Gajus - Fotolia
The problem with licensing and how to solve it
Licensing can be an expensive business. Niv Lilien explores license management solutions, and how to optimise licenses that you've already purchased
Each CISO knows these situations: sometimes a senior manager needs an operating license for a certain software, and all licenses are taken, a local copyright infringement that might put you at risk, or other licensing expense that has wasted tens of thousands of dollars on a software license that is never used.
Licenses for the average engineering software package are pricy. Using a perpetual license model, the average engineering software package will cost around 15,000 dollars. A cheap one will cost around 5,000 dollars and the expansive ones can go up to 60,000 dollars per license. Multiply that by dozens of units, and you will get a hefty expanse on your IT budget.
Secondary consequence of lack of available licenses, but economically significant to the Organisation just as well, is the downtime cost. The cost of employee who can't run the software he needs is dependent on the geographical location, role, seniority, etc. but with a crude estimate it can go up to a 100 dollars an hour per engineer (including salary, space rent and personal equipment costs). 10 such hours a month will be a thousand dollars lost. Of course, an employee can do other things while he or she waits for a license, but in real life it will usually mean zero productivity, and sometimes it will cause schedule delays for the entire company.
The solution to the problem is license management, optimisation and full utilisation of licenses already purchased, in a constant review process.
While examining license management, we need to take a look at several factors. Long session times, when parallel licenses are being used, is an indication that users are "sitting" on licenses and do not release them to the organisational pool.
Another stat that might seem simplistic, but is indicative of a healthy culture of licenses use, is the number of sessions per day. The higher the number, it means that users use the software correctly – As in using it when needed, and return it to the pool at the end of the session. This is the complement of the duration of the session. Denial of access rate may be negligible in a productive sense, but for the IT manager it is an important indicator. Low number of refusals suggest that software licenses are utilised optimally, whereas a high number of license activation refusals, will indicate an insufficient amount of licenses available.
The bottom line statistic is the general utilisation percentage. This formula is calculated as follows: Total licenses used in a session times the time of a single session divided by the number of licenses, times the length of observation period. Comparing this score of a specific software with the general average score, can give a good indication of the effectiveness of utilising software licenses, but that's not the only measure by which to operate. The other indicators could give an explanation for this index.
In a survey released recently by OpenLM, based on the usage statistics of hundreds of software in its server database, compiled some interesting data. For example, MATLAB, possesses average session time of 9 hours and 17 minutes, as well as a high number of daily sessions (9.4), and has a high license rejection number per hour – 2.06, which creates a relatively low overall efficiency score of 18%.
In contrast, Mathcad, another software that users are in no hurry to release its license – with 19 hours and 14 minutes average usage time, and a daily session amount of 16.4. but with a negligible rate of license activation rejections per hour of 0.01, which indicates a relatively good utilisation and an efficient licensing policy of the manufacturer.
Other popular software in organisations is Solidworks by Dassault Systems, that present different properties, with average session time of two hours and 17 minutes, a low daily session rate of 1.66 uses per day, yet the amount of hourly rejections is 1.03, which results in a rather low general utilisation percentage of 15.62 percent.
To avoid situations of under-licensing or over-licensing, there are software solutions for license management, as a part of a more general ITAM solution or tailored to task solutions, such as Asset Explorer, License Dashboard, or OpenLM.
These solutions can be implemented or managed as software as a service or installed on the client's local network (On Premise), which runs on the workstations in the organisation and monitors the level of efficiency of the actual licenses use. Some of these programs offer the possibility of free trials, so it is worth to install such a solution in order to get an estimate of your organisation's status regarding engineering software utilisation. It might save you on the next yearly budget discussion.