1. Sends (DM shares) — the single highest-weighted signal now. When someone privately sends a post to a friend, it's read as a genuine trust vouch. Likes barely move the needle anymore.
2. Saves — still valuable, especially for anything useful or reference-worthy (routines, checklists, explainers).
3. Watch time — how long people stay on a Reel, and whether they rewatch it. Getting past the first 3 seconds is the make-or-break moment.
4. Profile clicks — someone caring enough to go check out the account is a strong intent signal, and it's the step right before a follow or a DM enquiry.
Content built purely for likes. Hashtag stuffing (30+ tags now reads as spam). Reposting other people's content for "activity" — accounts doing this repeatedly are being excluded from recommendations entirely. Jumping between unrelated topics for reach — the algorithm now rewards accounts that stay tightly in their lane.
Original, specific, "send this to someone" content. Staying consistently in one topic (haircare/scalp health, in our case). Stories used actively — polls, questions, reactions — which now directly feed into how many followers see the next feed post. Real, unpolished human content over anything that looks AI-made or over-produced.
Their entire account is built around one person — a 90-year-old founder figure with a strong, recognisable personality. Every content format serves that one asset. It's less "a hair brand with a founder" and more "a founder with a hair brand attached." That's the structural insight worth borrowing — not the specific tone.
They borrow reach through PR and collaboration placements (podcasts, fintech brand partnerships) — a distribution advantage from scale, not content quality. And they run an active tagged-UGC program, leaning on real customers' before/after posts for social proof at no production cost.
One tactic we're deliberately not copying: they reshare unrelated viral content to stay "algorithmically active." Under current rules this actively hurts topical authority and can get an account excluded from recommendations — the opposite of what we want.
So the plan borrows the structure that's working — a consistent human face, recipe-style actionable content, real unscripted moments, light UGC — and runs it entirely through our own voice: warm but authoritative, direct not clinical, no overclaims, rooted in Priya's actual expertise.
Reels-first, hook in the first second, no logo intro. Lead with the specific frustration — "you've spent ₹15,000 on serums and nothing's changed" — before any solution appears. Filmed as real, unscripted customer moments from the store wherever possible, not scripted testimonials. Caption ends with a direct, specific ask — "send this to someone who's been told it's just stress."
Reframed from abstract education to one specific, actionable routine per carousel — long-form (8–10 slides), Priya's voice throughout, soft product mention only on the last slide. More slides means more chances for Instagram to re-show the carousel to people who didn't finish swiping.
A recurring, recognisable format — same simple visual frame every time, Priya answering one real question on camera. This is our answer to a "signature format" — consistency here is what the algorithm now rewards as topical authority, and it's the trust-asset a mascot-driven competitor can't actually match.
Unscripted, phone-shot reactions from real customers post-consultation, tagged and reshared — the store now makes this possible at zero production cost. This is where we start building the social-proof library the category leader already has.
This document is the research and direction only — no calendar dates or specific posts yet. Once the four pillars and the day-to-day changes above are signed off, the next deliverable is the monthly content calendar, with every post tagged by its job (reach / trust / convert) and built against the "would someone send this" filter from the start.