We created LicensePulse after years of working with enterprise software licensing and seeing how difficult it is for engineering organizations to understand software usage, control licensing costs, and prepare for vendor renewals.
LicensePulse exists to help engineering organizations make better software licensing decisions using accurate operational data rather than spreadsheets and guesswork.
The problem we set out to solve
Engineering software is among the most expensive line items a technical organization carries, and among the least measured. A simulation or CAD estate is licensed through concurrency, tokens, and pack stacking — mechanisms that are difficult to model and almost impossible to reconcile against a contract by hand. The data that would answer the question sits in license daemon logs that were never designed to be read by finance, procurement, or leadership.
So the questions that matter get answered by estimate. How many licenses do we actually need. Which of these are idle. What happens if we cut ten percent. What should we agree to at renewal. Each of those is a decision worth a great deal of money, and each is routinely made without the underlying numbers.
Why the waste is not carelessness
The common framing is that organizations waste a third of their software spend on idle licenses through poor hygiene. We think that reading is wrong, and it matters because it leads to the wrong product.
When an organization cannot measure its position, buying a cushion is the correct decision. A denial during a program-critical week costs more than a spare license. An audit finding with penalty multipliers costs far more. The surplus is not negligence — it is an insurance premium, priced under uncertainty, in the dark.
This is why we do not lead with savings. Asking an organization to cut its cushion without first reducing the uncertainty that justified it is asking it to cancel insurance while the risk stays the same. Certainty comes first. Savings follow from it.
The same logic explains something every license administrator knows: peak concurrency is contaminated data. An engineer denied a license once during a deadline will hold a session open for years afterwards. That behavior is individually rational and collectively expensive, and it quietly corrupts the exact number a renewal decision rests on. Separating real demand from defensive habit is part of what measurement is for.
What we built
- A read-only agent, roughly 200 lines of standard-library Python. It runs inside your network, polls your FlexLM, RLM, or Sentinel daemon every five minutes, and pushes summary records out over HTTPS. It opens no inbound port, accepts no commands, and has no remote-execution path. Ask us and we will send you the source to read before you run it.
- Concurrency, not seat counts. Peak concurrent usage, denial events, and utilization by hour of day — because how a solver behaves on night shift is a negotiating position, and one that cannot be argued without the data behind it.
- Numbers that show their work. Every figure traces back to the check-out records it came from. This is not an aesthetic preference. Your number will be challenged by someone whose job is to challenge it, and a figure you cannot defend is worse than no figure at all.
- Output shaped around the renewal. The product's job is to make you the best-prepared party in the room, well before the meeting happens.
What we will never do
Three commitments matter more to us than any feature, and we will write all of them into a contract.
We take no money from software vendors. No referral fees, no partner programs, no marketplace revenue share, no paid certifications. An auditor paid by the audited is not an auditor. This forgoes revenue that is available to us, which is precisely why it is worth stating.
We will never sell your data, and it will never move toward a vendor. Not anonymized, not aggregated, not in a research report. If we ever build pooled price benchmarks, participation will be opt-in, every contributor will benefit from it, and the data will only ever move in the buyer's direction.
We will never build individual surveillance. The same data that shows a product is over-provisioned could show which engineer held a license for four hours. We aggregate to teams and products, and we decline the work that would require anything narrower.
The full list, including how we handle exports, retention, and vendor subpoenas, is on our Trust page.
Where we are today
We would rather you hear this from us than discover it.
LicensePulse is early. There are no customer logos on this site because we do not yet have customers to name, and we have removed the invented statistics that were on these pages before. We are a small team, so we do not operate a 24/7 on-call rotation and we do not sell a service level we cannot measure. We are not SOC 2 certified; the controls are built and the audit begins when a customer requires it. Our security roadmap lists all of this plainly, including the parts that are unflattering.
What we can offer instead is that you are never trapped. Everything exports, in an open documented format, at any time, with no gate. If we were to disappear tomorrow you would keep your data, your history, and your evidence, and you could walk into your renewal without us. We would rather earn a renewal every year than hold anyone with an export button that does not work.
Who we build for
Engineering organizations of roughly 100 to 2,000 people running five or more engineering software vendors across one or more license servers, spending somewhere between $200K and $5M a year on them, where renewals are handled by someone who already has another full-time job.
If that describes your organization, we would like to talk — not to sell you something on the first call, but because our early customers will shape what this becomes, and we would rather build it alongside people who have lived the problem than guess at it.