Duration: 2

Campaign data comes from everywhere: ad platforms, the email & SMS platform, the CRM. Each system has its own names for channels, regions, and audiences, its own date format, and its own copy of the biggest campaigns. So nobody has a single, trustworthy view of what's launching, and teams end up hitting the same audience with competing offers in the same week.

In this hands-on lab you'll fix that for Meridian Stay, a fictional global hotel brand, just as its holiday season plan is going live. You'll land the raw campaign exports in open Apache Iceberg tables, use Snowflake CoCo to clean and standardize them, and surface every audience collision on a live collision heatmap. Then you'll hand the result to the whole team through Snowflake CoWork, so anyone can ask "What's launching in APAC next month?" and get an answer in seconds.

Duration: 3

To complete this lab, you'll need a Snowflake account. A free Snowflake trial account works well. To open one:

1. Navigate to https://signup.snowflake.com/. This link pre-selects AWS and the US East (Ohio) region for you.

2. Complete the first page of the form.

3. On the next section, set the Snowflake edition to Enterprise (Most popular).

4. Confirm AWS – Amazon Web Services is selected as the cloud provider (pre-filled by the link).

5. Confirm US East (Ohio) is selected as the region (pre-filled by the link).

6. Complete the rest of the form and click Get started.

Duration: 2

You're on the marketing operations team at Meridian Stay, and it's November 2026. Last year, the Loyalty team and Brand Marketing both targeted rewards members with competing Black Friday offers, and nobody noticed until the emails went out. Leadership wants to know: is it about to happen again?

The root cause is simple: campaign plans live in three systems that don't agree with each other.

1. Land the three exports, exactly as they arrived, in Apache Iceberg tables.
2. Clean them with CoCo into one governed campaign list, also stored as Iceberg.
3. Find every audience collision: two campaigns aimed at the same audience in the same region on overlapping dates.
4. See where collisions cluster on a live heatmap and put a dollar value on them.
5. Ask planning questions in plain language through a Cortex Agent in Snowflake CoWork.

Everything through step 4 runs from a single notebook (lab.ipynb) and a pre-built app in the companion repo. Step 5 is done in the Snowsight UI. Let's get started!

Duration: 5

There are no setup scripts to run and no SQL worksheets to open first. Your first action is to create a Git-backed Workspace that clones the companion repo. The repo contains the notebook you'll run (lab.ipynb), the three campaign exports (data/), and the collision heatmap app (campaign_timeline/).

Make sure you are signed into your trial account. Confirm your active role is ACCOUNTADMIN: click your name in the bottom-left corner of Snowsight to see the active role. If it shows a different role, choose Switch role → ACCOUNTADMIN.

1. In Snowsight, navigate to Projects » Workspaces.

2. Click the + icon next to Workspaces/Databases → Create new Git workspace.

3. Fill out the modal:
   • Repository URL: https://github.com/Snowflake-Labs/expedition-2026-day-2-hol
   • Workspace name: anything you like (e.g., campaign-planning)
   • API integration: click + / Create new and provide:
     • Name: GITHUB_MERIDIAN_LAB
     • Allowed prefixes: https://github.com/Snowflake-Labs
   • Check Public repository.

4. Click Create.

The Workspace opens with the repo's files visible in the file explorer on the left.

1. In the Workspace file explorer, open lab.ipynb.

2. Click Connect next to the Run button and select Create and connect. Wait for the status bar at the bottom of the notebook to show Connected before proceeding.

3. Set the notebook's active role and warehouse using the role & warehouse picker at the top of the Notebooks editor:

   • Role: ACCOUNTADMIN
   • Warehouse: COMPUTE_WH (the default warehouse in every trial account)

4. Work top-to-bottom through the notebook. Markdown cells explain each step; code cells contain the SQL, some of which you'll generate yourself with CoCo. Run a cell with ▶ (or Shift+Enter).

Open the CoCo chat panel from the Workspace toolbar. You'll send it prompts throughout the lab.

The first SQL cell in the notebook (setup) creates the MERIDIAN_STAY database with three schemas:

It also grants the Cortex Agent privileges your role needs later. Run the setup cell before continuing. You should see "Statement executed successfully."

Duration: 7

The first step is to get the raw exports into Snowflake without changing them, so you always have the original to compare against.

Apache Iceberg is an open table format. An Iceberg table's data is stored as Parquet files plus open metadata, so other engines (Spark, Trino, DuckDB, and more) can read the same tables through Snowflake Horizon Catalog. Your campaign data stays open instead of being locked into one tool, which matters when the data science team, an agency, or a partner platform needs it too.

In this lab you'll use Snowflake-managed Iceberg tables (CATALOG = 'SNOWFLAKE' and EXTERNAL_VOLUME = 'SNOWFLAKE_MANAGED'). Snowflake manages the storage for you, so there's no cloud bucket or IAM setup.

Run the raw_tables cell. It creates:

Run the upload_files Python cell. It copies the three CSVs from the repo's data/ folder into the stage. You should see three files with status UPLOADED.

COPY INTO is Snowflake's bulk-loading command. It reads files from a stage and inserts them into a table, and it works the same way whether the target is a standard table or an Iceberg table.

Send this prompt to CoCo. Compare the output against the expected output in the notebook. If they match, click Allow to run it.

You should see 16, 10, and 7 rows loaded. If CoCo's first attempt fails and it retries with a corrected statement, that's normal: it reads the error and fixes its own SQL.

Run the peek_raw cell. It lines up all 33 raw rows from the three exports in one grid, sorted by campaign name so copies of the same campaign sit next to each other. Scroll through and notice:

You can't find collisions until all three systems speak the same language, and that's next.

Duration: 12

Now you'll turn three messy exports into one governed campaign list. This usually takes a marketing ops analyst a day of spreadsheet work. With CoCo, you describe the rules in plain language and review the SQL it writes.

Before building anything, ask CoCo to compare the three sources. It queries the raw tables itself and shows you how differently each system describes the same things.

CoCo should map, for example, fb_ads, paid_social, and Social Media to Paid Social, and NA, US & Canada, and N. America to North America. It should also call out the three date formats, the text budgets, and the cancelled Veterans Day Weekend Blitz. It may mention the Draft Tokyo Business District Launch too; drafts are still planned, so it stays. The full expected mapping is in the notebook.

In the same CoCo conversation, send this prompt. CoCo reuses the mapping from STEP 2 and writes the SQL. Your SQL may not match the expected output in the notebook word for word, and that's fine; the check cell is what tells you it's right. Click Allow to run it.

Run the check_campaigns cell. You should see:

That's 33 raw rows, minus 5 duplicates and 1 cancelled campaign. If your numbers are different, ask CoCo to fix it rather than editing the SQL yourself:

Short on time? Paste the expected output from the notebook into the empty cell and run it.

Run the before_after cell. It shows the campaigns that needed the most cleaning, as they arrived in RAW and as they are now in CURATED:

The same happens for Black Friday Mega Sale (two copies become one), and Veterans Day Weekend Blitz has no "after" row at all. Six messy rows become two clean ones.

A collision is two different campaigns that target the same audience, in the same region, on overlapping dates. That's exactly what happened with last year's Black Friday offers.

Send this prompt to CoCo. Compare the output against the expected output in the notebook, then click Allow to run it. The prompt names the columns so that everyone's table matches the Semantic View and CoWork questions later.

Run the check_collisions cell. You should see 9 rows, with total_collisions = 9 and total_combined_budget = $557,500. If you see 18 rows, each pair is listed twice; ask CoCo to "list each pair only once." The collisions are (A and B may be swapped in your table):

Five of the nine are in November, including the answer to leadership's question: Loyalty Double Points Month and Black Friday Mega Sale are about to hit rewards members at the same time, again.

Notice what's not on the list: the cancelled Veterans Day Weekend Blitz. Without the clean-up step, it would have raised two false alarms against Thanksgiving Getaway Sale and Autumn Weekend Escapes.

Duration: 9

A table of collisions is useful. A view the whole team can scan in five seconds is better. The companion repo includes a pre-built Streamlit app that turns CURATED.CAMPAIGNS into a collision heatmap for the whole holiday season.

1. In the Workspace file explorer, open campaign_timeline/streamlit_app.py.

2. If a banner says "This file looks like a Streamlit app, but is missing configuration", click Convert to streamlit app. The Workspace adds the configuration files the app needs.

3. Click Run. The app runs privately for you on a container runtime and reads straight from MERIDIAN_STAY.CURATED.CAMPAIGNS.

Start with the three numbers at the top: 7 of 15 region and audience pairs are getting competing offers, and the busiest one, North America · Business Travelers, has 3 campaigns live at once, peaking Nov 15 – Nov 28. The caption under it reads 2 teams: Demand Gen, Loyalty + 1 unassigned. Now hover over one of the red cells in the top row: Demand Gen and Loyalty are both aimed at the same business travelers, and so is Amex Travel Partner Promo, a CRM campaign with no owner team. This isn't a one-off; it's a planning problem.

Use the Region filter to see one region's view: everything on the page, including the numbers at the top, follows it. Pick EMEA, and you'll see 1 of 4 EMEA pairs with competing offers. Turn on Show all audiences to see the pairs that have no overlaps. Use Inspect an audience below the heatmap to see every campaign for one row, with its owner, dates, and budget.

The heatmap shows where the plan collides. Leadership will ask how much is at stake. Ask CoCo to bring in the collisions table. With streamlit_app.py open, send this prompt:

Review the changes CoCo proposes, accept them, and click Run again. You should see:

Duration: 18

The data is clean and the heatmap is live. Now make it available to everyone, in plain language. You'll create a Semantic View with CoCo, a Cortex Agent backed by it, and investigate the plan in Snowflake CoWork. This happens in the Snowsight UI; no SQL is required.

A Semantic View describes your data in business terms: which columns are dimensions (things you filter and group by, like region or channel), which are facts (raw values, like budget), and which words people use for them ("territory" means region, "segment" means audience). It's the bridge that lets Cortex Analyst turn a question like "What's launching in APAC in January?" into correct SQL.

1. In Snowsight, navigate to AI & ML → Analyst.

2. Click Create in Workspaces in the top right.

3. Click Create with CoCo. Snowsight opens a new .sv.yaml file, and the CoCo panel asks for a name, a location, and the source tables.

4. Send CoCo this prompt:

5. Allow CoCo to create the Semantic View draft in the Workspace. This creates an editable draft; it does not publish the view yet.

6. Review the draft in the Semantic View editor, one table at a time. Confirm it contains:

   CAMPAIGNS

   • CAMPAIGN_ID as the unique key
   • Facts: BUDGET_USD
   • Time Dimensions: START_DATE and END_DATE
   • Synonyms: REGION (geography, market, territory), AUDIENCE (segment, target group), and CHANNEL (medium, platform)

CAMPAIGN_COLLISIONS

Verified queries

CoCo may also add metrics, such as a total budget, or extra synonyms. That's fine. If one of the items above differs, for example OVERLAP_DAYS landing under Dimensions, ask CoCo to fix it ("Make OVERLAP_DAYS a fact") before publishing.

7. Click Publish in the top right of the editor. In the dialog, confirm Name CAMPAIGN_PLANNING_SV, Database MERIDIAN_STAY, and Schema ANALYTICS, then click Publish.

Cortex Analyst is Snowflake's text-to-SQL engine; it reads the Semantic View to understand your data. The Cortex Agent receives questions, routes them to Cortex Analyst, and writes the answer.

1. In Snowsight, navigate to AI & ML → Agent Studio.

2. Click Create agent in the top right.

3. Configure:

   • Database and schema: MERIDIAN_STAY.ANALYTICS
   • Agent object name: CAMPAIGN_PLANNING_AGENT

4. Click Create.

5. Click Configuration near the top of the agent editor.

6. Under the General tab, set:

   • Description: I am the Meridian Stay Campaign Planning Agent. I answer questions about planned marketing campaigns across every channel, region, and audience, and I flag campaigns that target the same audience at the same time.
   • Example questions:
     • Which campaigns collide over the same audience between November 2026 and January 2027, and how much combined budget is involved?
     • Which collisions are happening in November 2026?
     • What's launching in APAC in January 2027?

7. Under the Instructions tab, set:
   • Orchestration instructions: Whenever you can answer visually with a chart, always choose to generate a chart even if the user didn't ask for one.
   • Response instructions: Give concise, accurate answers for marketing planners. Name campaigns explicitly and include dates, budgets, and owner teams when relevant.

8. Click Tools → Add semantic view.

9. Configure the tool:

   • Service database & schema: MERIDIAN_STAY.ANALYTICS
   • Select semantic view: CAMPAIGN_PLANNING_SV
   • Name: CAMPAIGN_PLANNING_ANALYST
   • Description: Answers questions about Meridian Stay's planned campaigns and audience collisions

10. Click Add, then click Save in the top right.

Since you created the agent through the UI, it's already available in Snowflake CoWork.

1. In Snowsight, navigate to AI & ML → Snowflake CoWork.

2. Select CAMPAIGN_PLANNING_AGENT from the agent list.

3. Ask:

   "Which campaigns collide over the same audience between November 2026 and January 2027, and how much combined budget is involved?"

The agent should list 9 collisions with a combined budget of $557,500.

The first answer tells you that there's a problem. Continue the same conversation to find out where to act, the way a planner would:

1. Focus on this month. Ask: "Which of those collisions are in November 2026? Chart them by region and audience."

   You should see 5 November collisions, and North America · Business Travelers stands out with 3 of them.

2. Look inside the pile-up. Ask: "For North America business travelers in November 2026, list each campaign's channel, dates, budget, and owner team."

   You should see Business Travel Year-End Push (Paid Search, Nov 1–30, $80,000, Demand Gen), Road Warrior Rewards (Email, Nov 16–30, $9,000, Loyalty), and Amex Travel Partner Promo (Email, Nov 9–23, $0). The Amex promo only exists in the CRM, so it has no owner team. That's a real finding: a partner campaign nobody on the marketing side owns.

3. Check the repeat offender. Ask: "Do the Loyalty Double Points Month and Black Friday Mega Sale campaigns overlap? For how many days, and what's the combined budget?"

   They overlap for 11 days (Nov 20–30) with $97,000 combined, so last year's Black Friday problem is set to repeat.

4. Look ahead. Ask: "What's launching in APAC in January 2027?"

   You should see Tokyo Cherry Blossom Preview, Sydney Summer Kickoff, and the draft Tokyo Business District Launch.

5. Turn it into action. Ask: "Draft a short Slack message to the North America Demand Gen and Loyalty teams explaining the November business-traveler overlap and suggesting how to stagger the three campaigns."

Your results may vary in wording and chart choice. Before acting on an answer, review the SQL and the rows behind it, especially for totals and date ranges.

The notebook produced clean, open Iceberg tables. The Semantic View gives them business meaning and the words your team actually uses. The Cortex Agent makes that model available in CoWork, where anyone on the team can go from "is there a problem?" to "who needs to talk to whom this week?" without writing a query or waiting on a report.

Duration: 1

Once you've finished the lab, run the teardown cell in the notebook, or execute the following in a SQL worksheet:

Duration: 1

Congratulations! You took Meridian Stay's holiday campaign plan from three disconnected exports to a shared collision heatmap and an agent anyone can ask, prompting CoCo along the way.