One of the advantages of building a writing tool that works across real writing contexts is that we see a lot of actual text. Not cleaned-up examples in textbooks or edited prose in published writing. We see emails drafted under time pressure, Slack messages that grew into paragraphs, Google Docs where a non-native speaker is trying to match the register of a document started by a native speaker on the same team.
After enough time looking at that kind of text, some patterns become very familiar. These are not random errors. They are consistent, predictable, and largely explainable by the structure of the languages these writers are coming from. Here is what we see most often.
Article confusion: a, an, the, and nothing
This is the single most common pattern across speakers of languages that do not use articles as English does. Russian, Polish, Chinese, Japanese, Korean, and Arabic all lack English-style definite and indefinite articles, or use them in ways that do not map cleanly onto English rules.
The results are sentences like "I spoke with manager about situation" or "Please review the attached document before meeting." Both are readable. Both carry small signals that English articles are not yet intuitive for the writer. In professional context, native readers do notice them, even if they rarely flag them.
What makes this hard to catch automatically is that the rules governing English articles are genuinely complex. The full rule involves countability, definiteness, specificity, and generic reference, and there are legitimate exceptions at every level. We can catch many article errors with confidence; we are much more careful about flagging ambiguous cases where the correct choice depends on context we cannot access.
Preposition errors
Prepositions in English are notoriously idiomatic. "Interested in" not "interested on." "Discuss the issue" not "discuss about the issue." "Agree with" and "agree on" are both correct but in different contexts. The patterns cannot be derived from logical rules. They are largely a matter of exposure.
Non-native writers often apply the preposition logic of their first language, which works for some structures and fails for others. Spanish speakers may write "depend of the outcome" (depend on). German speakers may write "invest on the project" (invest in). Mandarin speakers may omit prepositions in contexts where English requires them.
We find preposition errors more tractable than article errors for one reason: the incorrect forms are usually clearly wrong in a way that confident flagging is possible. "Discuss about" is an error in formal professional writing without significant ambiguity about context.
Subject-verb agreement at distance
This category appears most often in complex sentences with a long intervening clause between the subject and the verb. "The list of deliverables that were included in the original proposal have been updated" is a common pattern. The grammatical subject is "list" (singular), but the plural noun "deliverables" is closer to the verb and pulls it toward plural agreement.
Native speakers make this error too, particularly in speech and in first drafts. For non-native speakers, the pull toward proximity agreement is often stronger because the structural rule about subject-verb agreement at distance requires parsing the sentence's deep structure rather than just reading across it.
Tense and aspect
English has a more complex aspectual system than many languages, and the distinction between simple past and present perfect causes consistent confusion. "I have sent the file yesterday" mixes a present perfect form with a definite past time marker, which English does not permit. Simple past is required: "I sent the file yesterday."
The reverse error is also common: using simple past where present perfect is appropriate. "Did you see the email I sent?" when asking about a recent email that the recipient may not have opened yet sounds slightly off to native ears. "Have you seen the email I sent?" is the more natural phrasing for this context. The difference is subtle enough that many non-native speakers never encounter it corrected in their working lives.
Overly formal register
This one is not strictly a grammar error but it is one of the most consistent patterns we see among non-native speakers who learned English in formal educational settings. The register of their writing is calibrated to textbook English, which tends to be considerably more formal than contemporary professional writing in the US and UK.
"Kindly be informed that the meeting has been rescheduled" is a sentence structure that native speakers in most Western professional contexts no longer use, but that appears frequently in the writing of non-native speakers who learned English from materials that modeled this register. "The meeting has been rescheduled to 3pm on Thursday. Please let me know if that works for you" is how a native speaker in the same context would typically phrase it.
Register is not a simple binary. There are professional contexts where formal register is expected and appropriate. But for most workplace email, the gap between textbook English and current professional norms is significant enough to affect how the writer is perceived.
Gerund versus infinitive
English uses both gerunds (verb-ing) and infinitives (to-verb) after verbs, and the pattern for which one to use is mostly arbitrary from the perspective of non-native learners. "Enjoy doing" not "enjoy to do." "Plan to do" not "plan doing." "Stop talking" and "stop to talk" are both correct but mean different things.
The errors here tend to come in two flavors: wrong choice of form ("I suggest to review the proposal before Friday") and meaning-shifting error where the writer chooses the right form for the wrong meaning ("we stopped working on the feature" means completion; "we stopped to work on the feature" means we paused to do it).
What actually helps
We are not suggesting that non-native speakers writing English in professional contexts need to eliminate all traces of L1 interference from their writing. That is not a realistic goal, and a certain amount of stylistic variation is normal and unproblematic in most professional contexts.
What does help is targeted feedback on the patterns that consistently affect reader comprehension or that carry strong signals of non-native writing to native readers in formal professional contexts. Article errors and preposition errors are in this category. Register mismatches are in this category when the context makes formality expectations clear.
The goal of Linguix's work in this space is not to produce text that reads as native, which is neither necessary nor always desirable. The goal is to remove the friction points that distract readers from the content and to help writers say precisely what they intend to say, which is a challenge that belongs to everyone who writes under professional constraints.