A broken process can survive for years in plain sight. Everyone complains about it. Everyone knows where time gets wasted. The workaround has workarounds.
The team never changes the process because they’ve invested time in building competence around the painful thing—sunk cost bias at its best. What would they give up by switching? What would make one real workflow, with a rollback, feel safe to try? Frustration pushes, but the effort already spent pulls back harder.
Pain is only one force
From the outside, keeping the tool looks irrational. From inside the decision, staying may be perfectly sensible. The current system is slow, but people know how to recover when it fails. The team’s shared spreadsheet is brittle, but its logic is visible. The manual step is tedious, but one trusted person owns it. The old tool is ugly, but switching won’t threaten anyone’s role or expose how messy the process has become.
The four forces are Push, Pull, Habit, and Anxiety. Push and Pull move people from the status quo—the current workflow—toward a brighter future. Habit and Anxiety hold them in the status quo. Frustration describes the pain in the current workflow that creates Push; attraction describes the brighter future that creates Pull. Anxiety is mostly about the perceived pain of adopting something new. The workflow people already run is the real competitor, and it competes with more than pain.
Fill-in table
Backward forces
Name the Habit and Anxiety holding people in the status quo before you add more Pull.
| Force | What it sounds like | Where it shows up | What weakens it |
|---|---|---|---|
| Habit | “We’ve already set everything up.” “It would be a huge lift to change.” “We don’t have time to retrain everyone.” | Competence built around the current method: which fields can be ignored, who to ask, how to repair the export | Familiar language and objects kept around the new behavior, so nobody gives up a known recovery path to try it; their data imported, so nobody starts over |
| Anxiety | “What if it breaks?” “What if I can’t undo it?” “What if I can’t explain it to my boss?” | Unanswered questions about migration, made-up output, edge cases, accountability, adoption | A path that’s specific, bounded, and reversible, with a real win before the real commitment |
Habit is more than repetition
People know which fields can be ignored, who to ask, how to repair the export, when the number is probably wrong, and what leaders expect to see. That knowledge may never appear in documentation, but it makes the system survivable.
Workarounds become infrastructure: spreadsheets that feed the tool, naming rules everyone knows but nobody wrote down. Switching means rebuilding that invisible scaffolding.
A replacement asks users to surrender their competence and become beginners again. Nobody says “what if I look like a beginner again?” out loud, but it’s one of the strongest reasons people stay. A product promising simplicity can still threaten identity: “I’m the person who knows how this works” becomes “I’m the person who chose a tool nobody understands.”
The switch has social and emotional costs that a feature comparison can’t show.
Anxiety fills the gaps in your promise
Every unanswered question becomes a reason to stay. Will our data migrate cleanly? Will the AI make something up? What happens when the edge case hits? Can we get the work back out? Who’s accountable? Will the team actually adopt it? Are we going to spend a month configuring this and discover the demo was staged?
Marketing often responds by making the future sound even bigger. That can increase anxiety. The more transformative the promise, the more change the buyer has to imagine absorbing. Anxiety doesn’t need more promises. It needs a safer switch: a path that’s specific, bounded, and reversible.
A small, reversible first step reduces the cost of movement
Let people test the product with a real slice of their work. Preserve familiar language and objects. Show exactly what will change and what will stay. Provide a rollback path, so the work can come back out: an export that makes leaving possible, so entering feels safe. Demonstrate how the product handles the ugly case, not only the clean demo.
Make the first commitment small. A “no-brainer” offer isn’t cheap or flashy. It’s aligned with the buyer’s current desperation and low enough in risk that saying yes doesn’t feel reckless. The first step might be one imported account, one workflow, one team, one week, one reversible action. The user gets evidence without staking the entire operation.
The old way does something well, and the replacement has to keep it
The status quo usually contains hidden value. The replacement should carry that value forward or deliberately provide a better version of it.
Fill-in table
What the old way does well
List the reassurance the current method provides before you remove its friction.
| Old way | Hidden value the replacement must carry forward |
|---|---|
| Manual review | Accountability |
| Shared spreadsheet | Common visibility |
| Weekly meeting | Synchronized interpretation |
| An audit trail |
If you remove friction and accidentally remove reassurance, people will recreate the old process beside the new product. The team’s brittle spreadsheet is familiar because people know how to repair the export.
Example
A backward-forces audit for a team’s brittle spreadsheet
- Competence: people know how to repair the export and when the number is probably wrong.
- Responsibilities: one trusted person owns the manual step.
- Familiar failures: the team’s spreadsheet is brittle, but its logic is visible, so people know how to recover when it breaks.
- New uncertainty: will the data migrate cleanly, and will the AI make something up?
- Irreversible part: spending a month configuring the new product and finding the demo was staged.
- Proof before commitment: one workflow, one team, one week, with a rollback path.
- Property that must survive: common visibility.
A backward-forces audit
The audit identifies the competence and reassurance the new product must preserve, plus the uncertainty it must reduce before people can move.
For the behavior you want to replace:
0 of 7 done
