<- all tokdocs
2026.10.06@thejimmyreilly

Instagram's Own Insights Panel Shows the "9,500 Leads" Are the Comment Count, and Meta's API Caps Each Comment at One DM

Marketing
Watch on TikTok

View on TikTok ->

This is a 64-second vertical video (ffprobe reports 64.938005 s), 1080x1920, SDR, HEVC/H.265 video at a constant 30/1 fps with a 458,191 bps video stream, AAC audio at 44,100 Hz stereo and 48,125 bps, 513,086 bps overall container bitrate, 4,164,854 bytes in an MP4 container, served by TikTok as format bytevc1_1080p_513078-1; the fps field in the yt-dlp metadata is null, so the 30 fps figure comes from ffprobe on the downloaded file. It was posted 2026-10-06 at 23:00:00 UTC by uploader handle thejimmyreilly, channel nickname "Jimmy Reilly," over the audio track "original sound - Jimmy Reilly." The metadata was captured 2026-10-07 at 05:32:53 UTC, 6 hours and 33 minutes after posting and inside the first day, at which point the post had 612 views, 20 likes, 0 comments, 5 reposts, and 31 saves. I read all 32 frames in tokdoc_frames/ (sampled at 2-second intervals, so frame N sits at (N-1)x2 seconds) and all 206 words of transcript.txt. The frames show one setting throughout: a man with a red beard, round tortoiseshell glasses, a single stud earring, and a wedding band, in a heather-grey crew-neck t-shirt, seated at a wooden desk behind a black broadcast microphone on a boom arm, against a plain grey backdrop with a soft key light falling from camera left. Frames 1 and 2 carry a black-on-white burned-in title card reading "Claude Wrote A Reel That Pulled 9,500 Leads." Frame 3 begins word-by-word white captions ("Check"). Frames 4 through 7 overlay a dark Instagram Insights panel whose icon row reads, left to right, heart 6.2K, speech bubble 9.5K, repost arrows 93, paper plane 4.2K, bookmark 18.8K, with tabs "Overview / Engagement / Audience," a "Summary" header, and four tiles: "Views 341,725," "Viewers 270,019," "Average watch time 19s," "Follows 3,225." A red arrow points at the 9.5K speech-bubble figure in frame 5 and at "Follows 3,225" in frames 6 and 7. Frames 11 and 17 carry section cards reading "Problem Discovery" and "Fast-Solution Lead Magnet." Frames 22 through 27 replace the speaker entirely with a navy full-screen text card headed "THE YAP FORMAT," listing in orange and white: "0-3s — Curiosity hook with a clear benefit. 3-6s — Show proof (visual or audio). 6-10s — Re-hook with a new benefit. 10-12s — Preview the 3-step framework. 12-20s — Step 1 (leave an open loop). 20-28s — Step 2 (build curiosity). 28-36s — Step 3 (deliver the payoff). 36-40s — Pattern interrupt (screenshot or visual). 40-45s — Comment keyword CTA." Frames 28 through 32 return to the speaker with captions "simple," "lead," "automations," "sales," and "the free." There are no email mockups, no inbox screenshots, no spam-filter interfaces, and no deliverability dashboards in any of the 32 frames.

The 9,500 "Leads" in the Title Card and the 9.5K Comment Count in the Screenshot Are the Same Number

Frames 1 and 2 state "Claude Wrote A Reel That Pulled 9,500 Leads." The transcript opens with the same claim: "So I had Claude write one of my Instagram reels and it pulled in over 9,500 leads from my business."

Frames 4 through 7 supply the only evidence offered for that number, and it does not say leads. The Instagram Insights panel labels 9.5K against a speech-bubble icon, which is the comment count. The narration confirms the identification one sentence later: "300,000 views and over 9,000 comments asking for my giveaway." A red arrow in frame 5 points directly at the 9.5K speech bubble while the caption reads "asking." The 9,500 figure in the headline and the 9,500 comments in the screenshot are one number used twice under two different labels.

Two other narrated figures are rounded away from what the panel shows. The transcript says "300,000 views" where the panel tile reads "Views 341,725," a 13.9 percent understatement. The transcript says "3,000 new followers" where the panel tile reads "Follows 3,225." No frame shows an email list, a CRM record, a lead count, or a DM-sent count. The step between a comment and a captured lead is the step the video never shows.

Meta's Private Reply Rules Allow Exactly One DM Per Comment, Inside Seven Days

The transcript handles the entire capture mechanism in nine words: "Simple automations handle the rest and collect the lead." The named tool is never shown and never spelled out on screen.

Meta's own developer documentation sets the boundaries that any such automation runs inside. The Messenger Platform send-messages reference states that Private Replies allow "you to send a message to a person when the person publishes a comment on one of your posts or ads," that the reply "can only be a single message, which will automatically include a link to the post or comment," and that it "must be sent within seven days of the person publishing the post or comment." The Instagram Platform webhooks reference confirms the trigger side: the comments field delivers comment notifications, "Advanced Access is required to receive comments and live_comments webhook notifications," and "The Instagram professional account that owns the media objects must be public to receive notifications for comments or @mentions."

Those rules put a ceiling on the funnel. Each of the 9,500 comments earns at most one outbound DM, that DM expires after seven days, and a comment that already received a private reply cannot receive a second. Converting that DM into an email address requires the commenter to open the DM, tap the link, and complete a form. None of those three conversion steps appears in the video or in any frame.

Meta's Ranking Guidelines Define the Video's Own CTA Step as Engagement Bait

The YAP card in frames 22 through 27 ends with "40-45s — Comment keyword CTA." The transcript describes the same beat: "ask it to write a script about solving the most common pain points that also gets viewers to comment on the post in order to receive that fast solution lead magnet."

Meta's Transparency Center entry on engagement bait defines the category as "Posts that explicitly request engagement (such as votes, shares, comments, tags, likes or other reactions) for purposes other than a specific call to action," with carve-outs the page names for causes, natural disasters, and life-threatening situations. A lead magnet giveaway is not among the named carve-outs. Meta's page on lowering the distribution of problematic content lists engagement bait among the low-quality content types it addresses by limiting visibility in feeds.

I could not load Instagram's own Recommendations Guidelines page to confirm the Instagram-specific wording, so I am reporting only what the two Meta Transparency Center pages I fetched actually say. On those pages, the definition covers the tactic the video teaches, and the stated consequence is reduced distribution. The video does not mention this policy at any point in its 64 seconds.

Nothing in 64 Seconds Touches Email, and Gmail's Bulk Sender Rules Never Mention AI Authorship

The post is titled "Claude can write content that kills email leads" and tagged #emailmarketing, #digitalmartketing (misspelled in the source metadata), #leadgeneration, and #digitalproducts. Across 206 transcript words and 32 frames, the word "email" is never spoken and never appears on screen. The platform discussed is Instagram. The surfaces discussed are reels, comments, and DMs. The title and the content do not describe the same subject.

The video also cites no open rate, no click rate, no deliverability figure, and no spam-filter behaviour, so there is no email statistic here to check. The nearest primary source on what actually governs bulk inbox placement is Google's Email sender guidelines, which require senders of 5,000+ daily messages to "Set up SPF and DKIM email authentication for your domain," to "Set up DMARC email authentication for your sending domain," to support one-click unsubscribe with "a clearly visible unsubscribe link in the message body," and to "Keep spam rates reported in Postmaster Tools below 0.30%." That page contains no mention of artificial intelligence, AI-written text, or AI-generated message content. Gmail's published bar is authentication, list hygiene, and complaint rate. Authorship by a model is not on it.

One Claude capability claim in the video does hold up against vendor documentation. The first prompt asks the model to "search through reviews to find the most common problems and pain points." Anthropic's web search tool documentation states that the tool "gives Claude direct access to real-time web content," that "The API runs the searches and provides Claude with the results," that citations "are always enabled for web search," and that the feature is billed at "$10 per 1,000 searches, plus standard token costs for search-generated content." Server-side search is a real, documented, metered feature. The per-search cost is a line item the video does not mention.

The YAP Card On Screen Contradicts the YAP Framework on Reilly's Own Website

The card in frames 22 through 27 lays out nine beats on an absolute timeline ending at 45 seconds. Reilly's own write-up, How To Write Yaps That Sell, lays out a seven-component framework in durations: "Calm Hook (4-7 seconds)," "Visual Proof (3-5 seconds)," "Benefit Re-hook (5-8 seconds)," "Retention Statement (2-4 seconds)," "Named Steps (30-50 seconds)," "Save Bait Interrupt (3-5 seconds)," and "Call To Action (8-12 seconds)."

The two versions disagree on structure and on length. The on-screen card runs nine beats to 45 seconds. The website framework runs seven components summing to between 55 and 91 seconds. The video itself runs 64.9 seconds, so it matches neither its own card nor, in beat count, the framework on the author's site.

The claimed results also differ between the two artifacts. The website page claims "over 400,000 views" plus "thousands of leads and sales," and separately describes a reel with "60k views with 3k+ comments" from an account at "8,000 followers." The Insights panel in frames 4 through 7 shows 341,725 views and 9.5K comments. These are not the same reel's numbers, and the video does not say which set belongs to the reel Claude wrote.

Reilly Sells the Thing the Video Promotes, and the Free Blueprint Is Itself the Lead Magnet

The transcript closes with an offer: "if you want all the prompts and automations I use to generate leads and sales on Instagram you can download them at the free blueprint link in my bio."

That blueprint is gated. jimmyreilly.com/getfunnel is headed "Free: The Funnel Framework Our Top Affiliates Use to Turn Content Into Email Subscribers" and requires a first name and email address before delivering anything, with the form copy "Get the complete system used to capture, nurture, and follow up with email leads." The delivery hub at jimmyreilly.com/1bonusdelivery38ej lists the free assets, including the "$10M Digital Product Blueprint," the "$100k Launch Blueprint," a "DFY Lead Capture & Sales Funnel," and a "150 Hot-Selling Digital Product List (2026 Version)," alongside a paid upsell headed "Get Every Prompt I Use To Earn $4,091.70/Day Selling Digital Products." The YAP write-up page carries a "$7 limited promotion" for the prompts and automations, delivered by the same comment-a-keyword-then-DM mechanism the video teaches.

So the answer is yes. Reilly sells what this video promotes, and the free download at the end is the opt-in step of the exact funnel the video describes. The $4,091.70/day earnings claim is unsourced, carries no stated time period, sample, or verification, and is Unverified.

Key Takeaways

  • Correction: The "9,500 leads" in the title card is the comment count. Frames 4 through 7 show the Instagram Insights speech-bubble metric at 9.5K, and the narration itself calls them "over 9,000 comments asking for my giveaway." No lead count, email count, or DM-sent count appears anywhere in the video.
  • Partial correction: The narrated "300,000 views" understates the panel's "Views 341,725" by 13.9 percent, and "3,000 new followers" understates "Follows 3,225." The underlying screenshot is the stronger number in both cases.
  • Verified: Meta's Messenger Platform documentation states a private reply "can only be a single message" and "must be sent within seven days of the person publishing the post or comment." One comment buys one DM, and the window expires.
  • Verified: Meta's Transparency Center defines engagement bait as posts that "explicitly request engagement (such as votes, shares, comments, tags, likes or other reactions)," and its distribution page lists engagement bait among content types whose feed visibility is limited. The video's "40-45s — Comment keyword CTA" beat sits inside that definition.
  • Correction: The title says "email leads" and the post is tagged #emailmarketing, but the word "email" appears in neither the 206-word transcript nor any of the 32 frames. The video is about Instagram reels, comments, and DMs.
  • Verified: Google's Email sender guidelines set the bulk-sender bar at SPF, DKIM, DMARC, one-click unsubscribe, and spam rates "below 0.30%" in Postmaster Tools. The page says nothing about AI-written content. There is no published Gmail rule penalising authorship by a model.
  • Verified: Anthropic's web search tool documentation confirms Claude can run server-side searches with citations always enabled, billed at "$10 per 1,000 searches" plus token costs. The video's "ask it to search through reviews" step is technically supported and separately metered.
  • Correction: The on-screen "THE YAP FORMAT" card (nine beats to 45 seconds) does not match the seven-component framework on Reilly's own site (55 to 91 seconds of components). The video runs 64.9 seconds and matches neither.
  • Context: Reilly sells the method. getfunnel trades the "free blueprint" for a name and email, the delivery hub carries a paid prompt upsell, and the YAP page runs a $7 promotion delivered by comment-keyword DM.
  • Unverified: The "$4,091.70/Day Selling Digital Products" headline on Reilly's delivery hub has no stated period, sample, net-of-cost basis, or third-party verification attached to it.
  • Unverified: The claim that Claude specifically, rather than the giveaway offer or the account's existing distribution, caused the 341,725 views. The video shows one reel's panel with no control, no before-and-after baseline, and no comparison against a human-written reel on the same account.

Resources

Published October 6, 2026. Writeup generated from a favorited TikTok.