Case Study Comment Moderation June 2026

91% of Spam Comments Filtered in 28 Days

How a food & wellness creator with 38K followers eliminated manual spam deletion and improved their engagement rate by 68% — without touching a single comment manually.

91%

Spam Filtered

+68%

Engagement Rate

2.5 hrs

Saved / Week

28 days

To See Results

Account Overview

Creator Profile

Niche Food & Wellness
Followers 38,400
Account type Creator (Professional)
Primary content Reels, Carousels
Posting frequency 4–5 posts/week

Before InstaGrow

Spam comments/week ~163
Engagement rate 1.9%
Time on moderation 2.5 hrs/week
Manual filter in use Native only (partial)
Missed spam rate ~62%

The Problem

The creator's Reels had been growing steadily — recipe content and wellness tips were consistently hitting 10K–30K views. But as reach grew, so did inbound spam. By the time they connected InstaGrow, their comment sections were receiving an average of 163 spam comments per week — roughly 23 per day across all posts.

The spam fell into three main patterns: follow-for-follow bots ("Great content! Follow back @xyz"), DM-bait accounts pushing earn-from-home links, and emoji floods from freshly-created accounts with under 50 followers. Instagram's native Hidden Words filter was active, but the creator estimated it was catching fewer than 40% of what came in.

The real cost wasn't the spam itself — it was what it was doing to the numbers. Real follower comments were getting buried. The visible engagement rate sat at 1.9%, well below the 3–4% typical for food creators in this follower range. And manual cleanup was consuming 2.5 hours per week with no scalable path forward.

Spam breakdown before InstaGrow (weekly average)

Follow-for-follow bots 74 comments (45%)
DM-bait / promo accounts 54 comments (33%)
Emoji floods / new accounts 35 comments (22%)

The Setup

The creator built their full three-layer filter in 18 minutes after connecting InstaGrow. Here's the exact configuration they used.

1

Instagram native Hidden Words — enabled

The first step was simply enabling Instagram's built-in Hide Offensive Comments toggle (Settings → Privacy → Comments) and turning on the Manual Filter. This was already partially configured but had never been fully activated.
2

InstaGrow Keywords — 12 niche-specific rules

Added 12 keyword rules targeting the most common patterns for their niche. Unlike Instagram's Hidden Words (exact match only), InstaGrow's keyword matching is case-insensitive and catches partial matches automatically.

follow back → Hide
follow for follow → Hide
f4f → Hide
check my bio → Hide
check my profile → Hide
collab → Hide
dm for → Hide
earn from home → Delete
make money → Delete
buy followers → Delete
free followers → Delete
promo code → Hide
3

InstaGrow Rules — 2 multi-condition rules

The two Rules handled what keywords alone couldn't catch: emoji floods and buyer-intent comments.

// Rule 1: Emoji spam from new accounts

Emoji count > 8 AND Follower count < 75

→ Action: Delete

// Rule 2: Capture buyer intent (recipe queries)

Comment contains "recipe" OR "ingredients" AND Length > 8 chars

→ Action: Auto Reply ("Full recipe in our bio link!") + Send DM (with recipe link)

Rule 2 was set up not for spam, but to capture engagement. Anyone asking about a recipe got an automatic public reply and a DM — converting comment section activity into direct outreach.

28-Day Results

Spam Moderation

Total spam comments (28d) 712
Auto-filtered by InstaGrow 648 (91%)
Caught by keyword rules 431
Caught by Rules (emoji + bot) 217
Slipped through 64 (9%) — manually cleared
Time on manual moderation ~12 min/week (was 2.5 hrs)

Engagement & Growth

Engagement rate — before 1.9%
Engagement rate — after 3.2%
Improvement +68%
Auto-DMs sent (recipe rule) 53
Auto-replies posted 53
Manual comment responses 0 (all automated)

Before — Week 1

"Great content! Follow back @xyz_promos123" bot
"🔥🔥🔥🔥🔥🔥🔥🔥🔥🔥🔥🔥" emoji flood
"Earn $300/day from home! DM me" scam
"This recipe looks amazing, what are the ingredients?" real follower — buried on page 3

After — Week 4

"What protein do you use in this?" real follower
"Made this last night — the whole family loved it!"
"Can I substitute oat flour here?"
"What are the ingredients?" → auto-reply sent + DM triggered

Key Takeaways

Keyword rules do the heavy lifting

66% of filtered spam (431 of 648) was caught by keyword rules — not the AI. Simple phrases like "follow back" and "check my bio" cover the majority of bot patterns. Add these first.

Rules catch what keywords can't

The emoji flood rule (emoji > 8 AND followers < 75) caught 217 comments that had zero keywords to match on — pure emoji spam from bot accounts. Without multi-condition rules, these all get through.

Engagement rate reflects real interaction

The jump from 1.9% to 3.2% wasn't from more engagement — it was from less fake engagement diluting the real number. Spam removal reveals the actual quality of your audience.

Data note: All figures are based on InstaGrow platform analytics from a creator account in the food & wellness niche (30K–50K follower range), Q2 2026. Creator identity anonymized at their request. Results represent this account's experience over the 28-day measurement period and are not guaranteed for all accounts — spam volume and engagement lift vary based on niche, posting frequency, and audience composition.

Set up your spam filter in under 20 minutes

InstaGrow's comment moderation is free, uses the official Meta API, and takes the same 18-minute setup described in this case study.

Start Free — No Credit Card

Related reading