Instagram comment-to-DM for digital product sellers

The mechanic is the same three steps every campaign runs: comment, public reply, DM. What's specific to a product catalog is what sits behind it. The DM text you write goes out exactly as written to everyone who matches that keyword, so a code like SAVE20 typed straight into that message reaches every commenter with the same identical wording a form letter would use, just triggered by that person's own comment instead of scheduled in advance.

The file that follows the code is a sample, not the full product: a cropped preview, an excerpt, a lower-resolution version, whatever gives someone enough to decide before they buy. The link in the DM always points to whatever's attached to that campaign, so the sample and the code for a given product live in the one place: the campaign for that product.

A catalog of five presets, templates, or packs really needs five campaigns if each one has its own sample and its own code, since a campaign delivers one file. Naming the keyword after the product itself, PRESET1, SKU shorthand, the product's actual name, makes the Leads list and any exported data mean something later: a row that says which product someone asked for, not just that they commented something.

Worth knowing before it becomes a real problem: if two live campaigns could both technically match the same comment, only one fires, the older campaign wins, not both and not a random pick. Reusing the same keyword across two product campaigns, even by accident, means the newer one effectively never fires on that word, and there's no visible error, the comment just gets answered by the other product's campaign.

The 25 MB, PDF-only constraint applies here the same as anywhere else. For a written product, a template pack, a workbook, a swipe file, that's rarely a real limit: a sample chapter or a stripped-down version of the template fits comfortably. For something that isn't naturally a PDF, a preset pack, an audio loop set, the workaround already covered elsewhere applies: a one-page PDF describing the sample with a link out to where it actually lives.

The code goes in the DM text, not the file. That keeps it simple to change: editing the campaign's DM message updates the code for everyone who comments after that point, without touching the sample file at all. The trade-off worth naming plainly: the code is the same for every person who comments that keyword, since the message is fixed text on the campaign, not something generated per lead.

Set each product up as its own campaign: its own post, its own keyword, its own sample file, its own DM text with that product's code. Name the campaign after the product, not the generic offer, since that name is what shows up later in Leads, in a CSV, and in the tag an ESP sync applies.

The one thing worth planning around before launching several at once: engagement quota is shared across the whole workspace, not split per campaign. Five product campaigns each getting steady traffic all draw down the same monthly allowance, so a catalog launch across several posts in the same week can use up a plan's headroom faster than running one campaign alone would suggest.

Because each product is its own campaign, comparing matched-comment counts across campaigns is a direct read on which product the audience is actually asking about, no extra tagging required beyond naming the campaigns clearly in the first place. That's a real answer to "which of these five things should I make more of," sitting in data that already exists.

Repeat interest shows up the same way, just manually: the Leads list carries a username with every row, so someone who commented on both the preset pack and the template pack campaigns appears twice, once per product, both visible under the same username. Nothing bundles those two rows into a single "repeat customer" flag today, but the raw signal is there to look for. On Pro and Agency, an ESP sync tags each contact with the campaign it came from, so that same pattern is visible inside Mailchimp or Kit too, useful once there's an actual follow-up sequence to send.

Sharing a keyword across products is the most common one, covered above: it doesn't fail loudly, it just means one campaign quietly wins every match and the others never fire, so a product might look like it's getting zero interest when the comments were actually landing on a different campaign entirely.

Attaching the full product instead of a sample defeats the point of a preview. If the DM delivers exactly what someone would otherwise pay for, there's no reason left for them to buy, and "sample" stops meaning anything to future campaigns too.

The quota mistake is the quiet one: assuming five separate product campaigns means five separate monthly allowances. They don't. A workspace that's about to run several product launches in the same week is worth checking against the plan's total quota beforehand, not discovering mid-week that the fifth campaign is queuing instead of sending.