Quickstart: Create and Launch an Experiment

This guide walks you through the full experiment lifecycle using the Mida API — from creation to reading results. Copy-paste the curl commands and replace the placeholder values.

You'll need:

• Your Project Key — from Dashboard → Settings → API
• Your API Key — from Dashboard → Settings → API Keys
• Your region — us or eu

---

Step 1 — Create the experiment

Create an experiment in status 9 (draft). The experiment is saved but not yet running.

curl --request POST \
  --url "https://api-{region}.mida.so/v2/project/YOUR_PROJECT_KEY/create-experiment" \
  --header "Authorization: Bearer YOUR_GENERATED_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "test_name": "Homepage CTA Button Color Test",
    "url": "https://example.com/",
    "variants": [
      {
        "name": "Control",
        "data": []
      },
      {
        "name": "Red CTA Button",
        "customCSS": ".cta-button { background-color: #FF5733 !important; color: #fff !important; }",
        "customJS": "",
        "data": []
      }
    ],
    "primary_goal": {
      "goal_name": "CTA Button Click",
      "goal_type": "clickOnElement",
      "goal_value": ".cta-button"
    },
    "configuration": {
      "traffic_allocation": 100,
      "confidence_interval": 95
    }
  }'

Response:

{
  "success": true,
  "status": 9,
  "test_id": 28222,
  "company_id": 1001,
  "goal_profile_id": 8936,
  "secondary_goals_count": 0,
  "warnings": []
}

Save the test_id — you'll use it in every subsequent step.

Variants explained

• The first variant is always the Control (the original, unmodified page). Its data array must be empty.
• Add one object per variant you want to test. Use customCSS and/or customJS to inject changes.
• data: [] is required on every variant (even if empty). It holds visual editor DOM changes.

---

Step 2 — Launch the experiment

Set status to 1 (live) to begin splitting traffic between variants.

curl --request PATCH \
  --url "https://api-{region}.mida.so/v2/project/YOUR_PROJECT_KEY/experiment/28222/status" \
  --header "Authorization: Bearer YOUR_GENERATED_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{"status": 1}'

Response:

{
  "success": true,
  "test_id": 28222,
  "status": 1
}

The experiment is now live. Visitors to the test URL will be split between Control and your variant.

---

Step 3 — Check results

After visitors have arrived and conversions have been tracked, retrieve per-variant results:

curl "https://api-{region}.mida.so/v2/project/YOUR_PROJECT_KEY/experiment/28222/result" \
  -H "Authorization: Bearer YOUR_GENERATED_API_KEY"

Response:

{
  "success": true,
  "test_id": 28222,
  "test_name": "Homepage CTA Button Color Test",
  "status": 1,
  "days_running": 7,
  "total_visitors": 2980,
  "total_conversions": 268,
  "variants": [
    {
      "variant_id": "0",
      "name": "Control",
      "visitors": 1500,
      "conversions": 120,
      "conversion_rate": 8.0,
      "is_control": true
    },
    {
      "variant_id": "1",
      "name": "Red CTA Button",
      "visitors": 1480,
      "conversions": 148,
      "conversion_rate": 10.0,
      "improvement": 25.0,
      "is_control": false
    }
  ]
}

The improvement field shows relative lift vs the control. Here, the Red CTA Button variant converts 25% better than Control.

---

Step 4 — Deactivate the experiment

When you have a winner, deactivate the experiment with status 0:

curl --request PATCH \
  --url "https://api-{region}.mida.so/v2/project/YOUR_PROJECT_KEY/experiment/28222/status" \
  --header "Authorization: Bearer YOUR_GENERATED_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{"status": 0}'

Reactivating

Setting status 0 stops data collection but preserves all results. You can restart the experiment at any time by setting status back to 1.

---

Common patterns

Filter results by date range

curl "https://api-{region}.mida.so/v2/project/YOUR_PROJECT_KEY/experiment/28222/result?start_date=2024-01-15&end_date=2024-02-15" \
  -H "Authorization: Bearer YOUR_GENERATED_API_KEY"

Create experiment with a pre-existing goal

If you've already created a goal and have its key, reference it directly:

--data '{
  "test_name": "Pricing Page Headline Test",
  "url": "https://example.com/pricing",
  "variants": [
    { "name": "Control", "data": [] },
    { "name": "Variant 1", "customCSS": "h1 { font-size: 2.5rem; }", "data": [] }
  ],
  "primary_goal_key": "cta_button_click"
}'

Create experiment and launch immediately

Pass "status": 1 in the create request to skip the draft step:

--data '{
  "test_name": "Homepage Hero Test",
  "status": 1,
  "url": "https://example.com/",
  "variants": [...]
}'

Prevent duplicate experiments (idempotency)

Pass a unique idempotency_key to safely retry without creating duplicates. Same key returns the original response for 24 hours:

--data '{
  ...
  "idempotency_key": "deploy-2026-04-01-homepage-cta"
}'

---

Status codes reference

Code	Meaning
9	Draft — saved, not running
1	Live — actively running
0	Inactive — stopped, data preserved, can be restarted

---

Goal types reference

When setting a primary_goal inline, use one of these goal_type values:

goal_type	What it tracks	goal_value example
clickOnElement	Click on a CSS selector	.buy-btn
clickOnText	Click on an element with matching text	Add to cart
formSubmit	Form submission	form#signup
pageview	URL partial match	/thank-you
pageviewExact	Exact URL match	https://example.com/thank-you
pageviewWildcard	URL with * wildcards	/products/*/confirm
pageviewRegex	Regex URL pattern	^/checkout/.+/success$
script	Custom JS event trigger	purchase_complete

---

What's next

• Create Experiment — full reference for all fields including targeting, secondary goals, and advanced configuration
• Update Experiment Status — all status transitions and constraints
• Get Experiment Result — full result response schema
• Create Goal — create reusable goals to share across experiments