Instagram content analysis

@bahaish_kapoor

PDF

Jul 25, 2026 – Sep 28, 2026 · 78 posts · Asia/KolkataView on Instagram

Followers
346.5K
As provided
Posts/week
8.4
78 posts in 65 days
Engagement rate
5.18%

Above 1–3% benchmark

Avg/median
17.9K / 8.2K
engagement per post
Comments
0.7
per 100 likes (median)

@bahaish_kapoor: Vintage Bollywood and retro illustration memes drive most engagement, and the 5.18% average engagement rate is inflated by a few viral carousels; the typical post sits at 2.36%.

Account snapshot

Avg engagement / post

17.9K

median 8.2K

Comments per 100 likes

0.7

median post

Posts without caption

33%

26 of 78

Hidden like counts

0

rates are accurate

  • Profile and window: @bahaish_kapoor has 346.5K followers. The analysis covers 78 posts from 26 Jul 2026 to 28 Sep 2026, a 65-day window in Asia/Kolkata time.
  • Posting rate: The account publishes 8.4 posts per week and was active on 49 of 65 days. The busiest day was 20 Aug 2026 with 4 posts, and no day reached 5 or more.
  • Format mix: Carousels are 64 of 78 posts (82%), images 13 (17%) and Reel/Video just 1. The account is effectively a carousel and static-image operation.
  • Average is distorted. Average engagement is 17.9K per post, but the median is 8.2K, so a handful of viral posts pull the mean up. The median is the fairer baseline.
  • Rate vs benchmark: The average engagement rate is 5.18% and the median is 2.36%, against a usual 1-3% benchmark. The typical post sits inside the benchmark range and the average sits well above it.
  • Comment depth: The median post earns 0.73 comments per 100 likes, and average comments are 104 against 17.8K average likes. This is an inference, but the audience appears to like far more than it converses.
  • Captions and tags: 33% of posts have no caption, and none of the 78 posts uses hashtags, emoji or a call to action. No post is sponsored or location-tagged.

Content buckets

What the account posts, and which themes earn their share of engagement.

  • Share of posts
  • Share of engagement
Vintage Bollywood Still Memes1.31×
33.3%
43.8%
Original Typewriter Reflections0.40×
19.2%
7.8%
Retro Illustration Memes2.27×
14.1%
32.0%
Typewriter Quotes by Authors0.17×
14.1%
2.5%
Relatable Deadpan Film Memes0.73×
6.4%
4.7%
Mythology & Literary Reflections1.25×
6.4%
8.0%
Collabs & Brand Features0.20×
6.4%
1.3%
Index = share of engagement ÷ share of posts. Above 1.00× means the bucket earns more than its share.
  • Vintage Bollywood Still Memes: 26 posts (33%) produce 44% of engagement, an index of 1.31, with a median of 17.0K. These are classic film stills with dialogue-style jokes, all carousels. Standouts are Classic VI (112K on 26 Jul 2026) and Classics XV (77.8K, with 1,376 comments, the highest comment count among the top posts).
  • Original Typewriter Reflections: 15 posts (19%) bring 7.8% of engagement, an index of 0.40, with a median of 3.4K. These are unattributed typewriter aphorisms, split between 9 carousels and 6 images, and neither format lifts results.
  • Retro Illustration Memes: 11 posts (14%) generate 32% of engagement, an index of 2.27, the best of any bucket. The average of 40.7K is inflated by the 195.6K post of 10 Sep 2026, but the median of 13.8K is still 1.7 times the account median. Other standouts are 95.6K on 11 Sep 2026 and 62.9K on 20 Sep 2026.
  • Typewriter Quotes by Authors: 11 posts (14%) deliver 2.5% of engagement, an index of 0.17, with a median of 3.2K. Average and median are close, so the weakness is consistent rather than caused by a few misses.
  • Minor: Three buckets have 5 posts each, which is at the edge of what can be judged. Relatable Deadpan Film Memes has an index of 0.73 and a median of 13.2K, which is solid but a small sample. Mythology & Literary Reflections has an index of 1.25 but a median of only 7.9K, because one Persephone image post (92.4K) carries the average. Collabs & Brand Features has an index of 0.20 and a median of 3.2K.
BucketPosts% posts% engagementIndexMedian
Vintage Bollywood Still MemesBlack-and-white or colour stills from classic Hindi films with white caption bubbles carrying a dialogue-style joke or punchline, often captioned with series labels like Classic, Classics, Best of or Send help.2633.3%43.8%1.3117,033
Original Typewriter ReflectionsUnattributed typewriter-style text posts of short original aphorisms, thoughts or musings on healing, creativity, whimsy, love and life.1519.2%7.8%0.403,396
Retro Illustration MemesMemes built on vintage illustrations, calendar art, pulp or magazine covers, or classical paintings with speech-bubble text, including the Retrovandalism series.1114.1%32.0%2.2713,765
Typewriter Quotes by AuthorsTypewriter-style text images that quote a named writer, thinker or scientist with an attribution line.1114.1%2.5%0.173,161
Relatable Deadpan Film MemesMeme posts using film or character stills with a modern, relatable one-liner about work, trauma, habits or daily life, often with captions like No context, Character sketches or Films are people too.56.4%4.7%0.7313,199
Mythology & Literary ReflectionsText-only posts that reference myth, literature or storytelling, such as Persephone or Orpheus, or longer reflective prose on art and memory.56.4%8.0%1.257,946
Collabs & Brand FeaturesPosts about collaborations, physical displays or features of the account's work, such as the collab with @5rhythms or memes on the walls of Patna Museum café.56.4%1.3%0.203,196

Formats

FormatPosts% eng.IndexMedian
Image13 (17%)11%0.663,396
Carousel64 (82%)89%1.0810,180
Reel/Video1 (1%)0%0.061,038

Gaps & opportunities

  • Retro illustrations are under-used. The bucket is 14% of posts but 32% of engagement, with an index of 2.27. Even excluding the top post, its median of 13.8K is well above the account median of 8.2K.
  • Vintage Bollywood is the reliable engine. Its median of 17.0K is 2.1 times the account median, and it holds 8 of the top 10 posts by engagement counting both Bollywood and illustration posts. It is the only large bucket that combines volume with strong typical results.
  • Typewriter text posts dilute results. Original Typewriter Reflections (19% of posts) and Typewriter Quotes by Authors (14% of posts) have indexes of 0.40 and 0.17. Both have medians near 3.2K-3.4K, well below the account median.
  • Static images lag carousels. The 13 image posts have a median of 3.4K against 10.2K for carousels, an index of 0.66 against 1.08. Images are mostly the typewriter text posts, so this is partly a content effect rather than a pure format effect.
  • Uncaptioned posts underperform. The 26 posts without a caption have a median of 5.1K, against 10.8K for the 52 captioned posts. Caption presence overlaps with bucket, since series labels sit on the meme buckets, so the gap should not be read as purely a caption effect.
  • Reels are untested. One Reel/Video in 78 posts (1,038 engagement) says nothing about the format. The account has no evidence either way.

Trajectory

Average engagement 19.1K in the first half of the window vs 16.8K in the second (-12%).

Median engagement per post, by month

The typical post, not skewed by one viral hit.

Posts published, by month

How often the account posted.

  • Average fell, median held. The first half (37 posts) averaged 19.1K with a median of 8.3K; the second half (41 posts) averaged 16.8K with a median of 8.1K. The lower average likely reflects fewer viral hits, which is an inference, while typical post performance is essentially flat.
  • July is a partial month. July covers 6 posts from 26 Jul 2026, with an average of 41.2K and a median of 18.8K. That is too few posts to treat as a baseline, and it includes the 112K Classic VI post and the 92.4K Persephone post.
  • Carousel share rose. Carousels were 71% of August posts and 95% of September posts. The September median (7.9K) is below August (9.2K), so the shift to carousels did not lift the typical post.

Timing (Asia/Kolkata)

Median engagement by posting hour (Asia/Kolkata)
  • 5+ posts (ranked)
  • Fewer than 5 posts
Median engagement by day
Mon
16.3K
Tue
10.9K
Wed
5.4K
Thu
3.7K
Fri
10.6K
Sat
10.6K
Sun
10.6K
  • 21:00 is the top slot. 5 posts at 21:00 have an average of 56.9K and a median of 31.1K, the strongest of any hour with enough posts. It is a small sample, so it is a lead to test rather than a rule.
  • 23:00 is also strong. 8 posts at 23:00 have a median of 16.1K and an average of 30.5K. Midday is the most common slot (8 posts at 12:00) and is average at best, with a median of 10.0K.
  • Weak hours: 22:00 (5 posts, median 6.0K), 20:00 (6 posts, median 6.1K) and 18:00 (8 posts, median 7.3K) sit below the account median of 8.2K. Hours with fewer than 5 posts, such as 08:00 and 17:00, cannot be called best or weak.
  • Days of the week: Monday has the strongest median (16.3K over 10 posts). Thursday is the busiest day with 18 posts and a 21.9K average, but its median is only 3.7K, so a few hits distort it. Wednesday is the weakest, with a median of 5.4K over 8 posts.

Captions & hashtags

  • Captions help, with caveats. Captioned posts have a median of 10.8K against 5.1K for uncaptioned posts. Captions are largely series labels on meme posts, so this likely reflects the bucket rather than caption writing.
  • Series labels dominate top posts. Labels such as Classic VI, Classics XV, Send help XXIV and Classic XVII account for most of the top captioned posts. The numbered series format looks like a recognisable, repeatable hook.
  • Short captions only. 48 of the 52 captioned posts have 1-100 characters, with a median of 11.1K. Longer captions (3 posts at 101-300 characters, 1 at 301-800) are too few to judge.
  • Mentions underperform. The 4 posts with @mentions average 6.7K against 18.5K for the rest. The sample is too small for a firm conclusion, and these are mostly collab posts.
  • No emoji, CTA or questions. No post uses emoji or a call to action, and only one asks a question. Their effect cannot be tested, and the account's comment rate is low.
  • No hashtags used. All 78 posts carry zero hashtags, so the analysis cannot compare tagged and untagged performance.
  • Reach effect unknown. Posts still reach 17.9K average engagement without tags, but whether tags would add discovery cannot be determined from the account's posts. It is an untested variable.
Caption lengthPostsMedian
No caption265,071
1-100 chars4811,139
101-300 chars36,583
301-800 chars11,038

Collabs & audio

  • Collabs are too few to judge. 3 collab posts have a median of 6.6K against 8.3K for 75 solo posts, a multiple of 0.8. Partners were @5rhythms, @circle_bycred and @nicojournal, one post each.
  • Audio barely used. The only Reel uses licensed audio, A Taste Of Honey by Herb Alpert & The Tijuana Brass, and earned 1,038 engagement. There are no original-audio posts, so nothing can be concluded.
  • No sponsored content. All 78 posts are organic, so there is no paid-partnership benchmark to compare against.
  • No location tags. None of the posts is location-tagged, so the effect of tagging is unknown, including for the Patna Museum café feature.
CollaboratorPostsAvg eng.
@5rhythms16,634
@circle_bycred16,583
@nicojournal13,196

Top 10 posts

Ranked by engagement (likes + comments). Open a post to see it on Instagram.

DateFormatBucketLikesCommentsWhat it shows
Sep 10, 2026CarouselRetro Illustration Memes195,252385Retrovandalism family illustration with intelligence jokeOpen
Jul 26, 2026CarouselVintage Bollywood Still Memes111,749409Classic colour still with sea joke dialogueOpen
Sep 11, 2026CarouselRetro Illustration Memes95,404177Vintage calendar art with planet jokeOpen
Jul 30, 2026ImageMythology & Literary Reflections92,259181Text post referencing Persephone mythOpen
Aug 3, 2026CarouselVintage Bollywood Still Memes76,4561,376Classics colour still with nostril jokeOpen
Sep 20, 2026CarouselRetro Illustration Memes62,800124Barbier illustration man in robe with inflation jokeOpen
Aug 17, 2026CarouselVintage Bollywood Still Memes46,993176Send help: classic couple still with targeted ads jokeOpen
Aug 6, 2026CarouselVintage Bollywood Still Memes39,928154Classic B&W still with drink dialogue jokeOpen
Aug 9, 2026CarouselVintage Bollywood Still Memes30,876241Classic B&W still with spider man jokeOpen
Jul 29, 2026CarouselVintage Bollywood Still Memes30,545219Classics B&W still with Spanish lie jokeOpen

Recommendations

In priority order. Each one cites the numbers behind it.

  1. 01

    Increase Retro Illustration output

    The bucket has an index of 2.27 and a median of 13.8K from only 11 posts (14%), and it produced the top post at 195.6K. Raising its share is the clearest performance lever.

  2. 02

    Keep Vintage Bollywood as the core series

    26 posts (33%) deliver 44% of engagement with a median of 17.0K, 2.1 times the account median. Continue the numbered Classic and Classics labels.

  3. 03

    Cut or rework typewriter text posts

    Original Typewriter Reflections (index 0.40, median 3.4K) and Typewriter Quotes by Authors (index 0.17, median 3.2K) take 19% and 14% of posts. Reduce frequency and test a visual treatment.

  4. 04

    Add captions to every post

    26 uncaptioned posts have a median of 5.1K against 10.8K for captioned posts. Part of the gap is explained by bucket, so test captions on like-for-like content.

  5. 05

    Test a 21:00 and 23:00 posting window

    21:00 shows a median of 31.1K (5 posts) and 23:00 a median of 16.1K (8 posts). Both are small samples, so run controlled tests, and review the weaker 18:00, 20:00 and 22:00 slots (medians 6.0K-7.3K).

  6. 06

    Favor carousels over static images

    Carousels have a median of 10.2K against 3.4K for images, and 82% of posts are already carousels. Move remaining text content into carousel form.

  7. 07

    Run a small hashtag and Reel test

    No post uses hashtags and one Reel (1,038 engagement) is the only video. A few tagged posts and several Reels would give the first evidence on both.

  8. 08

    Judge engagement on medians

    The average rate of 5.18% is inflated by viral posts; the median rate is 2.36%. Report the median rate, and track it against the flat second-half median of 8.1K.

Based on publicly visible Instagram data collected without logging in. Engagement = likes + comments. Content buckets are assigned by AI and may misclassify individual posts. Shares, reach, impressions, saves and audience demographics are not public and are not estimated. 242 figures in the written analysis were checked against the data.