Analytics Made Simple

← All series & guides

Anthropic

Claude

16 of 31 parts live

Anthropic Claude from first login through product map, Claude Code, and Cowork — one continuous path.

Start with part 1 →

Getting started

  1. 1 What Claude is (plain English) Claude is Anthropic’s AI assistant family, not one single app. This Part 1 map separates company, models, and products so you can start on web or mobile without hype, ranking wars, or treating chat answers as a source of record.
  2. 2 Free vs Pro vs Max: what you actually get Free, Pro, and Max differ mainly in usage capacity and paid tools, not in magic truth. This Part 2 explains limits, Pro extras, why the API Console is separate, and how to choose a plan without FOMO. Check claude.ai for current prices.
  3. 3 Claude on web and mobile: first 30 minutes Part 3 of Learn Claude from scratch. A guided first half hour on claude.ai and mobile so you leave with one real output and a clear map of web, desktop, and phone.
  4. 4 Chats, Projects, files, and Research Part 4 of Learn Claude from scratch. When bare chat is enough, when a Project earns its keep, how files and Research fit, and a starter Project kit you can copy.
  5. 5 Writing, summarizing, studying, planning Claude is strong at four everyday jobs: writing, summarizing, studying, and planning. Part 5 of Learn Claude from scratch shows a simple prompt pattern and worked examples you can reuse Monday morning.
  6. 6 Memory, privacy, and what never to paste Chat history, memory, and project files are three different things. Part 6 of Learn Claude from scratch covers privacy settings, a paste ladder for what never goes into a model, and review habits that keep work data safer.
  7. 7 Claude for work / Microsoft 365 light A light tour of Claude for work and Microsoft 365: personal Pro versus Team or Enterprise, connectors and write-tool consent, and a safe policy-redact-draft-verify-human-send loop. Not an admin manual.
  8. 8 When Claude is the wrong tool Closes Learn Claude from scratch with judgment: four wrong-tool zones, a good-enough checklist, walk-away signals, and what comes next in the product map, Code, and Cowork series.

Product map

  1. 9 Claude.ai chat vs everything else Part 1 of Claude product map. A plain-English tour of Claude chat versus Code, Cowork, Design, Science, and the API Console so you know which door to open before you pay, paste, or set policy.
  2. 10 Claude Code: what it is for Part 2 of Claude product map. What Claude Code is for when you will touch a project: explore a repo, edit many files, run tests, and open PRs, without a full install walkthrough yet.
  3. 11 Claude Cowork: what it is for Claude Cowork is the agent surface for non-coding knowledge work: folders and tools you choose, multi-step tasks, deliverables you still review. Part 3 of the Claude product map separates Cowork from chat and Code, covers plans and usage burn, and teases the deep Cowork tutorial series.
  4. 12 Claude Design and Claude Science (where they fit) Claude Design handles visual work (decks, mockups, prototypes, brand systems, exports). Claude Science is a research workbench with tools, compute, and auditable artifacts, not peer review. Part 4 of the Claude product map places both next to chat, Code, and Cowork.
  5. 13 Models chooser: Haiku, Sonnet, Opus, and Fable (no hype) Part 5 of the Claude product map: pick Haiku, Sonnet, Opus, or Fable without FOMO. Start on the default, upgrade only when a clear prompt still fails, and treat Mythos as limited. Model names change; the decision rule does not.
  6. 14 API and console: only if you build apps Part 6 closes the Claude product map: chat subscriptions and the API Console are separate bills and separate jobs. Open platform.claude.com only if you build software that calls models. Most AMS readers stay in chat until that day.

Claude Code

  1. 15 Install / open Claude Code and first run Part 1 of the Claude Code tutorial. Install with native scripts, Homebrew, or WinGet, verify with claude --version, log in, pick a surface, and survive first run with default permissions and a tiny reviewed change.
  2. 16 Talking to a codebase: ask, explore, change Part 2 of the Claude Code tutorial. Talk to a repo in three beats: ask for a map with real paths, explore until you know what might break, then make a small change and review the diff before you commit.
  3. 17 Slash commands and common command patterns Part 3 of the Claude Code tutorial. Type slash to find commands, Tab to complete them, and use shell patterns like claude -c and claude -p so you stop reinventing the same session from scratch every morning. Scheduled · August 10, 2026
  4. 18 Skills: what they are, how to add and use Part 4 of the Claude Code tutorial. Skills are on-demand playbooks in SKILL.md files. Add them under .claude/skills or ~/.claude/skills, invoke with /name, or let Claude load them when the description matches. Scheduled · August 11, 2026
  5. 19 Memory and context: what sticks, what resets Part 5 of the Claude Code tutorial. Chat resets with /clear or a new session, CLAUDE.md reloads every time, auto memory can carry learnings if you verify it is on, and git plus files stay the real source of truth when the context window fills. Scheduled · August 12, 2026
  6. 20 Markdown and project instruction files (CLAUDE.md, AGENTS.md, and friends) Part 6 of the Claude Code tutorial. How CLAUDE.md, AGENTS.md, README, and related instruction files split work between humans and agents, with a sample CLAUDE.md skeleton and the rule that short true notes beat long aspirational essays. Scheduled · August 13, 2026
  7. 21 Plugins, connectors, and tools Part 7 of Claude Code tutorial. Map built-in tools, MCP connectors, and plugins that bundle skills, subagents, hooks, and settings. Climb the ladder from a plain task to a plugin only when you are stuck. Scheduled · August 14, 2026
  8. 22 Loops and agentic runs: multi-step work Part 8 of Claude Code tutorial. Run multi-step agentic work with a goal-plan-act-observe loop, checkpoints, and clear autonomy: watch, steer, or walk away only when rails are real. Scheduled · August 15, 2026
  9. 23 Reviewing diffs, undoing, not shipping blind Part 9 of the Claude Code tutorial. Review every agent change set: read the diff, run tests, check secrets and scope, then own the commit. Undo with restore and careful reset; treat force-push as dangerous on shared history. Scheduled · August 16, 2026
  10. 24 Team habits and safety rails Part 10 closes the Claude Code tutorial with team rails: shared CLAUDE.md and skills, required PRs and green CI, secrets policy, no silent prod deploys from agents, and teaching review before high autonomy. Next up: Claude Cowork tutorial, plus Learn Claude and the product map for orientation. Scheduled · August 17, 2026

Claude Cowork

  1. 25 Cowork vs chat vs Claude Code Chat, Claude Code, and Claude Cowork share a model family but not a job. Part 1 of the Claude Cowork tutorial separates conversation from software work from office file tasks so you pick the right surface before you burn half a day pasting between apps. Scheduled · August 18, 2026
  2. 26 Setup and first real file task Part 2 of the Claude Cowork tutorial walks paid-plan setup, a dedicated working folder, and one real file task you can finish and review. You leave with a first-task recipe, not another feature tour. Scheduled · August 19, 2026
  3. 27 Working on folders, docs, and spreadsheets Part 3 of Claude Cowork tutorial. Work folders, docs, and sheets with a clear recipe: goal, inputs, constraints, format, and done check. Turn receipts into expense workbooks, notes into reports, and rough inputs into real PPTX files you still review. Scheduled · August 20, 2026
  4. 28 Plugins and connectors for office work Part 4 of Claude Cowork tutorial. Plugins bundle skills, connectors, and sub-agents for office work. Start with folders only, then climb a ladder through instructions, connectors, plugins, and higher-risk browser or computer use. Scheduled · August 21, 2026
  5. 29 Autonomy: when to watch vs walk away Part 5 of the Claude Cowork tutorial. Choose Manual, Auto, or Skip with eyes open: deletion still needs Allow, Auto screens actions (and uses more quota), Skip does not. Watch high-stakes work; walk away only with tight folders, limited write tools, and a review afterward. Scheduled · August 22, 2026
  6. 30 Quality control and avoiding workslop Part 6 of the Claude Cowork tutorial. Catch workslop before it ships: skim structure, check numbers and names, verify scope, then human sign-off. Auto mode screens risky actions; it does not certify your metrics. You still own the send and the number on the slide. Scheduled · August 23, 2026
  7. 31 Team and enterprise notes Part 7 closes the Claude Cowork tutorial with Team and Enterprise notes: admin controls, spend and roles, OpenTelemetry monitoring, why personal Pro is not a policy waiver, and when to involve IT before connectors or computer use. Next on AMS: Phase P meta on-ramps or the ChatGPT track. Scheduled · August 24, 2026