> ## Documentation Index
> Fetch the complete documentation index at: https://docs.salvidia.com/llms.txt
> Use this file to discover all available pages before exploring further.
# How to collect your data
> Why data collection stalls, the order that gets you to a usable footprint fastest, who in your business holds what, and what year one should realistically look like.
Collection is the hard part. Not the maths, not the standards, not the software. The reason first footprints stall is almost always that someone is waiting on a number from a person who has not replied.
This page is about getting unstuck. The order you collect in matters more than the effort you put in, because the first pass tells you which categories are worth chasing and which are not.
## Why it stalls
Three failure modes account for most of it.
**Collecting alphabetically instead of by size.** Teams work through a list in order and spend three weeks on office supplies while their largest category sits untouched. You cannot know what is material until you have a rough total, which is why the first pass should be fast and incomplete on purpose.
**Waiting for one perfect source.** A landlord who has not replied blocks the entire assessment, because nobody wrote down what to do if the answer never comes. It is always better to estimate the site from floor area, note the assumption, and move on.
**One person chasing everything.** Carbon data lives in finance, operations, facilities, HR, and travel. A single person emailing five departments is a queue with one server. Naming an owner per category is the single biggest speed improvement available.
## Collect in this order
The sequence matters because each step tells you what to do next.
Notice that the first step produces no accurate numbers at all. Its job is to tell you which of the later steps are worth your time.
Connect your accounting platform or import your ledger. Every category you spend money on is now present, most of it estimated. Nothing is missing, which is the point.
You now have something more useful than accurate numbers: you know the approximate size of everything, so you know what to chase.
Electricity, gas, and fuel. These are usually few in number, held by one or two people, and they turn your least reliable estimates into measured figures. For most organisations this is a handful of documents.
Read the first pass and pick the biggest. Do not guess in advance which ones matter, because most organisations are wrong about it. Waste feels significant and is usually small. Purchased services feel trivial and are usually large.
While you still remember why. Every estimate needs its method and its reason recorded at the time you make it. This is the step people skip and the one auditors ask about first.
Note every source that was hard to get. That list is your procurement brief: the data to write into supplier contracts, and the reports to ask your landlord for annually rather than retrospectively.
## Who holds what
Collection goes faster when each category has a named owner rather than a single coordinator chasing everyone.
| Category | Usually held by | Typical form |
| --------------------------- | ---------------------------------------- | -------------------------------------------- |
| Spend across all categories | Finance | Accounting platform or general ledger export |
| Electricity and gas | Finance or facilities | Utility bills, retailer portal |
| Vehicle fuel | Fleet manager or finance | Fuel card statements |
| Refrigerants | Facilities or the maintenance contractor | Service reports |
| Flights and hotels | Travel manager or whoever books | Travel platform or agent report |
| Commuting | HR or People | A short staff survey |
| Waste | Facilities or office manager | Waste contractor reports |
| Freight | Logistics or operations | Carrier reports |
| Leased sites | Property or the landlord | Outgoings statements |
The two that catch people out are refrigerants and commuting. Refrigerant top-ups are invisible in your accounts because you are billed for a service call rather than for the gas, so you have to ask your maintenance contractor directly. Commuting exists in nobody's system at all and has to be surveyed.
## What year one actually looks like
Set the expectation early, because an unrealistic one is what causes teams to abandon a half-finished assessment.
**Days, not weeks, for the first complete draft.** With spend connected, you have a full-coverage footprint quickly. Most of that time is review rather than entry.
**A week or two to upgrade the big categories.** Electricity, fuel, and travel. This is mostly waiting on other people, which is why you start the requests on day one rather than when you get to that step.
**Some categories will not arrive at all.** That is normal and does not invalidate the footprint. Estimate them, document the estimate, and put them on next year's list.
Year two is substantially easier, and not because you get better at it. It is easier because you know which categories matter, you know who holds each source, and the awkward requests have already been made once.
## The rule that keeps you moving
**A documented estimate always beats a blank.** A gap left empty reads as zero, and zero is a claim that you have no emissions in that category. That claim is almost always false and much harder to defend than an approximation with a note attached.
This is accepted practice, not a shortcut. Every standard expects year one to contain estimates and expects them to be disclosed.
## Where to go next
* [What data you'll need for an organisation](/collecting-data/organisation-data)
* [Requesting data from other people](/collecting-data/requesting-data)
* [What good data looks like](/collecting-data/what-good-data-looks-like)