• What is Microcopy?

Microcopy

Microcopy refers to the small bits of text woven into a user interface or product - button labels, form hints, error messages, empty-state text, tooltip explanations, onboarding nudges, confirmation notices, 404 pages, the “no results” line, the “saved” indicator. The writing that nobody calls “content” but that quietly decides whether someone completes a task, trusts a product, or gives up.

The term gained popularity through Joshua Porter’s 2009 blog post (“The Greatest Button”) and Kinneret Yifrah’s 2017 book Microcopy: The Complete Guide. It’s now a recognised discipline in product and UX writing teams - though most B2B SaaS still treats it as an afterthought.

Where microcopy actually lives

Eight places worth naming, because they’re where microcopy tends to either help or quietly hurt:

Button labels. “Save draft” vs “Save” - one tells the user what’s about to happen, the other leaves them guessing. Clarity over cleverness, every time.

Form fields. Labels, placeholder text, inline hints, error messages. Forms are where microcopy has the most direct conversion impact. An error message that says “that email is already registered - try signing in?” converts; “invalid input” annoys.

Empty states. The page before the user has done anything. Most empty states are either blank or patronising. A good one guides toward the first useful action.

Loading and transition states. The moment between click and result. Longer than two seconds and the user needs a cue - “saving your draft,” “processing payment,” not just a spinner.

Confirmation and success messages. “Saved” is fine. “Draft saved - we’ll bring this back next time you log in” builds confidence.

Error and failure states. The single biggest microcopy blind spot in most products. An error message should name the problem, suggest a next action, and avoid blaming the user.

Onboarding tooltips. The single-sentence hint that explains what a feature is. Most are either missing or written like a Microsoft 2003 help file.

404 and edge pages. Often the last chance to recover a user. A good 404 redirects attention toward something useful.

The traits of good microcopy

Written from the user’s perspective. “We couldn’t save your draft” (product-centric) vs “Your draft didn’t save - check your connection?” (user-centric). The second feels like help. The first feels like a shrug.

Actionable. “Something went wrong” is useless. “Something went wrong - try again in a moment, or email support@” is useful.

In the brand’s voice. Mailchimp is playful, Stripe is precise, Notion is unpretentious. Microcopy is where voice lives or dies at the product level.

Short. Not aggressively short - readable short. The test: can a scanning user understand it in under a second? If not, it’s too long.

Where microcopy tends to fail

Written by engineers under time pressure. Not a dig - engineers shouldn’t be writing final microcopy any more than a writer should be writing the auth flow. Both happen in under-resourced teams, and the results are visible.

Translated from generic templates. The “Your session has expired” language that every SaaS shares. Replacing it with something specific and branded takes two minutes and improves every session for every user.

Inconsistent across the product. The save button says “Save” in one place, “Submit” in another, “Done” in a third. Each was a reasonable choice; together they’re noise.

Never tested. Microcopy is often treated as too small to A/B test. The classic Jared Spool example of “Register” vs “Continue” adding $300M to a revenue line is 16 years old and still gets cited because the pattern keeps repeating.

Microcopy sits outside Penfriend’s lane - on purpose

Penfriend writes long-form content - blog posts, category pages, glossary entries, comparison pages. We don’t write microcopy. The UX writer at your company who knows the product intimately, the edge cases, and your voice rules is going to out-perform an LLM on the button-label problem every time, because microcopy is all context and product knowledge.

What we do see in customer teams: writers who also own microcopy get more time for it when we take the long-form load off their plate. The 20 hours a week they used to spend drafting blog posts becomes 20 hours they can spend auditing error states, rewriting form hints, and getting the onboarding flow right. Indirect contribution, but real.

An example

A B2B SaaS team was seeing a 34% drop-off between the first screen of their signup flow and the second. The form asked for five fields. One was “company size” as a select dropdown with no default.

They rewrote three pieces of microcopy. The field label changed from “Company size” to “How many people work at your company?” The dropdown default became “Just me” instead of “- Select -”. The error message on blank submit changed from “This field is required” to “Pick whichever is closest - you can change it later.”

Drop-off on that screen fell to 11% over two weeks. Same form, same fields, three sentences rewritten. The team’s PM had never considered microcopy a conversion lever. It turns out to be one of the highest-impact places to edit text, because every user reads it and most never read a blog post.

Related terms