What is journey mapping, and what do you get out of it

On-Off Group

October 2026

Journey mapping is a session where you lay out every step a customer takes to get something done, using real evidence, and mark where they stop or wait. Done properly it produces a step-by-step map, the drop-off and turnaround time at each step, and a shortlist of fixes.

If someone has asked you to run a journey mapping workshop and you have to explain it to your boss first, this is the plain version. What journey mapping is, what comes out of it, and what to insist on so you do not end up with photos of a wall covered in sticky notes.

It is also for anyone who has sat in one of these sessions and then got asked, did anything actually change? And had no answer.

What is journey mapping

Journey mapping is a session where you lay out every step a customer takes to get one thing done, in the order they do it. Their steps, in their order, not your internal process. For a bank that might be: sees the ad, opens the form, uploads an ID, waits, gets a text, visits a branch, activates the card.

Under each step you write what the customer is trying to do, what they actually have to do, how long it takes, and what goes wrong. Then you mark the steps where people stop, wait or call.

The map covers the whole path, including the parts your team does not own, which is usually where the waiting happens. That is the main reason it is worth doing: the worst step in a customer journey is very often between two departments, so nobody has been looking at it.

Where these sessions usually go wrong

The session runs well. Plenty of sticky notes, good energy, a photo of the wall in the chat group afterwards. Someone types it up into a PowerPoint. Later on, someone asks whether the onboarding work paid off, and there is no honest answer, because nothing was ever counted.

We see this a lot. The map describes the journey accurately and still cannot be used to make a decision, because every step looks the same size. Nobody knows which step loses the most people or takes the longest.

The fix is small. Before you close the session, put a number against the steps you think are the worst: drop-off, turnaround time, calls per hundred applications, rework rate. Even a rough number from last month's report changes the conversation, because now the steps are ranked. We wrote more about that here: Instead of sticky notes on a wall, get a drop-off number out of the session.

What the session should produce

Four things, and you can check them off on the day.

A step-by-step map of the current journey, on one page, with the steps named the way a customer would describe them.
A number against each of the worst three or four steps. Turnaround time, drop-off percentage, volume of calls, whatever you can get.
A list of the moments that cause the pain, with evidence next to each one. A quote, a screenshot, a line from the call logs.
A shortlist of fixes, each with an owner and a rough estimate of the work involved.

What it should not produce is a future-state map of a journey nobody has tested. They are easy to draw and nobody can argue with them, so they sit in the slides and nothing happens.

If a step has no number and no evidence, mark it on the map as a guess. They are your research list.

Where the evidence comes from

If the map is built from memory, it only shows what your team believes is happening. That is a fine starting point, but you need something real to check it against.

The cheapest sources are usually already in the building: system data on how many people start and finish each step, turnaround times from operations, complaint and call reasons, and branch staff. Ask branch staff which step of the form customers cannot finish on their own. They know.

After that, watch people. Five customers on video trying the current journey will settle arguments the team keeps having. You do not need a large sample to find out that the ID upload fails on older phones.

If you would rather do this as an ongoing thing than a one-off, the trade-offs are set out in One-off customer research or monthly insight: which suits you.

What happens after the map

The map on its own changes nothing. Pick one journey and ship a change to it.

On a series of five-day design sprints with a large insurer, customer-facing web journeys that had previously taken six months or more to get out were designed and tested with customers in two weeks, and live in four. Twenty-six weeks down to four. Completion rates on the new journeys went up by 80 per cent and the old drop-off points disappeared. That was measured by the client, not by us, which is only possible because the before numbers existed.

The temptation after a good mapping session is to try to fix every journey at once in one big project. That is how the work stalls. Take the one journey with the worst number, fix it, measure the same number again, then widen. We set out that approach in Instead of a platform project: ship one customer journey, measure it, then widen.

One more thing, because it keeps coming up. If the plan is to put an AI assistant on a journey that already loses half the people at the ID step, you will automate the loss. Fix the customer journey before you put AI on it.

What to do this week

Pick one journey with a business owner who cares about it. Onboarding, claims, a card application.
Pull the numbers that already exist for that journey: starts, completions, turnaround time, call volumes. Write them down even if they are messy.
Ask two branch staff which step customers cannot finish without help.
Book the mapping session, and tell the room the output is a ranked list of steps with numbers, not a wall of sticky notes.
Agree one number you will report again after the fix, and write it in the invite.

Questions people ask

What do you actually get at the end of a journey mapping session?
A map of the steps a customer takes, a number against the worst steps such as drop-off or turnaround time, and a short list of fixes with an owner against each one.

Do we need customer research before we map?
You need something real: call logs, system data, branch staff notes, or a handful of recordings of people using the current flow. Mapping from memory alone produces a map of your assumptions.

Is journey mapping the same as usability testing?
No. Mapping shows the whole path end to end and where it breaks. Usability testing shows you one screen or flow in detail, with a person struggling in front of you.

Should we add AI to the journey while we are at it?
Fix the broken steps first. Putting AI on a journey that already loses people means you process the same failure faster and at more volume.

On-Off Group trains teams, tests products with real customers, finds where a transformation has stalled and builds what gets it moving, for banks, insurers and enterprises in the Philippines, since 2015. How we help with customer journey mapping.