★★★★★
The automation was set up in a clean and practical way by the Checkout Loom team. It saved me a lot of manual work and made the whole process much easier.
JanoUnited Kingdom
Verified
AI sales reporting
AI sales reporting for online stores that gives you one agreed set of numbers and tells you why they moved
AI sales reporting for online stores begins with the dull work that makes everything else believable, which is agreeing what revenue, margin, conversion and repeat rate actually mean, because the platform, the ad accounts and the accountant each count them differently and a report built on competing definitions starts arguments instead of decisions. We pull orders, refunds, ad spend and costs into one set of definitions, tie the totals to your ledger, produce daily and weekly sales reporting templates with commentary that explains what changed and why, and alert you when a number moves enough to matter.
- Fixed price from $1,920, paid once
- One definition for every metric, tied to your ledger
- Plain commentary on what moved and why
- Platforms supportedShopify, WooCommerce, BigCommerce and marketplaces, with Google Ads, Meta, GA4, Klaviyo and your accounting system as sources.
- Typical storeStores with a few hundred to many thousands of orders a month, several channels and a person who builds the report by hand each week.
- Report typesDaily flash, weekly performance, monthly management pack, product and channel reports and a sales reporting template for each.
- Stores coveredOne store group per plan, with extra stores, channels and languages available as priced options.
- DeliveryEmail, Slack, a shared sheet or Looker Studio, chosen by how your team reads reports.
- Live inThree to five weeks for the first reports and up to eight weeks for several channels and a management pack.
The costly disconnect
Why sales reports take hours to build and still do not settle the argument about the numbers
Most stores have plenty of data and little agreement, because Shopify, GA4, the ad platforms and the accounting system each define revenue, orders and attribution in their own way, and the report becomes whatever someone assembled on Monday morning.
Competing definitions
Gross sales, net sales, revenue after refunds and revenue after tax are all called revenue in different tools, and a report that mixes them is wrong in a way nobody can see. Many owners only notice this problem when a customer complains, so catching it earlier is one of the quiet benefits of sales reporting for online stores.
Analytics that do not match the store
GA4 and the store count orders differently because of consent, ad blockers, attribution windows and bot filtering, so a conversion rate from one divided by sessions from the other is unreliable. Left alone, this problem repeats with every order, which is why sales reporting for online stores is built to handle it by rule and not by memory.
Reports that arrive late
A weekly report built by hand arrives on Wednesday and describes last week, so decisions run a week behind the data. It is easy to overlook until a busy week exposes it, and by then the cost to online stores is already counted in refunds, hours or lost sales in the sales reporting work.
Changes noticed too late
A drop in conversion or a spike in refunds is visible in the data on the day it starts, and in the report only on the day someone looks. The fix is not more effort but a better rule, and that is exactly what the sales reporting build for online stores puts in place.
Numbers without explanations
A table of figures shows that revenue fell but not that a campaign paused, a product sold out or a payment method failed, so the reader has to investigate. Many owners only notice this problem when a customer complains, so catching it earlier is one of the quiet benefits of sales reporting for online stores.
Templates nobody trusts
A sales reporting template copied between spreadsheets accumulates errors, and each person edits their own version. Left alone, this problem repeats with every order, which is why sales reporting for online stores is built to handle it by rule and not by memory.
- 9hours a week is a typical saving for a store that replaces a hand-built weekly report with an automated one, indicative and varying by store.
- 1definition per metric is written down and applied everywhere, so a number means the same in every report.
- 100%of reports are tied to the ledger within a stated tolerance, with every difference explained.
Sources: Shopify, Google Analytics 4 and ad platform documentation on metric definitions and attribution; time savings come from stores we have worked with and are indicative.
What changes in your decisions once the numbers are agreed, timely and explained
You open a report that is already built, trust its figures and read what changed and why before you ask.
How sales reports work today
- Reports are assembled by hand each week.
- Tools disagree on revenue and conversion.
- Changes are noticed after the month ends.
- Figures arrive without causes.
- Templates are edited by many people.
How reporting works afterwards
- Daily and weekly reports arrive on schedule, built from the sources automatically.
- Every metric has one written definition applied to all sources.
- Alerts fire when a metric moves beyond its normal range.
- Commentary explains the moves from the events behind them.
- One governed template is used for every report.
What the management pack shows
- Hours returned each week.
- Agreement on the numbers.
- Earlier reactions.
- Reports that explain themselves.
- One trusted template.
What the system assembles and explains
- Pulling orders, refunds, ad spend and costs from every source.
- Applying one definition to each metric and tying totals to the ledger.
- Producing daily, weekly and monthly reports from a governed template.
- Writing commentary that links each move to its events.
- Alerting when a metric leaves its normal range.
What the owner acts on
- Agreeing the metric definitions.
- Deciding what to do about each alert.
- Setting targets and thresholds.
- Choosing which reports go to whom.
- Reviewing the commentary for context the data lacks.
Which online stores gain most from automated sales reporting
If someone spends hours each week building a report, or two people quote different revenue for the same day, automation pays back within a month.
Stores with a weekly manual report
The hours saved each week return as time for decisions. Owners in this position often say the biggest change is the time returned each week, and sales reporting is built to deliver that first for online stores.
Stores with several sales channels
A single report across the store, marketplaces and wholesale shows the true total and the real mix. This is the kind of situation where a fixed-price build is easier to justify than a monthly subscription for sales reporting for online stores.
Stores with several decision-makers
One set of definitions ends the argument about whose number is right. We look at this situation in the audit and give online stores an honest view of whether the build is worth doing now or later in the sales reporting work.
Stores with rapidly changing ad spend
Daily reporting and alerts catch spend and conversion shifts quickly. If this describes your business, the sales reporting work for online stores is likely to pay back quickly, and we will tell you plainly if it will not.
Stores with several currencies
Reports convert and label currencies consistently and show local and base figures. Owners in this position often say the biggest change is the time returned each week, and sales reporting is built to deliver that first for online stores.
Stores seeking investment or a sale
Clean, reconciled reporting history shortens diligence and builds confidence. This is the kind of situation where a fixed-price build is easier to justify than a monthly subscription for sales reporting for online stores.
The data events and schedules that build each report
Reports are built on schedules and on events, and every figure is traced back to its source records.
orders_syncedOrders are synced
Orders, lines, discounts, shipping and tax are loaded daily and checked against the store’s own totals. A short summary of how often this event occurs appears in the weekly report, giving online stores a clear view of how busy this part of the sales reporting work is.
refunds_syncedRefunds are synced
Refunds and returns are matched to their orders and dates, so net revenue is correct. Where the event depends on outside data, the data is checked for freshness first, because a stale figure is worse than no figure for sales reporting for online stores.
spend_syncedAd spend is synced
Spend by platform and campaign is loaded, so marketing efficiency is calculated from the same period. Each moment like this has its own rule, so sales reporting for online stores behaves the same way every time and never has to guess what to do next.
daily_reportThe daily flash is built
A short report of yesterday’s revenue, orders, average order value, margin and alerts arrives before the working day. The event is logged with the rule that handled it, which lets you check later exactly what sales reporting did for online stores and why it did it.
threshold_crossedA metric leaves its range
An alert names the metric, the move and the likely causes from the events of the period. Because this trigger is handled separately from the others, a problem with it never delays the rest of the sales reporting work for online stores.
month_closedThe month closes
The management pack is built and tied to the ledger, with differences explained. You decide how much of this runs without approval, and for online stores we usually begin with a review step until the first results look right in the sales reporting work.
One report followed from raw orders to a commented, tied-out summary
This shows how data becomes an agreed number, a commentary and an alert.
- 01
Sources are loaded
Overnight, orders and refunds are loaded from the store, spend from the ad accounts and fees from the processor. If the step cannot finish, the record says why and waits for you, so sales reporting for online stores never moves on with a half-completed task.
- 02
Definitions are applied
Net revenue is calculated as agreed, after discounts and refunds and excluding tax and shipping, and the same rule is applied to every channel. You can see the outcome of this step before the next one starts, which keeps the sales reporting build for online stores under your control throughout.
- 03
Totals are tied out
The day’s revenue is compared with the store’s own report and the processor payouts, and the small difference is explained as timing. The step uses only the facts you supplied, so nothing in the sales reporting result for online stores depends on guesses or on outside sources you did not approve.
- 04
The report is assembled
The governed template fills with revenue, orders, average order value, conversion, refund rate, margin and ad efficiency, each with its comparison to last week and last year. Timing for this step is paced to the limits of your tools, so sales reporting for online stores keeps running calmly even when volume is high.
- 05
The commentary is written
Revenue is down nine percent, and the commentary traces it to a paused campaign, a sold-out best seller and a weekend, with links to each. A check at the end of the step compares the result with the input, which catches the small mistakes that otherwise surface days later in the sales reporting build for online stores.
- 06
An alert is sent
Because the move is outside its normal range, a message names the metric and the three causes, and the owner decides to restore the campaign. Because the steps are separate, a change to one of them, such as a new rule for online stores, does not disturb the others in the sales reporting flow.
Pricing
Price your AI sales reporting
Choose a plan, tell us your sources and how you read reports, and add any extras, and the total and live date update as you go.
What AI sales reporting costs and what raises the price
Our fee is fixed and paid once, while the data warehouse, dashboard tool and model usage are billed by their own providers.
Number of sources
Each additional channel, ad platform or system adds a connection and tests. This is one of the factors we price in writing before work starts, so the sales reporting quote for online stores reflects your real situation and not an average.
Number of reports
Product, channel and cohort reports add templates and checks. If this factor is small for you, it adds little to the price, and the written quote shows exactly how much it did add for online stores in the sales reporting work.
Currencies and stores
Several stores and currencies add conversion rules and tie-outs. You can lower the cost by tidying this part first, and we say so plainly in the plan when the sales reporting work for online stores would benefit.
History to load
Loading and checking several years of history adds time. Anything extra that this factor creates later is quoted before it is built, so the fixed price for sales reporting never grows by surprise for online stores.
Audiences and permissions
Different reports for different people add delivery rules. We explain this factor in the quote in plain words, because online stores should understand what drives the price before agreeing to anything in the sales reporting work.
Commentary depth
Deeper causal commentary needs more event sources and tests. Changes to this factor after the quote are handled as a written change, with its own price, and never as a hidden charge on sales reporting for online stores.
Data warehouses, dashboard tools and model usage are billed by those providers.
Next step
Ready to price AI Sales Reporting for Online Stores?
AI sales reporting for online stores pulls orders, refunds, ad spend and costs into one agreed set of definitions, produces daily and weekly sales reporting templates with plain-language commentary on what changed and why, and alerts you to the movements that need a decision, so numbers are trusted and used.
- One fixed price from $1,920, paid once
- One definition for every metric
- Plain commentary on what moved and why
Three more reporting jobs the same system can do
Once the definitions and sources exist, related reports are small additions.
Product and category performance
A weekly report shows winners, laggards, margin and stock cover by product and category. The job follows the same approval rules as everything else, so nothing reaches a customer of online stores without passing the checks you already trust in the sales reporting work.
Channel and campaign efficiency
Revenue, spend and contribution margin by channel and campaign show where money works hardest. If the job does not pay back, it can be switched off without affecting the rest of the sales reporting build, and nothing needs to be unpicked for online stores.
Cohort and repeat purchase views
A monthly view shows repeat rate and value by first-order month, so retention becomes visible. The wording and the limits for this job come from the same rule table as the main sales reporting work, so one change updates both for online stores.
Which data the reports use and where each fact is the source of truth
A report is only as agreed as its sources, so each metric names the system that owns it.
| Store object | Fields we use | Direction | Source of truth |
|---|---|---|---|
| Orders and refunds | Lines, discounts, shipping, tax, returns | Loaded daily | Store and marketplaces |
| Ad spend | Spend by platform, campaign and day | Loaded daily | Google Ads, Meta and others |
| Costs | Product cost, fees, shipping cost | Loaded daily or monthly | Cost sheet and processor |
| Traffic | Sessions, source, consent state | Loaded daily | Google Analytics 4 |
| Definitions | Metric, formula, owner, version | Read before every build | Definition sheet |
| Reports | Figures, commentary, alerts, tie-out result | Written on every run | Report store |
Where analytics and the store disagree, the store’s own order data wins for revenue and orders, and analytics is used for traffic and behaviour with its limits noted.
Edge cases in sales reporting that we test before launch
Reports mislead at the edges of time, currency and definition, so we test those edges against your own history.
Refunds in a later period
A refund issued next month reduces that month’s net revenue and is flagged on the original order, so periods stay consistent. We list the cases that matter most for sales reporting in your handover guide, so anyone on your team can understand how online stores deals with them.
Currencies and exchange rates
Revenue is shown in the original currency and a base currency at a stated rate, with the rate source named. A rule for this case can be tightened or relaxed in minutes in the rule table, without waiting for a rebuild or a new quote in the sales reporting build for online stores.
Bundles and gift cards
Bundles are split into components, and gift card sales are recorded as liabilities until redeemed where your books require it. Handling the case early protects the margin and the reputation of online stores, which is worth more than the hours it takes to write the rule in the sales reporting work.
Late-arriving data
Data that arrives after a report was sent updates a restated figure, and the change is shown in the next report. We test this case before launch on your own records, so sales reporting for online stores has met it before it meets a real customer on a busy day.
Seasonal comparisons
Comparisons use the same weekdays and holidays of the prior year, so a moving holiday does not look like a trend. When the case appears, the workflow follows a written rule and records the choice it made, so you can review it afterwards if you wish in the sales reporting build for online stores.
Attribution gaps
Analytics numbers that depend on consent are labelled as partial, and the store’s data is used for totals. If a rule cannot settle the case safely, it goes to you with the record attached, and sales reporting for online stores waits instead of guessing.
Hand-built reports, a reporting tool and a built reporting system compared
Hand-built reports are flexible and slow, sales reporting software and sales reporting tools give dashboards quickly, and a built system agrees the definitions, ties the totals and writes the commentary.
| Hand-built | Reporting tool | Checkout Loom build | |
|---|---|---|---|
| One agreed definition per metric | – | Partly | ✓ |
| Tied to the ledger with differences explained | Partly | – | ✓ |
| Daily and weekly reports on schedule | – | ✓ | ✓ |
| Plain-language commentary on causes | – | – | ✓ |
| Alerts on unusual moves | – | Partly | ✓ |
| Cost pattern | Staff time | Monthly plan | A single build fee plus the tools you choose |
We use Looker Studio, Sheets or a tool such as Triple Whale or Polar where it fits, and we add the definitions, tie-out and commentary that they lack.
How the reporting system is built and what each part costs
We choose the lightest stack that is accurate at your volume and show the monthly cost beside our fixed fee.
A data layer
A hosted spreadsheet, BigQuery or a managed warehouse for orders, spend and costs, chosen by volume and the need for history. Everything this layer does for sales reporting is logged, so you can see what ran, when it ran and what it cost on any given week for online stores.
Source connections
Shopify, WooCommerce and BigCommerce APIs, Google Ads, Meta, GA4, Klaviyo and your accounting system, joined with official connectors where they exist. We test the layer on your own data before launch, so sales reporting for online stores is proven on real records and not on tidy examples.
A definition sheet
A written list of every metric, its formula, its source and its owner, with versions. If the layer fails or changes its terms, the handover guide explains how to swap it, so online stores is never stranded by one supplier in the sales reporting work.
Report delivery
Looker Studio, Google Sheets, email or Slack, with a governed template for each report. Its role is written in plain language in the handover guide, which means a new team member at online stores can understand it in a few minutes in the sales reporting work.
Commentary and alerts
A model that writes plain commentary from the events of the period and checks it against the numbers, plus alert rules by metric. The monthly price of this layer is written next to our fixed fee before you commit, so the true cost of running sales reporting for online stores is clear from day one.
Make, Zapier or n8n
The workflow layer that loads sources, builds reports, sends alerts and writes the log. You can replace this layer later with another tool, because the rules live in a table you own and are not locked inside any one product in the sales reporting build for online stores.
How we keep reports accurate and on time
A report that is late or wrong loses trust at once, so every build checks its inputs and its totals.
Totals tied before sending
A report is sent only when its totals match the source within the stated tolerance. When something goes wrong in sales reporting for online stores, the failure is named, retried within safe limits and listed for you, instead of vanishing into a log nobody reads.
Schedules with fallbacks
A late source delays the affected figure, not the whole report, and the gap is stated. This safeguard runs on every batch and not only on a schedule, so a problem in sales reporting is caught before it can spread to the next record for online stores.
Restatements shown
A figure that changes after sending is restated, with the change visible in the next report. Each alert names the rule, the record and the next step, which means you can fix the cause from the message alone without opening any tool in the sales reporting build for online stores.
Alerts that name the metric
Each alert gives the metric, the size of the move, the range and the likely causes. We compare the results of sales reporting with your own records on a regular basis, so a small drift for online stores is repaired before it becomes a large one.
Versioned definitions
A change to a definition is recorded, and old reports keep the definition they used. If a check fails, nothing is sent or changed for that record, and it waits in a short list for you, because a delay costs less than a wrong action in the sales reporting build for online stores.
Plain explanations after errors
If a report is wrong, the owner explains in writing what happened and the check that now prevents it. The safeguard is tested during the trial batch by feeding it deliberately awkward records, so online stores sees how it behaves before real customers are involved in the sales reporting work.
Financial data, access and sharing in sales reporting
Reports contain commercial secrets, so access is read-only wherever possible and sharing is controlled.
Read-only connections
Sources are connected with read access, so the reporting system cannot change orders, ad accounts or the ledger. We explain in plain words what each permission allows, so the person approving it for online stores knows exactly what they are agreeing to in the sales reporting work.
Controlled sharing
Reports are sent only to the people you name and are stored in accounts you own. At the end of the engagement you receive a list of every connection that was made, so nothing stays open by accident in the sales reporting build for online stores.
Keys in your accounts
Source keys sit in your accounts and can be revoked at any time. For sales reporting on online stores, access is limited to what the job needs and can be withdrawn in a minute, so ending the work leaves nothing behind.
Minimal customer data
Reports use order totals and counts and avoid customer names and contact details. Nothing in this arrangement requires you to share a password, because access is granted from your own account in a form you can revoke in the sales reporting build for online stores.
Named editors
Only named people can change definitions or templates, and changes are recorded. We keep no copy of your keys, records or customer details in systems that belong to us, and the handover guide shows you how to confirm that in the sales reporting build for online stores.
Handover and removal
You receive a list of every connection and report we built, and our own access is removed on the last day. Only the fields that sales reporting needs are read, and only for as long as the task takes, which keeps the exposure for online stores as small as it can be.
What you receive with your AI sales reporting
Each plan ends with agreed definitions, working reports and a template you can keep.
- A written definition sheet for every metric in your reports.
- Source connections for orders, refunds, spend, costs and traffic.
- A daily flash and a weekly performance report on schedule.
- A governed sales reporting template you own.
- Plain-language commentary linking each move to its events.
- Alerts with ranges tuned to your normal variation.
- A monthly tie-out to your ledger with differences explained.
- A short guide to reading the reports and changing a definition safely.
- A month of corrections after the reports go live.
Verified client reviews and the principles behind our reporting builds
The unedited words of verified clients of the Checkout Loom team appear below, followed by the principles every reporting build follows.
- 1definition per metric is applied in every report and every channel.
- 9hours a week is a typical saving, indicative and varying by store.
- 30 daysof corrections follow launch so the reports settle before we step back.
Principles follow platform and analytics documentation, and time savings come from stores we have worked with and are indicative.
★★★★★
I highly recommend ordering from the Checkout Loom team. I had slow responses and poor communication with other providers, but this team was great and I am very happy with their service.
HarmetUnited States
Verified
★★★★★
I am a returning customer because my first AI automation workflow worked so well. Once again everything was built exactly as requested, and the whole process was reliable from start to finish.
WilliamUnited States
Verified
★★★★★
The determination and commitment to quality from the Checkout Loom team were clear from day one. They understood my requirements perfectly and completed everything with professionalism and care, and it is rare to find a team this dependable.
LorcanColombia
Verified
★★★★★
It was an excellent experience from start to finish, and the automation works reliably with everything configured properly. It is clear that the team has strong knowledge of automation, and I will definitely be returning for future work.
EllieUnited States
Verified
★★★★★
The team carefully investigated the problem, tested different scenarios and delivered a dependable solution. A duplicate contact issue had been causing problems for quite some time, and it is finally resolved.
RichardUnited States
Verified
★★★★★
The Checkout Loom team were easy to work with and brilliant at the work itself. I highly recommend them to anyone who needs dependable automation.
IveraAutomation client
Verified
★★★★★
Everything was configured exactly as discussed, and the team made sure the automation was stable before delivering it. It has been a huge time-saver for our whole team, and it runs without any manual work.
BellamyUnited States
Verified
★★★★★
This was excellent automation work, and it is clear that the team has strong experience with AI and automation. I highly recommend the Checkout Loom team to other businesses.
FinnUnited States
Verified
How the reporting build runs from definitions to a trusted weekly routine
We settle the definitions first, because every later step depends on them.
- 01
Collect the reports you use
We gather the reports you build now, the questions they answer and the numbers people dispute. The time this step takes depends on how quickly your records and answers arrive, and we tell online stores early what we need to keep moving in the sales reporting work.
- 02
Agree the definitions
We write one definition per metric with its source and owner and agree them with you in writing. We use plain language in every document from this step onward, so anyone at online stores can read the plan and understand it in the sales reporting work.
- 03
Connect the sources
We connect orders, spend, costs and traffic with read access and test each load against its own totals. A short summary of the step is added to the handover guide, so the way sales reporting was built for online stores is never lost.
- 04
Build the templates
We build the daily and weekly templates and tune the alert ranges on your history. You see the result of this step in writing before the next one begins, which keeps the sales reporting build for online stores on schedule and under your control.
- 05
Run in parallel
We run the automated reports beside your manual ones for a few cycles and explain every difference. If this step shows a problem, we say so at once and propose the fix, and the date for sales reporting is moved in writing rather than quietly for online stores.
- 06
Switch over and hand over
We retire the manual report, hand over the guide and the definitions with a month of corrections and remove our access. Nothing in this step touches your live customers, so online stores can review it calmly and ask for changes before anything goes public in the sales reporting work.
Estimate the reporting time and cost automated sales reporting could save
Enter how long reports take now and the estimate shows the hours and the cost an automated report replaces.
Hours spent on reports each month today–
Hours saved each month–
Value of the time saved each year–
The estimate counts only building time, and the value of faster, trusted numbers is on top of it.
AI sales reporting questions, answered in full
These are the questions store owners ask about templates, tools, definitions, accuracy and cost before they order sales reporting.
What is AI sales reporting for online stores?
It is an automated system that pulls orders, refunds, ad spend and costs into one set of agreed definitions, builds daily and weekly reports from a governed template, writes plain commentary that links each move to its causes and alerts you when a metric leaves its normal range.
Why do my sales numbers differ between tools?
Each tool defines revenue, orders and attribution differently, and analytics tools such as GA4 also miss sessions because of consent and blockers, so we write one definition per metric, use the store’s order data for revenue and orders and label analytics figures with their limits.
What is a good sales reporting template?
A good sales reporting template shows revenue, orders, average order value, conversion, refunds, margin and marketing efficiency with comparisons to last week and last year, and it uses the same definitions every time. We build one for you that nobody can accidentally edit.
How is this different from sales reporting software?
Sales reporting software gives you dashboards on top of your data, and our build also agrees the definitions, ties the totals to your ledger and writes the commentary. We use dashboard tools for display where they fit.
Which sales reporting tools do you use?
We use the lightest sales reporting tools that are accurate at your volume, such as Google Sheets, Looker Studio, BigQuery or a store analytics product, and we compare their monthly cost for your data before choosing.
Does this replace sales reporting jobs?
It changes them, because an analyst or manager spends less time assembling reports and more on explaining and acting, and for a small store it often removes the need to hire for sales reporting jobs at all in the first year.
How accurate is the commentary?
The commentary is written from the events and numbers of the period, each statement is checked against the data and a person can review it, and where a cause cannot be established it says so instead of guessing.
Can reports include several stores and marketplaces?
Yes, stores and marketplaces can be loaded into one data layer with consistent definitions and currencies, and shown separately and combined.
When will the reports be live and what do you need from me?
The first reports are usually live in three to five weeks, after a parallel run. We need read access to your store, ad accounts and accounting system, copies of the reports you use now and a person to agree the definitions, and we send a written checklist after you order.
What will warehouse and dashboard tools cost me each month?
Our fee is charged once, while the warehouse, dashboard tool and model usage are billed by their providers, and we estimate the monthly total for your data volume before you decide.
Open one report you trust and know why your numbers moved
Choose a plan in the pricing calculator and continue to checkout, or request a quote with your store address, the reports you build now and the systems that hold your data. We reply in writing within one working day with a confirmed start date and a short checklist, so we can begin by collecting your current reports.
- One fixed price from $1,920, paid once
- One definition for every metric
- Plain commentary on what moved and why