Owning Your EV Charging Data
Installing the chargers is the capital project. Everything after it is a software business: pulling drivers in, keeping them, winning hosts, and finding the offer that brings a driver back to your bay instead of the one down the road. All of that runs on data about your own network, which is why who holds it decides how far the network can go.
Why a charging network becomes a software business the day it opens, when white labelling is the right answer and when it is not, what your data has to include, the three ways operators lose control of it, and how we move you off a platform you are already on.
For An operator planning to scale, or one already on a third-party platform and starting to wonder what leaving would cost.
Work out your exit costAfter the asset is built, this is a software business
The hardware is a one-off. The business that runs on it is not.
A charging site earns from repeat traffic, and repeat traffic is won with software: knowing which drivers came back and which did not, being able to make one of them an offer, being able to show a host what their site actually earned, being able to put something in a driver's hands that the network down the road cannot. A report on their battery, worked out from sessions they were having anyway, is an example of exactly that. It costs nothing per session and no competitor without the data can copy it.
- 1
Pulling drivers in
Visibility, pricing, reliability, and a reason to choose your bay. All decided in software.
- 2
Keeping them
First session to second session is the number that compounds, and it moves on follow-ups and offers rather than on hardware.
- 3
Winning hosts
A host signs with whoever can show them what their site will earn, and renews with whoever can show them what it did.
- 4
Building something others cannot
Added services, personalisation, insight for the driver. Every one of them is built on accumulated data about your own network.
None of that is possible on data you can see but cannot use, and none of it is possible on data you would lose by changing supplier.
What your data has to include
Operators usually think of this as session records. Session records are the smallest part of it, and on their own they are close to useless for anything except invoicing.
- 1
Drivers
Who they are, what they charge, where, how often, what they spend, and what happened when a session failed. The basis of every retention decision you will make.
- 2
Site hosts
Which sites, which units, what each earned, what was invoiced, what settled and the terms behind it.
- 3
Installers and field work
What was commissioned and how, what has been maintained, what was replaced and when.
- 4
The chargers
Asset records, firmware, configuration, fault history and the full session and diagnostic record per unit.
- 5
Tariffs, offers and transactions
What you charged, what you discounted, and what that did to behaviour. An offer with no record of its effect is a cost rather than a campaign.
All five, in a form you can query and export, or you do not have your network's data. You have its invoices.
Three ways operators lose control without noticing
- Export exists but is partial. Sessions come out; telemetry, configuration and fault history do not. You can bill from what you got and you cannot run a network with it.
- Export exists but is a report rather than data. A PDF or a summary CSV is a picture of the data, not the data. It cannot be loaded into anything.
- The chargers themselves are the lock. Even with a full export, if the fleet cannot be pointed at another platform remotely, the data is portable and the network is not.
The third is the one we know from the wrong side. We used to resell a shared third-party platform, and when a change finally became unavoidable the only way to deploy it was to send people to every charger in the field. The data was never the problem. The fleet was.
What ownership means in practice
Ownership is a legal word and it does very little work on its own. These four are what determine whether it means anything.
Swipe to compare
| Test | The question to ask | A weak answer |
|---|---|---|
| Completeness | Which categories are exportable, named individually? | All your data |
| Format | In what machine-readable format, with what schema? | We can provide reports |
| History | How far back, and what happens to data older than that? | The current period |
| Access | On demand via API, or by request to support? | Contact your account manager |
When white labelling is the right answer
Owning your data is not always the correct call, and it is worth saying so.
- 1
A small network, run locally
A handful of sites in one city, a known set of drivers, no plan to expand. White labelling a third-party platform is faster, cheaper and entirely reasonable. The data you are giving up is data you were not going to act on.
- 2
Scaling, or operating across markets
The moment you are planning to grow, operate in more than one country, or build services on top of charging, the position reverses. Every month of accumulated behaviour becomes an asset, and an asset held by your supplier is not one you can build on.
- 3
Somewhere in between
The question to settle early is not what you need now but what leaving would cost in two years. That number only goes up.
We supply both, which is a conflict rather than a coincidence. We say white label where white label is right because an operator sold the wrong thing is an operator who leaves.
Customer data is different
Session and telemetry data are commercial questions. Customer data is a legal one, and the contract cannot override it.
- Establish who is the controller and who is the processor, because that determines who carries the obligations
- Purpose limitation: what the platform may do with your customers' data beyond serving them
- Whether it may be aggregated into anything the platform sells or uses across its other customers
- Deletion on request, and what happens to data already in backups and derived aggregates
- Where it is stored, which increasingly matters for Indian deployments
Private tenancy and self-hosting
For some operators the answer is not a better export clause, it is not being a tenant.
A private tenancy or a self-hosted deployment puts the data in infrastructure you control. It costs more to run and it removes the whole class of questions above. It is worth it where the network is large enough that switching cost dominates, where a customer contract or a regulator requires data residency, or where charging data is genuinely strategic rather than operational. For a small network it is usually over-engineering.
Moving off the platform you are on
Most operators reading this are already on something, and the reason they stay is usually not the platform. It is the belief that moving means downtime, lost history, and a fleet of chargers that have to be visited.
- 1
The history comes with you
Sessions, drivers, hosts, assets and the fault record migrated rather than abandoned, so the network does not restart at zero and the reporting does not have a hole in it.
- 2
The chargers repoint without being visited
Where they can be reconfigured remotely, they are. The failure mode no operator wants is the one we lived through ourselves, which was sending people to every unit in the field to deploy a change.
- 3
The two run in parallel
Chargers move in groups with both platforms live, so the network keeps charging throughout and a group that misbehaves goes back rather than taking the site with it.
- 4
Settlement is reconciled across the cut
Sessions that start on one and finish on the other are the part that generates disputes, and they are handled deliberately rather than discovered at month end.
We do this work as a service, including off platforms that are not ours. It is the most common reason an operator talks to us for the first time.
What we commit to, and why
We were the supplier in an arrangement that served the supplier better than the customer, so these are the terms we wish we had offered then.
- Your data is yours and exportable in full, including telemetry, fault history and configuration, not only sessions
- Machine-readable, via API, on demand rather than by request
- Your fleet can be pointed at another platform remotely, without anyone visiting a charger
- Changes you need are not negotiated against other customers on the same system
If any of that is missing from what you are being offered by anyone, including us, ask why. A vendor with no answer has told you something.
Want this applied to your own site?
Work out your exit costTechnically reviewed by Deepu Joy, Director of Products and Delivery. Last reviewed 2026-09-10.
Frequently asked questions
Our contract says we own our data. Is that enough?
No. Ownership is a legal word. What matters is completeness, format, history and access. A vendor can honour an ownership clause with a PDF report and leave you unable to do anything with it.
What is most often missing from an export?
Telemetry, fault history and charger configuration. Sessions almost always come out, because they are the billing record. The rest is what you actually need to run or migrate a network.
Is data portability enough to switch platforms?
No. If the fleet cannot be pointed at another platform remotely, you can take your data and still not move. Portability of the chargers matters as much as portability of the data. Where both are in place, a typical migration takes about three months.
Should we self-host?
It removes this whole class of question and costs more to run. Worth it for large networks, for data residency requirements, or where charging data is strategic. Usually over-engineering for a small network.
What about derived data like utilisation and health scores?
This is where arguments happen. Establish explicitly whether derived outputs are your data processed or the platform's product. Vendors differ and few contracts address it.
On a third-party platform and wondering what leaving costs?
Tell us what you are on and how many units are in the field. We will tell you what the move actually involves, including where it is not worth doing.
Work out your exit cost