Dashboard Creation with ShedBox Agent
Create data dashboards using natural languageβno BI tool experience needed.
Describe Your Dashboardβ
You: "Create a sales dashboard with key metrics"
ShedBox Agent:
β Connected to your sales data
β Generating dashboard...
π Sales Dashboard
βββββββββββββββββββββββββββββββββββββββββββββββββββ
β Revenue β Orders β AOV β
β $487,500 β 2,847 β $171.25 β
β β 12% MoM β β 8% MoM β β 4% MoM β
βββββββββββββββββββββββββββββββββββββββββββββββββββ
[Revenue Trend - Line Chart - Last 12 Months]
[Orders by Region - Bar Chart]
[Top Products - Table]
[Conversion Funnel - Funnel Chart]
Would you like to:
1. Customize any chart
2. Add more metrics
3. Schedule daily refresh
4. Export or share
Dashboard Typesβ
Executive Dashboardβ
"Create an executive dashboard with company KPIs"
Key metrics at a glance for leadership.
Sales Dashboardβ
"Build a sales performance dashboard"
Pipeline, revenue, rep performance, forecasts.
Marketing Dashboardβ
"Create a marketing analytics dashboard"
Campaigns, channels, CAC, conversions.
Product Dashboardβ
"Build a product usage dashboard"
Active users, feature adoption, retention.
Operations Dashboardβ
"Create an ops dashboard for daily monitoring"
Orders, inventory, fulfillment, SLAs.
Visualization Typesβ
| Chart Type | Best For |
|---|---|
| Line Chart | Trends over time |
| Bar Chart | Category comparisons |
| Pie/Donut | Part-to-whole |
| Table | Detailed data |
| Metric Card | Single KPI values |
| Funnel | Conversion stages |
| Heat Map | Patterns in data |
| Scatter Plot | Correlations |
Example Workflowβ
You: "I need a dashboard to track customer health"
ShedBox Agent:
What data sources should I connect?
You: "Our PostgreSQL database and Zendesk"
ShedBox Agent:
β Connected to PostgreSQL
β Connected to Zendesk
π Customer Health Dashboard
ββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Health Overview β
ββββββββββββββββββββββββββββββββββββββββββββββββββββββ€
β Healthy: 78% β At Risk: 15% β Churning: 7% β
ββββββββββββββββββββββββββββββββββββββββββββββββββββββ
[Customer Health Trend - 90 Days]
Shows movement between health categories over time
[At-Risk Customers - Table]
| Customer | Health Score | Last Activity | Open Tickets |
|----------|-------------|---------------|--------------|
| Acme Inc | 45 | 21 days ago | 3 |
| TechCo | 52 | 14 days ago | 2 |
...
[Support Ticket Volume by Customer Segment]
Bar chart showing ticket distribution
[NPS by Health Category]
Average NPS: Healthy (62), At Risk (28), Churning (-12)
You: "Add alerts for customers dropping below 50 health score"
ShedBox Agent:
β Added alert configuration
When health_score < 50:
- Slack notification to #customer-success
- Email to account owner
- Add to "At Risk" queue
Generated Pipelineβ
Dashboards are powered by pipelines:
data_sources:
customers:
type: postgresql
connection_env: DATABASE_URL
query: |
SELECT
customer_id,
name,
health_score,
last_activity_date,
mrr
FROM customers
WHERE status = 'active'
tickets:
type: rest_api
url: https://your-domain.zendesk.com/api/v2/tickets
auth:
type: basic
username_env: ZENDESK_USER
password_env: ZENDESK_TOKEN
processing:
join:
sources: [customers, tickets]
left_key: customer_id
right_key: requester_id
type: left
aggregate:
group_by: customer_id
metrics:
- open_tickets: count(status == 'open')
- avg_ticket_age: avg(age_days)
transform:
- operation: add_field
name: health_category
expression: |
CASE
WHEN health_score >= 70 THEN 'Healthy'
WHEN health_score >= 40 THEN 'At Risk'
ELSE 'Churning'
END
schedule:
cron: "0 */4 * * *" # Every 4 hours
alerts:
- condition: health_score < 50 AND previous_health_score >= 50
notify:
slack: "#customer-success"
email: "{{account_owner_email}}"
output:
type: dashboard
refresh: 4h
Sharing & Exportβ
Export Optionsβ
"Export this dashboard as PDF"
"Create a shareable link"
"Send this to stakeholders@company.com every Monday"
Embeddingβ
Embed dashboards in other tools:
<iframe src="https://shedbox.ai/embed/dashboard/abc123" />
Get Startedβ
Create dashboards with natural language.