Operator take: Reels reach in India dropped 18–25% across our test accounts in early March. The algorithm is now weighting “saves” and “watch-throughs” significantly higher than likes and comments. Creators leaning into “send” and “save” CTAs are recovering reach within 14 days.
Something shifted in the Instagram Reels algorithm in early March. Brands across our client roster, and creator partners we work with on campaigns, started seeing reach drop sharply — sometimes for accounts that had been steady for months.
It’s not a content issue. The same accounts publishing the same content as the previous month, with the same hashtags and the same posting schedule, are seeing different distribution.
Here’s what we know after three weeks of testing on 14 accounts (8 Indian D2C brands, 6 creators) and what’s working to recover reach.
What changed
Three signals appear to be re-weighted upward in the algorithm:
- Saves — saving a Reel is now valued roughly 3× higher than a like, vs. ~1.5× previously.
- Sends/shares — sending a Reel via DM is the highest-value signal, valued ~5× a like.
- Watch-through to 80%+ — Reels with high completion rate get amplified into broader audiences much faster than before.
And two are downweighted:
- Likes alone — Reels with high like ratios but low save/share ratios are not promoted.
- Comments without sentiment depth — short, low-content comments (“nice”, “great”) are essentially ignored as ranking signals.
Who’s most affected
| Account type | Reach delta |
|---|---|
| D2C product brands (visual-first content) | −14 to −22% |
| Lifestyle creators (lookbook/transition content) | −25 to −35% |
| Educational creators (carousel-style narration) | +8 to +18% |
| Comedy/entertainment | −4 to +10% |
| Local food/restaurant content | −12 to −18% |
The accounts gaining are those producing content where saves and sends naturally happen — recipe videos, “do this when X” tutorials, restaurant location reveals.
The response that’s working
1. Move “save” CTAs into the first 3 seconds
“Save this for later” or “Save before it’s gone” delivered in the opening hook generates 3–4× more saves than the same CTA at the end. We’ve A/B tested across 18 ads. Front-loading the save CTA recovers ~60% of lost reach.
2. Build content that creates “send” moments
Content that earns a “send to friend” share is structurally specific:
- Tutorials someone needs (“how to clean stainless steel” beats “stainless steel facts”)
- Funny/relatable observations a specific friend would appreciate
- Recommendations (“places to eat in HSR” → high send rate)
- Couple-relatable or family-relatable content
Brands that produced 1 “send-bait” Reel per week saw reach recovery within 21 days.
3. First 3 seconds = stop scrolling, second 3 seconds = stay
The algorithm now reads watch-through to 80% as the primary engagement signal. Hooks that get a stop-scroll but don’t earn the second 3 seconds underperform now in ways they didn’t 90 days ago.
Re-cut every Reel for two arcs: first arc earns the stop, second arc earns the watch-through.
4. Drop the “swipe up” / “link in bio” mid-Reel CTAs
External-action CTAs reduce watch-through. Move the “link in bio” to the end card or the caption only. Don’t break the watch experience.
5. Reduce posting frequency, increase quality
Counter-intuitive but consistent: accounts that dropped from 5 Reels/week to 3 Reels/week (with more production effort per Reel) recovered reach faster than accounts that doubled posting frequency.
What not to do
- Don’t buy “engagement boosting” services. The likes generated are downweighted; the engagement-pod patterns are flagged.
- Don’t add 30 hashtags. The hashtag signal is roughly half what it was 18 months ago.
- Don’t migrate budget out of Reels. Reach is down but conversion-per-reach is still strong. Audit conversions, not impressions.
- Don’t change account focus mid-shift. Sudden niche pivots compound the algorithmic uncertainty.
The 30-day recovery template
- Audit your last 30 Reels. Sort by save-rate. The top 3 are your template; the bottom 5 are mistakes.
- Re-cut your next 8 Reels with save CTAs in the first 3 seconds.
- Produce 1 “send-bait” Reel per week.
- Drop posting frequency to 3-4/week, raising production value.
- Track watch-through to 80% as your primary KPI for 30 days.
Forecast
The shift looks like a permanent re-weighting, not a temporary algorithm test. Indian creator and brand accounts that adapt quickly will recover most lost reach within 30–45 days. Accounts that don’t adapt will see reach plateau at the new baseline through Q3.
If you want our team to audit your account against the new algorithm and produce a 30-day recovery brief, our first call is free.
Webfluence is a Bangalore-based performance marketing studio running paid, SEO and creative for 30+ Indian brands. If you’d like a working session on what any of this means for your brand, our team takes free 30-minute calls from our HSR Layout office.
Want more like this? Subscribe to Pulse — daily intelligence from the Indian marketing front lines.
Leave a Reply