@pinelab/vendure-plugin-metrics
v1.7.0
Published
Vendure plugin measuring and visualizing e-commerce metrics
Downloads
909
Readme
Vendure Metrics plugin
Official documentation here
A plugin to visualize your shops most important metrics of the past year.
Getting started
- Configure the plugin in
vendure-config.ts
:
import { MetricsPlugin } from "@pinelab/vendure-plugin-metrics";
plugins: [
...
MetricsPlugin.init({
// Consider displaying fewer months for shops with a lot of orders
displayPastMonths: 13
}),
AdminUiPlugin.init({
port: 3002,
route: 'admin',
app: compileUiExtensions({
outputPath: path.join(__dirname, '__admin-ui'),
extensions: [MetricsPlugin.ui],
}),
}),
...
]
- Rebuild your Admin UI
- Start your Vendure server and login as administrator
- You should now be able to add the widget
metrics
on your dashboard.
Metric results are cached in memory to prevent heavy database queries every time a user opens its dashboard.
Built-in Metrics
- Revenue (per product): The total revenue per month, or the revenue generated by specific variants if a variant is selected.
- Average Order Value (AOV): The average of
order.totalWithTax
of the orders per week/month - Units sold: The number of units sold for the selected variant(s).
Custom Metrics
You can implement the MetricStrategy
interface and pass it to the MetricsPlugin.init()
function to have your custom metric visible in the Widget.
// Fictional example that displays the average value per order line per month in a chart
import {
Injector,
OrderLine,
ProductVariant,
RequestContext,
TransactionalConnection,
} from '@vendure/core';
import {
MetricStrategy,
NamedDatapoint,
AdvancedMetricType,
} from '@pinelab/vendure-plugin-metrics';
export class AverageOrderLineValue implements MetricStrategy<OrderLine> {
readonly metricType: AdvancedMetricType = AdvancedMetricType.Currency;
readonly code = 'average-orderline-value';
/**
* Determines if this metric allows filtering by variants
*/
readonly allowProductSelection = true;
getTitle(ctx: RequestContext): string {
return `Average Order Line Value`;
}
getSortableField(entity: OrderLine): Date {
return entity.order.orderPlacedAt ?? entity.order.updatedAt;
}
// Here you fetch your order lines
async loadEntities(
ctx: RequestContext,
injector: Injector,
from: Date,
to: Date,
variants: ProductVariant[]
): Promise<OrderLine[]> {
let skip = 0;
const take = 1000;
let hasMoreOrderLines = true;
const lines: OrderLine[] = [];
while (hasMoreOrderLines) {
let query = injector
.get(TransactionalConnection)
.getRepository(ctx, OrderLine)
.createQueryBuilder('orderLine')
.leftJoin('orderLine.productVariant', 'productVariant')
.addSelect(['productVariant.sku', 'productVariant.id'])
.leftJoinAndSelect('orderLine.order', 'order')
.leftJoin('order.channels', 'channel')
.where(`channel.id=:channelId`, { channelId: ctx.channelId })
.andWhere(`order.orderPlacedAt >= :from`, {
from: from.toISOString(),
})
.andWhere(`order.orderPlacedAt <= :to`, {
to: to.toISOString(),
})
.skip(skip)
.take(take);
if (variants.length) {
query = query.andWhere(`productVariant.id IN(:...variantIds)`, {
variantIds: variants.map((v) => v.id),
});
}
const [items, totalItems] = await query.getManyAndCount();
lines.push(...items);
skip += items.length;
if (lines.length >= totalItems) {
hasMoreOrderLines = false;
}
}
return lines;
}
// This is where you return the actual data points
calculateDataPoints(
ctx: RequestContext,
lines: OrderLine[],
// Variants are given when a user is filtering based on variants in the chart widget
variants: ProductVariant[]
): NamedDatapoint[] {
const legendLabel = variants.length
? `Order lines with ${variants.map((v) => v.name).join(', ')}`
: 'Average order line value';
if (!lines.length) {
// Return 0 as average if no order lines
return [
{
legendLabel,
value: 0,
},
];
}
const total = lines
.map((l) => l.linePriceWithTax)
.reduce((total, current) => total + current, 0);
const average = Math.round(total / lines.length) / 100;
return [
{
legendLabel,
value: average,
},
];
}
}
Contributions
Thanks @dalyathan for his contributions on this plugin.