AIAP Tool Guides·Claude·16:06

Claude Multimodal and Long Context: Analysing Images, Summarising a 64-Page Report, and Mining a YouTube CSV with Code

Karan Rana Brand/marketing manager, YouTuber, podcaster and course creator ("a non-tech guy"); presents the AI for Marketing, NotebookLM and Manus tool guides

▶LMS lesson

The short version

Attach an image with the plus button: a rival cat-food poster analysed for messaging with differentiation ideas; a stage photo read for body language (0:00-0:06).
Long context: a product-marketing industry report (Claude counted 64 pages) summarised in five bullets (0:06-0:09).
A YouTube Shorts CSV: Claude wrote and ran code, found brand-mistake stories and brand names in titles drive views, and suggested five next topics - with the human supplying context (0:09-0:15).

At a glance, three clicks deep

Skim here first: the closed row is the glance, open is the study card with the key points and timestamps, and the ↓ link drops to that concept's full write-up below.

01Long context on data: CSV analysis with visible codeUpload thousands of rows, ask for what works and what to do next;›

Upload thousands of rows, ask for what works and what to do next; Claude parses the file with code you can watch, then summarises.

Image analysis (0:01-0:06)

PDF summary (0:06-0:09)

CSV (0:09-0:14)

↓ Full write-up of this concept

02Images in, strategy outModality means input format;›

Modality means input format; attach an image and ask a business question about it.

Poster (0:01-0:04)

Body language (0:04-0:06)

↓ Full write-up of this concept

The concepts in full

Long context on data: CSV analysis with visible code

"You're saving hours by investing just five minutes."

Prompt (as spoken): "analyze my YouTube channel data, summarize which videos are working and which are not, and basis that suggest me next five video topics; keep the format structured and easy to read." Result: ~500 videos analysed (471 Shorts); a Rolex-mistake story led with ~7 lakh views and 82% retention; brand names in titles meant roughly ten times the views; 352 videos under 100 views - which he explains with context Claude lacked (the sheet covered only a year). Suggestions included "how Nike's biggest mistake made them billions".

Why it matters

Data-backed content decisions without a spreadsheet expert.

This idea elsewhereparallelAI for Marketing: briefing Claude like an analyst on a CSV

Images in, strategy out

"Slide-ready pointers which you can propose to your CMOs."

Prompt: "attaching a poster of my rival cat food brand - analyze the messaging strategy and tactics used, give me tips on how I can differentiate my company's poster." Claude read the headline and found emotional appeal and vet authority, then suggested life-stage lines, premium small-batch positioning and scientific claims. For a stage photo it described each speaker's posture and gestures.

Why it matters

Fast competitive and visual analysis.

This idea elsewhereparallelGemini: the same competitor-poster exercise

Tools referenced

Tool	Coverage	Moment	Context
Claude	demonstrated
ChatGPT	mentioned
DeepSeek	mentioned

Action items

Resources mentioned

Resources

docLesson page on the platform

Extraction notes

This page was built from an auto-generated transcript, which garbles product and people's names. Those were corrected silently in everything above and logged here for transparency. The warnings flag claims that were true on the recording day but change fast.

Transcript corrections applied

The transcript says	The trainer actually means
cloud	Claude
shots	Shorts
pma	PMA (Product Marketing Alliance)

True on recording day — verify before relying

Vendor UI walkthrough: menu names, layout, plan limits and pricing change monthly - check the live product before following click paths.
"Long context as Claude's differentiator versus ChatGPT or DeepSeek" is a mid-2025 claim; he expected a free-plan upload error that did not occur (0:06-0:07).