← AIAP Topic CoursesAll programsHomeSearch
AIAP Topic Courses·Data Analysis·20:53

Numerous AI: ChatGPT Inside Google Sheets - Sentiment, Keywords and Issue Buckets on Amazon Reviews

Data Analysis trainer (unnamed) Marketing-analytics practitioner; runs four AI data tools on one e-commerce orders dataset
▶LMS lesson
← PreviousMultivariate Analysis: Regression, PCA, Clustering, and a Three-Variable Pivot
Lesson 8 of 11 in Data Analysis
Next →Google Colab with Gemini: One Prompt to a Full Python Notebook, and Fixing a KeyError

The short version

  1. Numerous AI is an add-on, not an upload: =NAI() prompts as cell formulas, a sidebar, point-and-click tasks; the vendor says it cannot see your data. Install: sign up (~60 free credits) > Extensions > Add-ons > Get add-ons > Numerous AI > open as sidebar (0:00-0:11).
  2. Put the analytical hat on first: what can these columns answer - date trend, rating split (3 = neutral), platform, review text for sentiment, issue buckets routed to QA or logistics (0:04-0:08).
  3. Demos on Epigamia yogurt reviews: Generate formula for domain extraction (a regex, two tries); sentiment into column N then drag down; top-5 comma-separated keywords for SEO; classify into quality / delivery / no issue. 'The output is only as good as the prompt' (0:12-0:20).

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.

01Numerous AI: an in-sheet AI formula add-onPrompt as a formula;›

Prompt as a formula; data stays in place.

Positioning and install (0:00-0:11)

Generate formula (0:12-0:13)

↓ Full write-up of this concept

02Prompt-driven columns: sentiment, keywords, issue bucketsSentiment -> keywords -> buckets, one column each.›

Sentiment -> keywords -> buckets, one column each.

Sentiment (0:14-0:15)

Keywords (0:16-0:18)

Buckets (0:19-0:20)

↓ Full write-up of this concept

The concepts in full

01

Numerous AI: an in-sheet AI formula add-on

The data never leaves the spreadsheet. The prompt lives in a cell and drags down like any formula.

A Google Sheets (and Excel) extension adding =NAI(...) so a GPT prompt is a formula, plus a sidebar UI for common tasks. PROCEDURE: sign up (about 60 free credits) > Extensions > Add-ons > Get add-ons > search Numerous AI > Install > Extensions > show as sidebar. Privacy: the vendor claims it cannot see sheet data (a vendor claim, not an audit). 'Generate formulas' turns 'extract domain names from the URL in column L' into a REGEXEXTRACT-style formula (a couple of retries). Before prompting, list the questions the columns can answer.

Why it matters

The lightest-touch AI data tool in the series, for people who live in Sheets.

02

Prompt-driven columns: sentiment, keywords, issue buckets

One prompt classifies every review; a second pulls the five words customers use; a third routes complaints to the right team.

Sentiment: 'analyse the sentiment of this column as positive, negative or neutral' into N2 under a 'sentiment' header, then drag down - about a minute, no Python. Keywords: 'extract top 5 keywords from the review in column M, comma separated, into O2' - the phrases customers associate with the brand ('best strawberry yogurt') for SEO, then frequency or a word cloud. Issue buckets: 'classify the reviews in column M as quality, delivery or no issue into P2' - filter quality to QA, delivery to logistics. Prompt quality decides output quality.

Why it matters

Three reusable column recipes for any review or feedback export.

This idea elsewhereparallelContent Library on audience mining

Tools referenced

ToolCoverageMomentContext
Numerous AIdemonstratedInstall, =NAI, sentiment, keywords, buckets
Google Sheetsdemonstrated
ChatGPTmentioned

Action items

    Resources mentioned

    Resources
    • docAmazon reviews export (Epigamia yogurt)

    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 saysThe trainer actually means
    Numerous dot a INumerous.ai
    Epic gamerEpigamia
    Google Soap resultsGoogle SERP results
    blanketBlinkit
    Charge GPT / Chart GPT / ChargeGPDChatGPT
    cloud / Cloud AIClaude
    Mishou / Misho / MISOMeesho
    main dot com / mail dot comMake.com

    True on recording day — verify before relying

    ← PreviousMultivariate Analysis: Regression, PCA, Clustering, and a Three-Variable PivotNext →Google Colab with Gemini: One Prompt to a Full Python Notebook, and Fixing a KeyError