In short
- The vendor's setup work is rarely the critical path. Assembling a year of evidence is.
- Plan backwards from the sign-off date, not forwards from the kick-off call.
- Name a person and a date per data category before you load anything, or the schedule is fiction.
Configuring a carbon platform is not the part that takes time. Assembling twelve months of evidence is. Entities, periods, users and reporting outputs are set up once and largely up front. What sets the finish date is how fast other people send you their records, and those people have other jobs. So if you need a date you can defend to a board, build it backwards from the day the figure has to be signed, treat data collection as the critical path, and put a name and a deadline against every category before you load anything.
What actually happens, in what order?
Eight stages. Only two are software work, and neither is what you will be waiting on.
| Stage | Who really does it | What makes it quick | What stretches it |
|---|---|---|---|
| Decide the boundary | Owner and sponsor | A simple group structure, one reporting requirement | Joint ventures, leased sites, a mid-year acquisition |
| Configure the account | Owner | Few entities, one currency, one financial year | Several entities with different year ends |
| Map the sources | Owner and data owners | Someone already knows which system holds what | Nobody has ever written the list down |
| Request the data | Data owners | A named person and a due date per category | Requests sent to a shared inbox |
| Load and map | Owner | Exports with units in them | Scanned invoices and landlord-billed energy |
| Reconcile | Owner and finance | Spend that ties back to the ledger | Category totals nobody can trace |
| Review and close | Reviewer and approver | Sign-off booked before it is needed | Waiting for the next board cycle |
| Produce the output | Owner | The required format was known at the start | The format is discovered at the end |
Read down the fourth column. Almost none of it is a vendor problem, which is why two companies buying the same product on the same day finish at different times.
Why does data collection set the date?
Because it is the only stage that depends on people who did not choose to be part of this. The owner is motivated. The data owners are being asked for a favour on top of their actual job, usually for a figure they will never use, so every category becomes a small negotiation that runs on their calendar rather than your plan.
It is also the stage that reveals what you do not have. Electricity billed through a landlord, a depot whose meter reads were never kept, a waste contractor who reports in containers rather than tonnes. None of that shows up in a demo. All of it shows up as soon as collection starts.
Our own customers say this in public. A small business reviewer wrote on G2 in August 2026 that it takes a lot of manual labour to load data, and that once the data is there it works perfectly, but getting it loaded is the challenging part. That is the honest shape of an implementation in this market, ours included.
What actually determines how long yours takes?
Six things, in the order they matter.
Whether the year is closed. Chasing a period that is still open means chasing the same category twice.
How many entities and currencies. Each entity adds a boundary, a data owner set and a consolidation step.
Whether a prior footprint exists. If someone has done this before, the source list already exists and half the schedule disappears.
How scope 3 is being done. A spend-based screen runs off one procurement export. Supplier-specific data means writing to suppliers and waiting for them, which is a different order of magnitude and belongs in its own plan. Our scope 3 consulting work exists mostly because of that gap.
Whether the output format is fixed. Building for a known deliverable is fast. Building for whatever the customer eventually asks for is not.
Whether assurance is involved. An assured figure needs evidence attached to entries as you go, which is cheap during collection and expensive afterwards.
Which parts can run in parallel?
More than most plans assume, and this is where a schedule is actually won. Account configuration can happen while data requests are already out. Scope 1 and scope 2 can be finished before scope 3 arrives. Data owners work independently of each other by definition, so eight categories can be in flight at once if eight people have been asked properly.
What cannot be parallelised is reconciliation and sign-off. Those need the whole picture and the owner's full attention, and they come last. Leave slack there rather than in the middle, because that is where a late surprise costs you the deadline.
So: decide the boundary first because everything depends on it, send every data request on the same day rather than one category at a time, and start loading whatever arrives first instead of waiting for a complete set. Our beginner's guide to building a footprint sets out a sensible order for the categories themselves.
How do you plan backwards from a real deadline?
Start at the date the number leaves the building, then subtract in this order: publication and formatting, sign-off, review and reconciliation, loading, collection, configuration, boundary. Whatever is left at the front is your slack, and if there is none, the plan is already wrong.
If the remaining time is short, triage rather than compress. Scope 1 and scope 2 are fast, evidenced and rarely disputed, so complete them properly. Do a spend-based screen for scope 3 and say in writing that it is a screen. A defensible partial inventory with a documented method beats a complete one built on guesses, and it gives you something to improve against next year rather than something to restate.
What quietly adds time?
Four things, and all of them are avoidable.
Waiting for perfect data. Every category has a version that is good enough to report and improvable later. Holding out for the ideal export is the most common cause of a missed date.
Changing the boundary halfway. Adding an entity after loading has started means revisiting every category already done.
Discovering the format late. A tender or customer format found in the final week can force a recalculation of categories you had already closed.
One person doing everything. A single owner loading eight categories alone is a serial process pretending to be a parallel one.
What does Hedgehog do to shorten this, and where does it not help?
The platform attacks the early stages directly. An AI guide takes you through GHG Protocol setup, inventory building, data upload and reporting, and you can chat with it to identify your data sources, data owners and documents, which is exactly the source-mapping step most first implementations improvise. Human GHG experts are reachable in-app when the guide runs out. Entity management covers locations and sites with roles for data owners, auditors and managers, so collection can genuinely run in parallel instead of through one inbox. A free account needs no sales call, and Pro starts at EUR 1,200 a year.
Three honest limits, because they all affect a schedule.
Loading is still manual work. The August 2026 small business review quoted above is ours, and we have not solved it. Roles spread the effort. They do not remove it.
There are few integrations today. A mid-market reviewer said on G2 in June 2026 that they would like more integrations with other software. If your plan assumes emissions data flows automatically out of an ERP, test that before you build a schedule on it.
Product footprints are a separate track. The platform does organisational footprints. LCA, EPD, MKI and PCF work is delivered as consultancy with its own timeline, so do not fold a product study into an organisational implementation plan.
What should you do first?
Before you compare vendors, write two lists: every emissions category you need, and the person who holds the records for each one. That single page is what an implementation schedule is actually made of, and most companies build it three weeks late.
Then test the worst category on real data rather than in a demo. You can start a free account with no sales call and find out what your evidence actually looks like. If you are still choosing between products, our guide to choosing carbon accounting software covers the criteria, and you can book a meeting for a second opinion on a date you have already promised someone.
Sources: Hedgehog platform, Hedgehog on G2. Verified 27 August 2026.
Facts on this page were last verified on 2026-09-17.
