Switch Interview

Ask why someone uses a product and what usually comes back is a reason assembled on the spot. The interface is clean. Collaboration is easy. Nobody is lying — explaining a decision made months ago in terms of today's satisfaction is what people naturally do. The trouble is that those answers don't tell you what to build next.

Pagedeck collected a list like that once. Sofia asked eight recently signed customers why they picked Pagedeck, and eight of them handed back nearly the same sentence. Grace then asked why, after three years on their old tool, they had switched last month specifically. That question produced something else entirely: in a quarterly review someone had put up an outdated file and read a wrong number out loud. That answer came with a date, a room and a consequence attached.

A switch interview works that difference on purpose. Instead of asking why someone is satisfied now, it asks when and because of what they moved. You get a person whose memory is still intact to walk backward through time, which restores the order the decision actually happened in.

Start by finding people who switched recently. Accounts that paid or rolled out within the last 90 days are the baseline. Past that, the scene of the day is gone and only a smooth reason is left.

There are six steps.

1. Pick the participants. Switched within 90 days, and involved in the decision. A user who joined after the rollout wasn't in the room.
2. Set the starting point at the day they decided and walk backward. Ask what happened the morning they paid, then the week before that, then the month before that.
3. Fill in four forces: the friction pushing them off what they had, the expectation pulling them toward the new thing, the habit keeping them where they were, and the anxiety about moving and getting it wrong. The first two can both be strong and the person still won't move. A demo works on the pulling force and leaves the other two untouched.
4. Dig at the hesitation. How many days the decision sat, what they had to verify before paying. This is where the interview earns its keep.
5. Write down what they compared it against. Competing products aren't the only entries. A spreadsheet, or doing nothing at all, turns up as the real rival often enough.
6. Sort what you heard into three layers: the functional job, the task that has to get finished; the emotional job, what they want to feel while doing it; and the social job, how they want to be seen by others. Bundling those into one sentence gives you a job story.

There's one trick in how you ask. Ask for scenes, not opinions. Rather than what feature they needed, ask who was looking at that screen with them that day, and who they first said "we have to change this" to. A question anchored in a time, a place and a person is hard to answer with an invention.

Use it in a market where people have already switched. If there's an existing way to get the job done and people move between options, there are scenes left to recover. It's the first thing to reach for when deals aren't growing and nobody knows why, especially when demos go well and the decision stalls at payment. What's blocking it is often not a feature but anxiety or habit, and those two surface only in this kind of conversation.

Three situations where it doesn't fit.

1. A brand-new category nobody uses yet. With no one who switched, there's no moment to walk back through.
2. Products where the decision has no sharp moment. If people arrive free and pay months later, they themselves don't know when they made up their minds.
3. When you want a proportion. Eight conversations can't tell you how common something is across everyone.

Two limits come with it. The first is that memory gets rebuilt. Dates, the emails exchanged that week, the document they were working in — put those on the table alongside the question and the scene snaps back into place. The second is that you only meet people who switched. Eight stories from people who moved contain nothing about those who looked and stayed. Mixing in two or three who took a quote and didn't move makes the set better. A deal that stalled is usually the clearest description of what the anxiety side is missing.

"What happened the morning you paid? Start with who you talked to about it that day."

Trying to restore the scene the decision happened in rather than current satisfaction.

"The hesitation matters more right now than what they liked. That's where conversion stops."

Praise doesn't change the next decision; the part where someone stalled does.

"What this customer compared us against wasn't a competitor. It was a spreadsheet."

Which changes what the value has to be explained against.

Mia and Sofia went back to the eight customers signed most recently. The pushing force came out in the same shape at five of the eight: putting up an old file in a meeting because nobody knew which version was current. The pulling force was the expectation of editing together on one screen, which was already known.

What was new sat on the anxiety side. All eight said they had asked, before paying, how many days it would take to move three years of documents and whether anything would be lost along the way. So what got built that quarter was not a new editing feature but an import tool that pulls existing documents over wholesale, and a process where Sofia's team performs the first migration for the customer. The share of demos that reached payment rose noticeably that quarter.

The most common reading is that this is another name for a customer interview. Sitting with one person for an hour looks the same from outside. What differs is who and when: this one meets only people who switched recently, and it asks about one decision rather than the whole experience of the product. The 90-day window matters as much as the questions do. Erase that difference and the questions drift back into the usual shape, leaving what caused the decision unrecorded.

Reading it as a place to collect compliments is the other one. It's an interview about why someone chose you, so praise feels like the expected output. But praise confirms what's already built and can't change next quarter's list. Hesitation can. If eight people tried to verify the same thing before paying, everyone who hasn't paid is standing at that spot right now.