Dashboards: A Guide to Neuro’s Numbers

If you’ve ever gazed upon a dashboard and felt that the numbers were speaking an obscure dialect, take heart. Neuro’s dashboards are designed to turn operational noise into quiet competence - illuminating orders, fulfilments, plugins, webhooks, and channel inventory so you can catch problems before they do something theatrical.

This guide will waltz you through what each panel means, how the maths is computed, and where to prod when figures behave strangely.

Part I: Filters That Rule Them All

Before we count anything, we must decide which “somewhere” and “sometime” we are counting.

• Date Range: The time window. Only records updated inside this window are included - no time-travellers allowed.

• Comparison Range (optional): Provide a start and end date to see percentage change against the primary window. Useful for “Are we better than last week?” conversations.

• Store (optional): Restrict everything to one store. If you have many stores, choose wisely.

Pro tip:

If a number looks suspiciously small, the Date Range is often the guilty party.

Part II: Overview Dashboard - The Early Warning System

The Overview is your operational seismograph: calm numbers mean business as usual; spikes signal “go investigate”.

Sync Status Panels (by entity)

Each panel totals records updated within the chosen date range, grouped by status. Think of it as traffic lights with more nuance.

Comparison badges (if enabled)

We display percentage change between the current period and the comparison period. If the previous count is zero, you’ll see 100% - because any growth from nothing is, mathematically speaking, infinite enthusiasm trimmed to civility.

Failed Log Trends

For connoisseurs of gremlin-hunting:

• Primary metric: count of failed jobs over time (daily, monthly, or yearly).

• Grouping: by Log Type (e.g., “inventory update failed”, “order acknowledgement failed”) and Source Type (Channel, Plugin, Webhook - including the specific source name).

• Series data: each point shows value (failures that day) and, if a comparison range is set, the parallel compare value and rate of change.

• Totals: sum for the main period, plus comparison delta when applicable.

Interpretation tip:

A sudden spike in “order acknowledgement failed” for a specific channel usually points to credential or endpoint issues. Check the integration settings first; investigate payload second.

Part III: Sales Dashboard - Follow the Money

Here we measure revenue performance across time, channels, countries, and products. Same filters apply: Date Range, Comparison Range, and Store.

Metric Cards (top row)

Each card can show a comparison badge using the same formula as in Overview. If your AOV drops while Orders rise, you’ve likely discounted generously or sold more lower-priced items - rarely a catastrophe, often a strategy.

Revenue Over Time

• Period Type: auto-selects day, month, or year.

• Series: revenue per tick; revenue_compare if comparison is supplied.

• Totals: total_revenue, total_revenue_compare, and total_revenue_compare_rate.

Use this for trend conversations - seasonality, campaign effects, or “did our weekend promo do anything other than excite Marketing?”

Revenue by Sales Channel

• Top 5 channels listed individually with revenue and comparison rate.

• Other channels aggregated into other_channels_revenue with its own comparison.

• Totals include overall revenue and change.

If one channel’s revenue collapses while others remain sprightly, suspect an ingestion or credentials hiccup for that channel rather than a universal bout of customer apathy.

Revenue by Country

The same structural logic as channels, grouped by shipping country (ISO code + name). Wonderful for localisation decisions and shipping policy sanity checks.

Products Revenue (Top Products)

Per SKU:

• Title & Image: pulled from channel listings where available.

• Sales Quantity: value and comparison.

• Revenue: value and comparison.

• Channel breakdown: units, revenue, comparisons per channel.

Sorted by revenue, highest to lowest. If a top SKU’s quantity skyrockets while revenue limps, you may be suffering from discounts, mis-priced variants, or a promotion that forgot the profit margin.

Part IV: Inventory Dashboard - Where Stock Meets Sanity

Predict demand, prevent stockouts, and ensure capital isn’t napping on the shelf.

Inventory KPI Tiles

Percentages show with a % sign; if the denominator is zero, we sensibly display 0%.

Inventory Value Trend

Monthly inventory value for the last 12 months. Each point is the sum of price × quantity for that month; total displays the cumulative amount across the plotted period. Handy for cashflow conversations and “why is our warehouse so full?” musings.

Products to Reorder

Three views, same columns; different stock sources:

• Store Stock: Neuro product quantities (products.quantity).

• Plugin Stock: quantities at connected plugins/3PLs.

• API/Webhook Stock: quantities from external integrations.

Column anatomy:

• Store: store name and UUID.

• SKU / Title: identifiers from products or channel listings.

• Avg Sales Count: average daily sales over the last 30 days = total sales in 30 days ÷ 30.

• Avg Normal / Kit / Replacement (Plugin/API views): same 30-day average, segmented:
  • Normal: direct SKU sales
  • Kit: sales from kit bundles
  • Replacement: sales triggered by replacement rules

• Quantity / Available Quantity: current stock in the source system.

• Time to Out of Stock (TTO): days until stock hits zero = ⌊Quantity ÷ Avg Total Sales⌋. If average sales are 0, we display ∞ (infinite).

• Required Stock: extra units needed for the forecast window = (Avg Total Sales × Prediction Days) − Current Quantity, clipped at 0.

Prediction Days

Defaults to 14 and reflects a gentle, pragmatic philosophy: order enough to cover supplier lead time. Adjust this value in the filter bar; all downstream calculations update instantly.

Sorting & Pagination

Rows are sorted by shortest TTO first - triage made civilised. Pagination lets you browse; exports email you the entire dataset in CSV, including comparison values where applicable.

Part V: Troubleshooting & Good Habits

• Filters first: dates, store, prediction days. Mis-set filters are the mother of all misunderstandings.

• Sudden zeros: verify imports and syncs ran during the selected window (channels, stock feeds, plugins, webhooks).

• Use exports for sharing: they mirror the on-screen columns and preserve comparison values - no screenshots required.

• Investigate spikes intelligently:
  • Failures by Log Type: look for payload changes, endpoint drift, or rate limiting.
  • Channel/Product anomalies: check credentials, listing mappings, and pricing rules.
  • Fulfilment delays: examine queue backlogs and retry policies before blaming couriers (they have feelings, allegedly).

Part VI: Interpreting Comparison Badges Like a Pro

Comparison badges show the percentage change between the current window and the comparison window. Growth from zero displays as 100% - not because we’re cavalier with infinity, but because you need a clear signal that something new happened. Pair the percentage with absolute counts to avoid “100% of very little” euphoria.

Closing Thought: Clarity Is a Feature

Dashboards are not merely decorative scoreboards; they are decision engines. With the right filters and a touch of curiosity, Neuro’s dashboards will tell you what’s humming, what’s hiccuping, and where to intervene so operations glide and revenue sings.

Now, onwards - go make the numbers behave.