Comment-to-DM Keyword Planner

Because matching is a string test, not a reading test. Inrō documents that keyword matching ignores case and accents and, on its default, looks for your word anywhere in the comment. That is generous in the right direction and dangerous in the wrong one. If it already sits inside sentences your audience writes, ordinary comments fire the automation and people who asked for nothing receive a DM. If it shares a run of letters with another live keyword, a comment meant for that campaign fires yours as well. Neither failure has anything to do with spelling, and neither shows up on a list of recommended words.

Yes, and it is the normal case. Inrō's product page states that you can configure multiple keywords on one post, each sending a different DM. The thing to watch is what happens when one comment matches more than one. Inrō documents that scenarios which are not in a priority group all run independently, so both fire and the person receives two messages. Inside a priority group, only the highest-priority matching scenario runs. The fix is straightforward once you know the rule exists: pick words that share no run of letters, or put the overlapping scenarios in one priority group with the most specific flow at the top.

Only when you want the comment to be the keyword and nothing else. Inrō documents strict match as firing only when the message is the keyword with no extra text. That removes false triggers, because a sentence containing your word no longer counts. It also removes everyone who typed the word with a full stop, an emoji or a greeting attached, which on a busy post is a large share of the people trying to reach you. The usual answer is to leave matching on its default and spend the effort choosing a word ordinary comments do not contain. Strict match suits a qualification step, not a lead magnet.

Inrō supports emoji triggers alongside keywords, but read the constraint before you print one in a caption. Inrō documents that some triggers fire only when the message is nothing but emojis, and gives the example that hi followed by an emoji does not match. People rarely comment an emoji on its own. They wrap it in a word, a name, or three more emoji. Underneath that sits a second problem: an emoji can carry a skin tone modifier or a variation selector, which are extra characters, so the version your audience sends is not always the one you set. A plain word is the safer instruction.

No. Inrō documents that keyword matching ignores case and accents, and gives cafe matching cafe with an accent as its example. Printing the word in capitals is therefore a readability choice, not a matching requirement, and someone who types it in lower case still receives the DM. What does not fold is styled text. The decorative letters people paste from font generators are separate Unicode characters, not styled versions of ordinary ones, so a word written in them is a different string. Punctuation stuck to the word is a smaller version of the same problem, which is why the planner rejects anything but plain letters and digits.

Rarely, and the reason is mechanical. Inrō lets you attach more than one keyword to a trigger and choose OR, where any keyword is enough, or AND, where all must be present. A two-word phrase behaves like AND: both strings have to turn up in the comment before anything sends. Every extra word is another chance for a typo or an autocorrect to break the match, and the person never finds out why nothing arrived. The exception is a phrase so distinctive that no part of it is safe alone. If you are reaching for one because your short word keeps clashing, change the word instead.

Two separate rules apply. Meta's private replies documentation states that only one message can be sent to the commenter, so a single comment buys a single reply and the follow-up waits until they answer. Inrō adds its own guard and publishes it: one DM per user per post by default, so someone who comments five times is not messaged five times. Age is the part people forget. Meta allows the reply within seven days of the comment, and that clock starts at each new comment rather than at the post. An old post therefore keeps delivering whatever its keyword is wired to until you switch it off.

We read six of them in August 2026. Every page recommends words, none sources the performance figures printed beside them, and several recommend the words most likely to misfire: one names GUIDE, LINK, YES, PRICE and FREE as choices that work well, which are exactly the words people already type in ordinary comments. Only one of the six, Wati's, named the contains and exact matching options. That same page defines contains as triggering if the keyword appears anywhere in the message, then advises that DROP and DROPS are different enough, which its own definition contradicts: a comment reading DROPS contains DROP. None of the six tested a word against the reader's own comments.

It is free, there is no sign-up, and nothing you type leaves your browser. The six checks run in a script on this page, so the candidate word, the comments you paste and your keyword list are never sent to a server and are gone when you close the tab. That matters more here than on most tools, because those lists describe your account in a way you would not publish. Run as many words as you like, with no cap. The one thing this page cannot do is set the trigger, since that needs a connected Instagram professional account. The free Inrō plan covers 100 contacts and 3 automations.