By explicitly separating the two: name the specific low-value friction point you'd remove, and separately name the safeguard or control you'd keep. That's what makes a recommendation read as a considered trade-off, not a blanket "remove all friction" instinct.
Why this is a judgment move, not just vocabulary
This is a genuine business-judgment move, distinct from simply knowing the friction/safeguard vocabulary: given a process with both a low-value friction point and a protective control, the strongest recommendation names both pieces separately rather than treating "reduce friction" as an unqualified goal. Knowing that friction and safeguard are different words is necessary but not enough -- the actual skill is being able to look at a real process and correctly sort its steps into "this one isn't earning its cost" and "this one is."
The shape of the recommendation
The shape of the move is consistent: identify the friction that isn't earning its cost, propose removing exactly that, and explicitly state what protective mechanism stays in place and why -- so a reader can't mistake the recommendation for "remove everything that slows this down."
"We can drop the second confirmation email in the signup flow -- it's not preventing anything and it's just adding friction. The identity-verification step stays; that's the safeguard actually catching fraudulent signups."
"Cutting the redundant address-re-entry field would remove friction with no real cost. The manager sign-off on refunds over $500 should stay -- that's the control that's actually catching pricing errors."
"Recommend removing the duplicate manager approval in the low-risk expense tier -- pure friction with no protective value there -- while keeping it in place for the high-risk tier, where it's genuinely catching errors."
Want to learn "Friction" in depth?
Lyra Practice teaches advanced non-native professionals the nuance of high-value expressions like this one, then has you practice using them in realistic work scenarios.
Start learning for free →Notice the pattern across all three
Each example does three things in sequence: names the specific low-value friction, proposes removing exactly that (not "reducing friction" broadly), and names the specific safeguard that stays and what it protects against. That third piece is the one most often skipped -- and it's the piece that turns a vague efficiency suggestion into a recommendation a risk-aware reader can actually trust.
The mistake to avoid
The mistake to avoid is a recommendation that only argues for removing friction without ever naming what stays and why -- that reads as "remove all friction is always good," which is a risk, not a defensible business position. A recommendation that only lists what to cut, with no mention of what stays protected, invites the reasonable question "did you consider what this step was actually for?" Answering that question up front, in the same recommendation, is what makes the trade-off explicit instead of implied.
Why this move is worth practicing deliberately
This move is worth practicing on purpose, because the two habits it fights are both natural defaults. Left unchecked, most people either argue for removing everything that slows a process down, or defend every existing step out of caution, without weighing each one individually. The recommendation format above forces a middle path: look at each friction point on its own evidence, decide whether it's earning its cost, and say so explicitly either way. That's a more defensible position in front of a skeptical stakeholder than either extreme, because it shows the reasoning, not just the conclusion.
Practice scenarios
Practice using friction in situations like:
- recommending removal of one specific low-value friction point
- explicitly naming the safeguard or control that stays in place, and why
- avoiding a recommendation that reads as 'remove all friction' with no trade-off named
Useful practice phrases:
- "This step is pure friction with no protective value -- recommend removing it."
- "The verification step stays; that's the safeguard catching fraud."
- "Remove the low-value friction here, but keep the control that's catching real errors there."
A recommendation that names both what to cut and what to protect reads as judgment. One that only names what to cut reads as a shortcut.
Lyra Practice helps advanced non-native English professionals learn the nuance of high-value workplace expressions and practice using them in realistic scenarios, so their English sounds natural, precise, and senior at work. Try Lyra Practice.