Quarter Replay: How to Analyze Last Quarter the Right Way

Your last quarter is not a grade. It’s a crime scene. If you only read totals, you miss the fingerprints that show repeatable wins. This guide teaches a Quarter Replay that turns messy history into three clear plays for the next 90 days.

Quarterly analysis works when you compare like with like. Build cohorts by first-order month and SKU launch, label time series with events like promos or stockouts, and measure cadence changes such as reorder intervals. Translate the findings into three specific plays with owners, targets, and start dates.

What problem are we solving?

Most quarter-end reviews stop at revenue and margin, then devolve into guesswork. Promotions, stockouts, price changes, and territory moves distort the picture. Without cohorts and event labels, you can’t separate signal from noise—so you fund the wrong things next quarter. Quarter Replay fixes that by standardising comparisons, adding context to every chart, and extracting cadence metrics that reliably predict revenue.

Set the right comparison set

If you want a clean signal, the first job is to compare apples with apples.

Same quarter last year vs trailing three

YoY same-quarter (e.g., Q2 this year vs Q2 last year) controls for seasonality. Great for categories with strong annual cycles and recurring promo windows.
Trailing three quarters (Q-1, Q-2, Q-3 vs QX) highlights momentum and operational change. Useful when you’ve recently changed price, packaging, or go-to-market.

Pro tip: Use both. Start with YoY for seasonality, then sanity-check against the trailing three to catch recent structural shifts.

Build a clean baseline without major events

Before you argue about causes, construct a “no-events” baseline:

Export weekly revenue/orders for the last quarters (at least 6).
Remove weeks with major disruptions: promotions, stockouts, supplier delays, pricing changes, competitor shocks, big territory reshuffles.
Interpolate a baseline trend from the remaining weeks.
Compare actuals to baseline to quantify event impacts (promo lift, stockout loss, price effect).

Now you can say “Q2 underperformed baseline by 6%, 4 points of which came from stockouts,” instead of “it felt slow.”

Cohort views that reveal shape

Totals blur structure.

Cohorts reveal shape—who behaved similarly, and why.

Customer cohorts by first order and reactivation

First-order month: Group accounts by the month they placed their first order. New customers share similar onboarding dynamics, discount exposure, and education gaps.
Reactivation month: Dormant accounts that returned this quarter behave differently from brand-new customers. Track them separately to avoid overstating new-logo momentum.
Segments: Layer SMB/Mid-Market/Enterprise, or industry verticals.
Questions to ask:

What good looks like: A healthy Quarter Replay shows rising second-order conversion and tightening time-to-second-order for recent cohorts.

SKU cohorts by launch quarter

Launch quarter cohort: Group SKUs by the quarter they launched (e.g., 2025-Q1 launches). Early demand curves often follow predictable arcs.
Track adoption by segment (e.g., enterprise adopts slower but retains longer) and attach rate (what do customers buy with this SKU?).
Use this to plan inventory and to decide which SKUs deserve promo support vs normal lifecycle management.

What good looks like: Launch cohorts that hit time-boxed attach rate and repeat rate thresholds without heavy promo dependency.

Event context on every chart

A time series without context tells half the story. Overlay event labels on every view.

Promotions, stockouts, supplier delays, pricing changes

Promotions: Record start/end dates, discount depth, channel, and targeted segments. Measure incremental lift vs the baseline.
Stockouts: Note affected SKUs, backorder policies, and substitution rates. Quantify lost orders and deferred orders separately.
Supplier delays: Capture lead-time spikes and the SKUs affected. Watch for spillover where delays cascade into cancellations.
Pricing changes: Record magnitude and SKU scope. Break the impact into mix vs price vs volume using a simple driver tree.

When you rerun these labels next quarter, you’ll know if a result was repeatable or just ride-along on a one-time event.

Competitor and territory changes

Competitor moves: New entrant, price match, bundle change, or a large account loss/win in a region.
Territory redesign: If account ownership changed, label it. Expect a temporary wobble in cadence while relationships reset.

Pro tip: Keep event labels short, standardised, and reusable. “Promo: 15% sitewide (email, SMB)” is better than “Summer sale blast!”

Cadence metrics that predict revenue

Cadence is the rhythm of your book of business. Get this right, and you can forecast and design plays with confidence.

Reorder interval by segment and SKU

Compute median days between orders by segment (regions, customers, channels) and by SKU family.
Track how the median shifts for the latest customer cohorts vs prior quarters.
Flag cadence breaks: for example, SMB in Region East slipped from a 28-day median to 39 days for the Cleaning family—likely inventory or service related.

Actions tied to signal:

If median is lengthening, deploy a replenishment nudge (email + SDR call) at 70% of the typical interval.
If median is shortening on a new SKU family without promos, consider light availability campaigns or bundle suggestions.

Quote-to-close time and cycle compression

Track median days open for quotes by segment and SKU family.
Identify bottlenecks (legal, credit, out-of-stock).
Measure cycle compression tactics (pre-approved terms, instant samples, guided pricing) against median movement, not just win rate.

What good looks like: A Quarter Replay that converts cadence insights into triggered outreach—not generic “call more” directives.

From charts to plays

Analysis doesn’t pay the bills; plays do. A play = segment/cohort + motion with an owner, a target, and a start date.

Define a play: segment plus motion

Use this template:

Segment/Cohort: Be precise (e.g., “Reactivated SMB, first-order 2024-09, East region”).
Motion: The thing you’ll do (e.g., “Bundle Cleaning + Filters, 10% for 30 days, email → SDR follow-up”).
Trigger: The signal that fires it (e.g., “No order at 21 days for this cohort”).
Offer/Assets: What the rep or workflow needs (script, email, landing page, stock check).
Measurement: The cadence metric you expect to move and by how much.

Owners, targets, start dates

Turn the analysis into a 90-day plan with three plays:

Play #1: Replenishment for new SMB cohort
Play #2: Launch cohort attach-rate push
Play #3: Cycle compression in Enterprise

Document these in a shared tracker (or use the worksheet below) and review weekly.

Step-by-step workflow

You can do this in any BI tool or spreadsheet. To make it simple, grab our template: Quarter Replay Worksheet (Excel). For an example of how our template works, take a look at our pre-filled example sheet.

Set comparisons
Create a baseline
Build cohorts
Overlay event context
Measure cadence
Write three plays

Example questions to interrogate your data

Seasonality vs promo impact: If YoY is flat but trailing-three is up, did a QX promo just mask an underlying slowdown?
Mix vs price: Did margin rise because of mix shift to higher-margin SKUs, or because you raised price? Use a simple driver tree to separate volume / mix / price.
Cohort health: Are 2025-05 first-orders converting to a second order faster than 2025-03? If slower, what changed in onboarding?
Cadence drift: Which region’s reorder interval widened? Does that correlate with stockouts or a service issue?
Launch curve: Which launch cohort failed to hit attach-rate targets without promo support?

Quick checklist

YoY and trailing-three comparisons built
“No-events” baseline computed
Events labelled on every chart
Customer cohorts by first order & reactivation
SKU cohorts by launch quarter
Cadence metrics: reorder interval, quote-to-close, lead-time variability
Three 90-day plays with owners, targets, dates