. Studies show your next plan can still be just as late.

In one study, students guessed they would finish a big project in 34 days. It took them 55.

If you are the only marketer at a software company, you know this feeling. The founder asks for a launch page by Friday. You think it through and say yes. It goes live next Wednesday.

Nobody shouts. But the founder notices. Next time, they quietly add a few days to your dates. And you start to think you are just slow.

You are not the only one. In July 2026, Knak asked over 300 marketing leaders about this. Knak is a company that sells an email and landing page builder. 85 in 100 said their team missed at least one launch date in the past year. (Knak survey, PR Newswire, 28 July 2026)

What most planning advice leaves out

Most advice on this says one thing. Look at your past. That advice is half right. It leaves out one detail, and we will get to it.

What you will learn:

  • Why smart, careful people guess short, again and again
  • Why knowing you were late last time does not make you faster
  • The one number most people plan from, and why it lets them down
  • Five steps to give a date you will actually hit

What happened when students guessed their own finish date

In 1994, three researchers asked university students a simple question. When will you finish your final year project? The researchers were Roger Buehler, Dale Griffin and Michael Ross.

The students guessed 34 days on average. It took them 55 days. Only about 3 in 10 finished by the day they guessed.

The researchers also asked for a worst case guess. What if everything goes wrong? The students said about 49 days. Even that was too short.

These were not lazy students. This was their most important piece of work that year. They still guessed short.

So when your launch page slips, it does not mean you are bad at your job. Guessing short is normal. It happens to almost everyone. (Buehler, Griffin and Ross, Journal of Personality and Social Psychology, 1994)

But surely knowing this helps? That is what the next test checked.

Why knowing you were late before does not make you faster

The same researchers ran a second test. Students had a computer assignment with a deadline. They were split into three groups.

  • Group 1 just guessed when they would finish.
  • Group 2 first remembered how long past assignments took. Then they guessed.
  • Group 3 also remembered past assignments. Then they had to picture this one finishing as late as the last ones did.

Groups 1 and 2 were just as late as each other. Remembering the past did nothing on its own. Only group 3 got less late on average.

A later study looked at people in Canada filing their tax forms. People knew they usually filed late. They still expected to be quicker this time. (Buehler, Griffin and MacDonald, 1997)

So knowing you were late is not enough. Even experts fall into this. The next story shows how far it can go.

The expert who had the answer in the room

Daniel Kahneman won a Nobel Prize for his work on how people make choices. He tells this story in his 2011 book, Thinking, Fast and Slow.

In the 1970s, he was in a small team in Israel. They were writing a new textbook for schools. After about a year, he asked everyone how long it would take to finish. The guesses were between 18 and 30 months.

Then he asked one team member a different question. This man was an expert in school courses. How long did other teams like ours take?

The expert said about 4 in 10 teams never finished at all. The rest took 7 to 10 years.

The team kept going anyway. The book took 8 years.

The answer was in the room the whole time. It was the real time other teams took. The team heard it, and still planned from their own feeling.

So look at the past. But which past?

Why the number in your head is the wrong number

Here is the detail most planning advice leaves out.

Ask yourself how long the last launch page took. The answer comes from memory. And that memory is too short.

In 2005, Michael Roy, Nicholas Christenfeld and Craig McKenzie looked at many studies on this. People remember past jobs as quicker than they really were. Then they plan the next job from that memory. (Roy, Christenfeld and McKenzie, Psychological Bulletin, 2005)

So trying harder to remember does not help. It feeds the same small number back into your plan.

This may explain the students in group 2. They did look back. But they may have looked back at a memory that was already too short.

So if memory is the problem, what fixes it?

What fixes it, and the one catch

In 2008, Roy and two colleagues tried a fix. They showed people the real time their earlier task took. They did this in three different tests. Each time, people's guesses got more accurate. (Roy, Mitten and Christenfeld, Journal of Experimental Psychology: Applied, 2008)

There is one catch. In 2018, Kevin Thomas and Cornelius Koenig ran three more tests with 40, 80 and 80 people. A real time helped only when the next job was the same kind of job. The time from a different task did not help. (Thomas and Koenig, Frontiers in Psychology, 2018)

Think of a kitchen timer. It tells you how long the rice really took. It does not tell you how long the curry will take.

So here is the answer to the question we started with. Plan from a time you wrote down when the job ended, for the same kind of job. Not from what you remember.

Marketing project planning: how to give a date you will actually hit

You do not need new software. A notebook works. So does a simple sheet on your laptop.

  1. When a job ends, write down how long it took. Count from the first request to the day it went live. Include the days you waited for someone to say yes.
  2. Write the kind of job next to it. Launch page. Newsletter. Webinar. Case study.
  3. Next time, look up that kind of job. Plan from the real times on your list.
  4. Doing a kind of job for the first time? Ask someone who has done it how long it really took them.
  5. Give your date from the list. If someone pushes back, show them the list.

Each job of the same kind you add makes the list more useful. Your dates start to match your real weeks.

What this research cannot tell you

These studies used short, simple tasks. Counting pieces of paper. Writing a short essay. Nobody has timed a real launch page in a lab.

A written list will not stop new requests from landing on your desk. It only helps you give honest dates for the work you say yes to.

The 85 in 100 number comes from Knak. Knak sells a tool for this kind of work. Treat the number as a sign, not as proof.

The 2008 tests showed guesses got more accurate. They did not show that people stopped being late altogether.

Sources

Frequently asked questions

Should I just add extra time to every guess?

No study we found gives an amount that works for everyone. Your own list of real times is a better guide than a fixed extra amount. It also tells you which jobs run long and which do not.

What if every job I do is different?

Most jobs fall into a few kinds. Group them by kind, like launch pages or newsletters. For a new kind of job, ask someone who has done it how long it took.

How long before my list is useful?

The studies do not give an exact number. Each job of the same kind you add makes the list more useful. Start with the jobs you do most often.

Does this work for a team too?

Yes. The same idea works for a team. Keep one shared list, and write down the real time for each kind of job.

What is one job you do every month where you have never checked how long it really takes?