Data Analytics and Visualization with AI is a hands-on, no-code workshop in using AI as an analyst's assistant for content and audience data from many platforms, from cleaning and combining exports to finding insights and building a Looker Studio dashboard. It suits content, social media and marketing teams, and you leave with a dashboard and an insight brief for management.
Content and digital media teams now have data pouring in from many platforms at once: Facebook, Instagram, YouTube, TikTok, LINE and their own websites. Each platform exports files in a different layout and counts its metrics differently, so most of the time goes into collecting and reshaping data instead of answering the questions that matter: which content works, why, and what to do next.
This course uses AI as an analyst's assistant across the whole data workflow, from setting questions and KPIs, cleaning and combining data from several brands and platforms, finding insights and classifying Thai-language text, through to building a dashboard in Looker Studio and writing a summary for management. The core skill throughout is checking whether the numbers AI produces are correct. By the end of the day each learner has built their own Content Performance Dashboard and a one-page Insight Brief, without writing any code. (1 day, 6 hours, Beginner to Intermediate level.)
Course Details
A 1-day, 6-hour hands-on, no-code course (09:00-16:00) at Beginner to Intermediate level for content, social media, marketing and digital media teams. In one day it covers the full path from platform exports to a Looker Studio dashboard and an executive insight brief, with AI as the assistant. It differs from Claude AI for Data Analyst, which covers general data analysis with Claude over 2 days, from Google Data Studio
for Data Visualization, which goes deep into Looker Studio itself over 2 days, and from NLP and Thai Text Analytics with LLMs, which is a technical course for developers. Learners use their own notebook, Google account and an AI account that can analyse files, either their own or their organisation's. Workshops use a mock multi-brand, multi-platform dataset. Learners take home their dashboard, insight brief, a prompt template set and a checklist for verifying AI output.
Day 1: From Multi-platform Data to Dashboards and Insights with AI
The data analytics lifecycle: Ask, Prepare, Process, Analyze, Share and Act, and where AI saves the most time
Four levels of analysis: descriptive, diagnostic, predictive and prescriptive, with example questions from content work
Strengths and limits of AI with numbers: hallucination, and the difference between AI that calculates with code and AI that guesses from text
How to review AI output before using it for decisions
Content KPIs: reach, impressions, engagement rate, watch time, view-through rate, CTR and follower growth
Agreeing one definition for each metric across the organisation when platforms count differently
Data governance and PDPA: what data may and may not go into AI, and personal versus organisational AI accounts
Workshop 1: set 3 business questions from a mock scenario, turn them into KPIs and draft the data dictionary used all day
The structure of files exported from Meta Business Suite, YouTube Studio and TikTok
Common problems: mismatched date formats, different column names and numbers stored as text
The Role, Context, Task, Format and Constraint prompt pattern for data work
Ask AI to show its steps or code so the work can be checked and repeated
Handle blanks, duplicates, outliers and inconsistent Thai spellings of brand and content category names
Design a central tidy-data schema and have AI map the columns from each source
Check after cleaning: spot checks, reconciling totals with the source files, and an AI summary of what changed
Workshop 2: combine mock data from 3 platforms for 3 brands into one central table and save the prompts as team templates
EDA with AI file analysis: overview, distributions, trends over time and relationships between variables
Compare performance by brand, platform, content format, topic and posting time
Diagnostic analysis: use AI to propose why some content stood out or dropped, then test it with data
Lab: find the 3 best-performing content patterns for each brand, with reasons you can verify
Detect anomalies and make simple trend forecasts, and understand the limits of short data histories
Use AI to classify content topics from Thai headlines
Analyse the sentiment of audience comments and spot-check how accurate the AI is
Lab: build a topic and sentiment table and join it back to the content performance table
Recheck key numbers calculated by AI with a pivot table or formulas in Google Sheets or Excel
Separate facts from assumptions and state the confidence level of each conclusion
Workshop 3: answer the 3 business questions from Workshop 1 and summarise them as 5 verified key findings
Match the question to the chart: comparison, trend, composition, distribution and relationship
Charts to avoid and why
Declutter, use colour to highlight what matters, and write chart titles as insight sentences
Order the layout so managers grasp the main point straight away
Connect the central table from Google Sheets to Looker Studio
Build scorecards, time series, bar charts and tables with filters for brand, platform and date range
Ask AI to propose a dashboard layout for each audience and help write calculated fields
Lab: have AI critique the finished dashboard and apply the suggestions you agree with
A structure for telling stories with data: context, findings, meaning and recommendations
Use AI to draft an executive summary that cites real figures from the dashboard
Workshop 4: a one-page Content Performance Dashboard and a one-page insight brief, presented in 3 minutes per group with feedback
Wrap-up, how to apply this to your organisation's real data, and Q&A
No public rounds are open right now. Join the waiting list and we will contact you first when the next round opens, or ask us on LINE. Or call 02-570-8449 or 088-807-9770
Claude AI for Data Analyst is a 2-day course that goes deep into general data analysis with Claude, Google Data Studio for Data Visualization is a 2-day course on Looker Studio itself, and NLP and Thai Text Analytics with LLMs is technical work for developers. This course is a 1-day, no-code workshop for content, social media and marketing teams. It covers the full path from multi-platform exports to a Looker Studio dashboard and an executive insight brief, with AI assisting at every step.
No. You should be comfortable with basic spreadsheet work such as opening a CSV file, sorting, filtering and simple formulas like SUM or AVERAGE, and have used an AI chat tool before. AI handles the technical calculations, and you learn how to check its numbers with pivot tables or spreadsheet formulas.
The institute does not provide accounts. You use an AI account that can upload and analyse files, either your own or your organisation's, such as Claude, ChatGPT or Gemini, plus a Google account for Google Sheets and Looker Studio. File analysis features and usage limits vary by plan. If you only have a free plan that lacks a feature, the instructor demonstrates that part. If your organisation has a business AI account, we recommend using it in line with company policy.
All workshops use a mock multi-brand, multi-platform dataset built to resemble real data, not data from any real organisation. If you want to try your own data, remove or anonymise personal data first and use only AI accounts your organisation allows, following PDPA and your organisation's AI policy, both of which are covered in the course.
Yes. The prompt patterns, verification habits and dashboard design principles apply to any tool. Learners with Copilot in Excel through their own or their organisation's Microsoft 365 plan can use it for the data preparation and analysis parts. The dashboard work in class uses Looker Studio, which is free with a Google account, and the same ideas carry over to Power BI Desktop.
You leave with the Content Performance Dashboard you built in Looker Studio, which can be pointed at real data in Google Sheets, a one-page insight brief as a model for management reports, prompt templates for preparing data, analysing it and writing summaries, a checklist for verifying AI output, do's and don'ts for using AI with organisational data, the course handbook and a certificate of completion.
Every organization now holds a vast amount of data, so the real problem is not a lack of data but turning it into decisions fast enough. Many analytics teams still spend most of their time on repetitive work such as cleaning data, formatting tables, building…
Many organisations still build reports by copying data out of several systems into Excel every week and passing the files around by email. The numbers never quite match, managers wait for their reports, and the people who make them spend their time on…
Customer text is one of the most valuable and least used sources of data: product reviews, social media comments, chats with admins and open-ended survey answers. Thai text is harder to analyse than English because words are not separated by spaces, and it is…
