Who is coming
What the session is for

Finding out what the room already knows

What your group can already do decides half of what you should plan. This chapter gets that out of them before the day, and gives you a thirty-second opening check for the ones you could not reach.

Start here Essentials

This chapter asks you to find out one thing about your group before the day, and to have a plan for the people you could not reach.

  1. Write the one question you most need answered before the day. One, not six (section 3).
  2. Send it where they are already replying to you. The sign-up confirmation, the group thread, the email that told them the room (section 3).
  3. Plan what you will do with each likely answer. Including the answer that nobody replies (section 3).
  4. Write a thirty-second opening check. One question answered by every person in the first five minutes (section 4).

Minimum time: about 20 minutes, plus the minute it takes to send. Done when: one question is sent, and you have written down what you will change if the answers come back at either extreme.

1. Why what they know decides your plan Understand

Two sessions on the same subject, for the same number of people, in the same room, can need different demonstrations, different practice and different checks, and the thing that decides it is what the group can already do. This is not a courtesy. It is the input that most changes the plan, and it is cheap to get.

Carnegie Mellon's Eberly Center puts it first in its list of learning principles: prior knowledge can help or hinder learning. Convention Both halves matter. What people already know is the hook that new material attaches to, and it is also, when it is wrong or when it comes from a neighboring subject, the thing that quietly deflects everything you say. Someone who has sewn with a machine for years brings habits to a new machine. Someone who has used one spreadsheet brings an idea of what a formula is. You are not writing on a blank page.

Three things change in your plan depending on the answer:

If the group isThe demonstrationThe practiceThe check
All beginnersSlower, with the steps you do without thinking spelled outFewer attempts at a smaller performanceCatches people who are lost, so it has to be answerable without jargon
MixedSame demonstration, with a named thing for the experienced to watch forPairs, deliberately made, rather than left to the roomHas to distinguish "knew it already" from "just learned it", or it tells you nothing
All experiencedShorter, aimed at the step they are most likely doing differentlyMore attempts at a harder version, or one attempt and a comparisonAbout the edge case rather than the basics
UnknownBuilt for the beginner, with a named shortcut for anyone aheadDesigned so a fast finisher has somewhere to goEarly, so you find out in minute ten rather than minute ninety

The last row is the most common and the least planned for. It is also survivable: a session built for the beginner with an escape route for the experienced works adequately for both, while a session built for the middle works for neither.

2. The expert blind spot

There is a second reason to ask, and it has nothing to do with the group. It is about you.

The University of Wisconsin-La Crosse's teaching center, in its guide on reducing cognitive load, calls the thing you need "cognitive empathy": teachers are experts and their students are novices, and the gap is hard for an expert to remember. Physical principle Skill that has become automatic stops being available to introspection. You do not experience yourself making the twelve decisions you make while shaping a loaf, because they stopped being decisions years ago. When you describe the task, you describe the four steps you still notice, and the other eight are invisible to you and load-bearing for everyone else.

That is why "what is hard about this?" is a question to ask other people rather than yourself. Three cheap ways to get the answer:

  • Ask someone who learned it recently. Not the person who learned it twenty years ago. Somebody six months in can still name the thing that confused them, because it is still fresh.
  • Ask someone to do it while you watch, and say nothing. Where their hands stop is a step you merged. This is uncomfortable and it takes ten minutes.
  • Write the steps, then do the task and count them. You will have written seven and performed fifteen. Showing someone how, so they can see it turns this into the demonstration breakdown.

The same page describes cognitive task analysis as the formal version of this: working out the knowledge and decisions an expert uses, including the ones they do not report. You do not need the formal version for a two-hour class. You need one person who learned it recently and ten minutes of their time.

3. The one question, and where to ask it

Here is where most pre-session information gathering dies. Somebody builds a survey with nine questions, sends it to thirty people, four reply, and the four who replied are the four who were going to be fine anyway.

Convention One question, sent where people are already replying to you, outperforms a form. The place matters as much as the wording: the reply to the sign-up confirmation, the group thread where the room was announced, the text message that will reach a volunteer. A link to a form is a decision to open a browser; a question in a message is a decision to type four words.

Writing the question

A good pre-session question has three properties. It is answerable in one line. It asks about what people have done rather than what they know. And its answer changes something you would otherwise do.

Weak questionWhy it failsStronger version
What is your experience level with bread?Self-rating. People who have baked twice say "beginner" and people who have baked twice say "intermediate"Have you shaped a loaf by hand before? If yes, roughly how many times?
What do you hope to get out of the session?Produces aspirations, which you cannot plan againstWhat is the thing you most recently tried and could not get right?
Rate your comfort with spreadsheets, 1 to 5A number with no shared meaning, and nobody knows what a 3 isHave you written a formula that refers to another sheet? Yes, no, or not sure
Any dietary requirements or accessibility needs?Two different questions bolted together, and the second one gets lostSeparate them, and ask the access question as its own line (section 6)

Planning the answers before you send

Write what you will do for each likely answer before you send the question. Three columns: if most say yes, if most say no, if they split. If the answer would not change anything you do, you have found a question worth deleting.

Limits

A pre-session question tells you about the people who answered, and about them on the day they answered. It does not tell you who turns up. Treat it as a hypothesis the first five minutes will confirm or overturn, which is what the opening check is for. And if nobody answers, that is information too: it usually means people have not thought about the session yet, which tells you something about where to start.

4. The thirty-second opening check

Whatever you learned in advance, open with a question every person answers at once. It costs half a minute, it corrects whatever you got wrong, and it has a second effect: the room has spoken once, in the first five minutes, which makes the second time easier.

Three forms that work:

  • Hands. "Hands up if you have done this before. Keep them up if you have done it more than five times." Two questions, one gesture, and you can see the shape of the room.
  • Stand where you are. For a group that is already standing, physical sorting reads faster than counting hands.
  • Show me. "Everyone pick up the tool the way you normally hold it." You get the real answer rather than the reported one, which matters because reported experience and actual technique come apart.

Write down what you will do at each result. If more of the room has done it before than you expected, you shorten the demonstration and add an edge case. If fewer, you cut an objective, which is why the not-today list from Deciding whether a workshop is the right format is worth having with you.

The University of Connecticut's list of 50 classroom assessment techniques, attributed on the page to Angelo and Cross, includes the Background Knowledge Probe for exactly this job. Teaching moves and checks, side by side lists it with the others and what each costs in minutes.

5. Longer probes, and prior knowledge that gets in the way Go deeper

When you have more time, or when the session is expensive enough to be worth it, there are longer instruments.

  • The Background Knowledge Probe. A short set of questions on the specific content, given before teaching, so you can see what is already there. The point is to use the result: a probe you do not read changes nothing.
  • Asking for an artifact. "Bring the last one you made" or "send me the spreadsheet that broke." This gives you real evidence of what people do rather than what they say, and it gives you worked examples for the session that belong to the people in it.
  • Talking to whoever asked you to teach. Often the person who booked you has a theory about what the group needs. It is worth hearing and worth checking, because it is frequently a theory about what the group should want.

Prior knowledge that actively hinders

The Eberly Center's principle says prior knowledge can help or hinder. The hindering case is worth planning for because it is invisible: somebody who already has a working method will hear your method as a criticism, or will map your terms onto their terms and lose the distinction that matters. Two moves help. Name the other method out loud and say where yours differs and why, rather than pretending it does not exist. And ask the experienced person to do it their way first, then yours, and compare, which converts the conflict into the comparison that teaches.

"All levels welcome"

A session advertised as "all levels welcome" is usually a session built for nobody. It is an honest thing to write when you genuinely do not know who is coming, and it is worth pairing with a line that tells people what the session assumes: "we start from a mixed bowl of dough, so if you have not made dough before, that part is done for you." That one sentence does more for the right people turning up than any amount of welcoming language.

When people's experience is known to you but not to each other

If you know that three of the twelve are experienced, the room does not. That matters when you ask them to pair up, because people pair with whoever is next to them. Making the pairs yourself, out loud, takes twenty seconds and is worth doing: "if you have done this before, find somebody who has not." It also prevents the common failure where the three experienced people sit together and finish in four minutes.

6. What to ask at sign-up about access Go deeper

The W3C's checklist for making events accessible puts one item on the organizer's list that belongs in this chapter rather than in the room chapter: ask about accessibility requirements at registration. It belongs here because the time to ask is while people are signing up, not while they are arriving.

A workable line, which you can paste into a sign-up confirmation: "If there is anything that would make it easier for you to take part, tell us by [date] and we will see what we can arrange." Asking early is what gives you time to arrange an interpreter, print large copy, or move the session to a room with a step-free route. Asking at the door gives you an apology.

What to do with the answers, what a venue is responsible for and who decides are in Access: what to ask, what to arrange, who decides and in Setting up a room people can actually work in. This manual states what federal guidance says and does not tell you whether your event is covered by it; that is a question for your venue and, where it matters, a lawyer.

7. See it See it

Fictional example

Marisol and the spreadsheet session. Marisol Vance is running a three-hour spreadsheet session for twenty-four staff in a borrowed conference room. Everyone will have their own laptop. Her manager wrote the objectives and two of them are not measurable.

Known: twenty-four people, three hours, laptops, one co-facilitator (Teo), mixed experience by reputation.

Her one question, sent as a reply-to-this line on the calendar invitation: "Before Thursday: have you ever written a formula that pulls a number from another sheet? Yes, no, or not sure. One word is fine."

What came back: nine replies out of twenty-four. Four yes, three no, two not sure.

Known from that: at least four people in the room can do the thing the second half was built around.

Assumed, and whose assumption it is: Marisol assumes the fifteen who did not reply are more like the "no" group than the "yes" group, on the theory that people who can do a thing answer a question about it faster. That is her assumption and it could be wrong; she plans the opening check so that it is corrected in minute four rather than in minute ninety.

Her opening check: "Hands up if you have written a formula that reaches another sheet. Keep them up if you have done it this month."

What she changed: she wrote one harder task for anyone whose hand stayed up, so that the four experienced people have somewhere to go while the others catch up, and she asked Teo to sit with them.

Open, and who settles it: whether the two unmeasurable objectives her manager wrote stay. Her manager settles that, and Writing learning objectives: what they will be able to do has the conversation to have.

8. Practice Practice

Minimum: write one question and send it. Time: about 20 minutes to write, a minute to send. Materials: the list of people coming, and whatever channel they already answer on.

  1. Write three candidate questions. Each one line, each about something people have done rather than what they know.
  2. Test each against the "so what" rule. For each, write what you would do differently if the answer came back yes, and if it came back no. Delete the ones where the two answers lead to the same session.
  3. Send the survivor where they are already replying to you. Not a form, unless a form is genuinely where this group lives.
  4. Write the opening check. One question, answered by everybody at once, in under thirty seconds, that would correct you if the replies misled you.
  5. Write the two changes. What you cut if the room is less experienced than you thought, and what you add if it is more. Put both in the objectives sheet of your session plan, in the "in or out" column.
Go deeper

The ten-minute version of the expert blind spot exercise: ask somebody who learned your skill in the last year to do it while you watch, and hold your tongue completely, including the encouraging noises. Write down every place their hands hesitate. Those hesitations are the steps you merged, and they go straight into the demonstration breakdown in your demonstration and practice sheets.

9. Your session Your session

10. Check your work Check your work

Criteria

  • Is your question answerable in one line, by someone reading on a phone?
  • Does it ask about something people have done, rather than asking them to rate themselves?
  • Have you written what you will do for each likely answer, including no answer?
  • Does your opening check reach every person in the room, rather than the person who speaks first?
  • Have you asked somebody who learned the skill recently what was hard about it?

Worked example

Marisol's question is one line, asks about a thing done rather than a level, and has a written consequence for each answer. Her opening check reaches all twenty-four hands at once and corrects the reply bias she knows she has. Her gap is the last criterion: she has not asked anybody who learned spreadsheets recently what was hard, and the obvious person to ask is Teo, who learned two years ago.

Common mistakes

  • A survey instead of a question. Nine questions produce four replies; one question produces nine.
  • Asking people to rate themselves. Self-ratings on a 1 to 5 scale are not comparable between two people, so the average means nothing.
  • Collecting answers and not using them. If the plan is the same whatever comes back, the question was decoration.
  • Treating the repliers as the room. The people who reply are a sample selected for being organized and interested.
  • Asking the access question too late. At the door it is an apology; three weeks out it is an arrangement.

What is still unresolved

This chapter does not tell you what to do when the room splits, which is a question about practice design and pairing: Designing the part where they do it and Asking, waiting and handling what comes back both deal with it. It does not tell you how to turn what you learned into objectives, which is Writing learning objectives: what they will be able to do. And the access answers you collect turn into arrangements in Setting up a room people can actually work in.

11. Knowledge check and scenario

Why is "rate your experience 1 to 5" a weak pre-session question?
Because the numbers are not comparable between people. Two people with identical experience will pick different numbers depending on who they compare themselves with, and confident people and cautious people answer differently about the same history. A question about something done, "have you shaped a loaf by hand, and roughly how many times", gives you a fact instead of a self-image.
What is the expert blind spot, and why does asking yourself not fix it?
Skill that has become automatic is no longer available to introspection, so when you list the steps you list the ones you still notice. UW-La Crosse's guide calls the needed quality cognitive empathy and says the gap is hard for an expert to remember. Asking yourself returns the same incomplete list, because the list is produced by the same automatic process. Watching somebody else do it, or asking somebody who learned recently, returns the missing steps.
Nine of twenty-four people reply to your question. What do you actually know?
You know about nine people on the day they replied, and you know that those nine are a sample selected for organization and interest. You do not know about the other fifteen, and assuming they resemble the repliers is a guess. The honest response is to treat the replies as a hypothesis and to build an opening check that corrects it in the first five minutes.
(Go deeper) Why make the pairs yourself instead of letting the room pair up?
People pair with whoever is beside them, which usually means the experienced sit together and finish early while two beginners work slowly with nobody to ask. If you know the spread and the room does not, twenty seconds of instruction puts that knowledge to work: ask anyone who has done it before to find somebody who has not.
You sent one question to twelve people and got two replies, both saying they have not done the thing. The session is Thursday. What is the best use of the next ten minutes?
A reminder might get you one more reply, and it costs you goodwill and ten minutes you could spend on the plan. Two replies out of twelve is a weak signal, and a third reply would not make it a strong one. The information you want is available for free in the first five minutes of the session.
This is the robust plan. A session built for the beginner with a named escape route for anyone ahead works adequately for both, and the opening check corrects you at minute four, when you can still act, rather than at the end. Write the harder task now, because you will not invent one in the room.
A reasonable hunch, and it points the wrong way. People who can do a thing tend to answer a question about it quickly, so silence is weak evidence of inexperience rather than of experience. Even if it were a coin toss, planning for the beginner carries less risk than planning for the expert.

Checklist

Records your learning only; it is not a qualification.