Category: White Papers

We recently looked into how long a software’s free trial should last. In doing so, we analyzed numerous competitors who also offer room booking, scheduling, and appointment software.

As part of our research, we were quite surprised at the number of vendors who didn’t display pricing information for their software anywhere on their websites.

In fact, out of the 72 vendors we analyzed, 31% of them didn’t display pricing on their websites.

Software Vendors Displaying Pricing Information On Their Websites
Software Vendors Displaying Pricing Information On Their Websites

Of the 22 vendors who didn’t publish pricing, 10 actually made no reference to the cost of their software at all!

The 12 remaining vendors encouraged those interested in purchasing or subscribing to their software to instead contact them for a price.

Why do some software vendors not display any prices on their websites?

There are a number of reasons why a software vendor might choose not to display prices on their website:

  1. Their software is customized for each customer and the price will depend on the specific requirements of the customer.
  2. The software is sold through a network of resellers or distributors. The final price will depend on the specific terms and conditions agreed upon between the vendor and the reseller.
  3. The software is offered on a subscription basis. The price will depend on the length of the subscription and the number of users.
  4. The vendor is trying to create a sense of exclusivity or scarcity around their product. They believe that by not displaying prices they can generate more interest and demand.
  5. The vendor gives different pricing deals to different customers. If they think they can get away with charging certain customers more than others for the same product/service, they won’t publish pricing on their website.

Let’s take a closer look at each of these…

The software is customized for each customer

All of the software vendors we looked at offered standard “off the shelf” software solutions. By this, we mean that the exact same software product is supplied to multiple customers. Whilst some vendors may offer additional one-off bespoke customizations, there were no indications that any vendor supplied each and every customer with bespoke software.

Here are MIDAS, we don’t understand therefore how this could be a justification for not displaying prices.

Whilst we ourselves occasionally do the odd bespoke customization for customers, any such customization is an additional cost. It does not affect the basic price of our software, and doesn’t prevent us from displaying prices.

The software is sold through resellers

Out of the 72 software vendors we analyzed, we could only find one who didn’t sell their software directly to customers through their own website. Instead, for this vendor a “How to buy” link on their website led to a “Find a local reseller” form.

All other vendors sold directly to end-customers themselves.

Therefore, aside from the one vendor who sells through resellers, this shouldn’t be a reason for other vendors to hide their prices.

The software is offered on a subscription basis and has different pricing tiers

Some subscription based software is offered at a fixed monthly or annual price for all customers. Other vendors may adopt a simple “tiered” approach – offing perhaps a very limited “Basic” package to start with. Followed by “Standard” and “Pro” packages at increasing cost.

Other vendors may offer a wider range of prices based upon specific criteria. For instance, when it comes to booking or scheduling software solutions, the price may depend on the number of bookings that can be made over a given period of time.

However, even if there are multiple pricing points, this shouldn’t be a barrier preventing a vendor from displaying prices. If pricing is dependent upon the number of bookings, users, spaces, etc, it’s not that difficult for a vendor to provide a calculator on their website. That way, customers can see the price they’ll pay for their specific service configuration.

Creating a sense of exclusivity by not displaying prices

By not displaying any publicly visible prices, prospective customers are then forced to make contact with the vendor. This initiates a dialog to between prospective customer and software vendor. During the dialog, the vendor can employ sales tactics to “hook” the prospective customer and convince them that their software is both right for them and they’re getting a great deal.

Giving different “deals” to different customers

Some software vendors may even “inflate” their prices depending upon the type of customer. For example, if a vendor thinks that a particular organization is likely to have a big budget, they may feel that they can get away charging that customer more for the same product that they typically charge another customer with a limited budget.

If the vendor was to display pricing on their website, they wouldn’t be able to arbitrarily charge different prices to different customers.

By not displaying any prices, this gives the vendor the ability to vary the price of their software at will. The customer has no “base price” to reference the price they’re quoted back to.


Ultimately, the decision to not display prices on a website is primarily a marketing and sales strategy. Software vendors may have a variety of reasons for choosing such this approach.

How we approach pricing at MIDAS

Here at MIDAS, we’re totally transparent and upfront about the pricing of our room booking & resource scheduling software.

We clearly display pricing on our website at mid.as/pricing. There’s also a “Pricing” link at the top of every page of our site too.


Like many of our competitors, we offer a no obligation “free trial” of our room and resource scheduling software.

Back in 2017 we increased the length of the free trial periods of our MIDAS room booking system from 14 days to 30.

Six years on, and we were curious to revisit this and try to understand if this is still the most suitable duration to offer for a software trial.

To that end, we began by researching what our competitors currently offer in this regard.

We analyzed 64 online software vendors who develop room booking / scheduling / appointment systems.

The results were quite surprising!

Length of Software Free Trial Periods
Length of Software Free Trial Periods

Nealy 44% of vendors we looked at didn’t offer any form of “free trial”!

Of those who did, the most common length of free trial offered was 30 days, with 18 businesses offering this. 30 days is also the length of free trial that we currently offer for MIDAS.

The second most common length of free trial was 14 days – which is what we previously used to offer for MIDAS. A total of 10 out of the 64 software vendors analyzed offered a 14 day (two week) free trial.

Of the remaining vendors we looked at, the shortest free trial period offered was just 3 days. The longest on offer was 90 days.

What is the best duration to offer for a free software trial?

The best duration for a free software trial largely depends on the complexity and capabilities of the booking system being evaluated. As we found, some software vendors offer trials that last just a few days, while others offer trials that last for several months.

Now some more “basic” scheduling systems may be easy for prospective customers to evaluate within a few days.

Others may require more time to fully test and evaluate all the product’s features and capabilities.

This is primarily the reason why we previously increased our free trial period from 14 to 30 days.

We were finding that some prospective customers were reaching the end of their 14 day trial and were then requesting a 1-2 week extension so they could complete their evaluation. In most cases, we were generally able to extend 14 day trials upon request. However, over time as we added more features and functionality to our booking software, we decided that there was quite a lot to evaluate and digest in just 14 days.

We certainly didn’t want prospective customers to feel “rushed” into evaluating our product.

That’s why we increased our free trial period length to 30 days, to allow extra time for users to get a good feel for what MIDAS is all about.

Since we did this, we’ve had far fewer requests for extensions to free trial periods.

Would an even longer free trial period not be better?

Looking at our our own data, we don’t believe so. On average, those prospective customers who evaluate a free MIDAS trial make their decision within 21 days of the start of their trial period.

That’s why we still feel that offering a 30 day trial is still the best length.

Some vendors offer considerably longer trial periods. One of the reasons for very lengthy trial periods may be because purchasing their solution ties you in to a long and potentially expensive contract, and so potential customers want to take extra time to evaluate their potential long-term commitment.

However, there are potential drawbacks to offering longer trial periods (a couple of our competitors offer 90 day trials for instance);

  1. Firstly, people have limited attention spans and may not be willing to to commit to a long trial period. If a prospective customer can’t get a good “feel” for your product within 30 days, it’s unlikely that they will. (or it’s likely that the software is far too complex for them to be able to readily get to grips with)
  2. Secondly, the computing resources needed. These days, it’s fair to say that the majority of room booking, scheduling, and appointment systems are online. The days of stand-alone, downloadable software packages have all but gone. Therefore, for each free software trial, that system has to be “hosted” somewhere, at a cost to the vendor. A free trial that lasts for 90 days will cost a vendor three times as much as a trial that lasts for 30 days.

What about those vendors who don’t offer ANY free trial?

Going back to our analysis of 64 software vendors, 28 of them (44%) didn’t offer any “free trial” period.

We did however find that 22 (79%) of these vendors instead encouraged prospective customers to “Book a demo” with a sales representative to schedule a live “guided demo” presentation of their software.

Software vendors not offering a Free Trial, but a scheduled "Guided Demo" instead
Software vendors not offering a Free Trial, but a scheduled “Guided Demo” instead

For completeness, we also found that 11 out of the 63 vendors offered both a free trial and a “guided demo” option.

Is a “guided demo” better than a free software trial?

Free software trials and guided demos can both be effective ways to showcase a product to potential customers and help them evaluate whether a product is a good fit for their needs.

Free trials allow potential customers to try out the software at their own pace and in their own time. It allows prospective customers to experience the software firsthand, which can be more effective than just watching a demo.

Guided demos, on the other hand, can be a good way to answer questions and address any concerns potential customers might have about the software.

Ultimately, both free software trials and guided demos can be effective tools for showcasing software to potential customers and helping them decide whether it’s suitable for their particular needs.

Do prospective customers want to see a “guided demo”?

A recent poll on Reddit, revealed that of the 444 people who voted, 300 would want to see a software demo on their first interaction with the vendor. 133 people would want to see a demo after 2-3 interactions, and 11 people wanted a demo after 4 interactions.

Looking at the poll’s comments though, many people want to evaluate software without any interaction with the vendor.

One person commented that “Sales-led demos only exist to give the sales people leverage in the interaction.

Another commented “Software selection is a part of my job description — if I can’t see a demo without needing to put a meeting on my calendar, it’s an automatic no-go.

Whilst another commented “Users should be able to signup for [a] free trial without having to talk to sales“.

We’re in agreement with these sentiments and feel that a “free trial” a customer can explore at their own pace is far better than “guided sales demos”.

In summary…

For prospective customers who insist on a “guided demo”, we can accommodate, however here at MIDAS we prefer to actively promote our FREE 30 day room booking system trial.

We believe that trying a fully-functional copy of our software free for 30 days yourself – with no credit card required, and with obligation to buy – is the best way to get to know our software and to explore its extensive capabilities.

…and our responsive sales team are always happy to answer any questions you have on our software before, during, or after your free trial.


MariaDB vs MySQL

Any software application that “stores” data in some way shape or form, needs a reliable and efficient way to do so. That’s where a database comes in!

MIDAS – our web based room booking and resource scheduling software – is no exception. Since 2012, for its database, MIDAS has supported MySQL. More recently we started supporting MariaDB too.

Last week we announced that we’d migrated all our cloud hosted customers to MariaDB.

In this post we thought we’d look a little bit closer as to the differences between MySQL and MariaDB.

MySQL What is MySQL?

MySQL is a database engine released under a GNU General Public License, and also available under a variety of proprietary licenses.

MySQL was originally owned and sponsored by the Swedish company MySQL AB, which was subsequently bought by Sun Microsystems, and which then went on to be ultimately acquired by Oracle Corporation.

In 2009, when it was announced that Oracle was to acquire Sun, the founder of MySQL wasn’t happy and so “forked” the MySQL project and created MariaDB… more on that later.

Today, MySQL is the second most popular database engine in the world.

Prominent users of MySQL include Facebook, Pinterest, Airbnb, Sony, BBC, Symantec, GitHub, and booking.com.

MariaDB What is MariaDB?

MariaDB is a community-developed fork of MySQL. One of its key attractions is that it is intended to always remain free under a GNU General Public License.

It was created by one of the founders of MySQL, who forked it due to concerns over MySQL’s acquisition by Oracle Corporation in 2009. The concerns centered around rumors that Oracle (who already developed a competing self-titled database product) wanted to kill off MySQL in order to let their own “Oracle” database thrive. In the end, that didn’t actually happen, but that fear was the initial main driving factor behind MariaDB initially.

MariaDB intended to maintain high compatibility with MySQL, ensuring it would be a “drop-in replacement” for MySQL. Whilst that’s still broadly true today, since its inception, the features of the two database engines have slowly diverged more.

MariaDB version numbers initially followed MySQL’s numbering scheme up to version 5.5. As such, MariaDB 5.5 for example offers all of the features that MySQL 5.5 offers.

Specific new features have since been developed in MariaDB, so its developers decided that a major version number change was necessary. MariaDB 10.0 was released in November 2012, whilst MySQL at that time remained at version 5.

New features are added more frequently to MariaDB than to MySQL. MariaDB development is also led by some of the original developers of MySQL, with many other contributors. MySQL on the other hand is now developed almost exclusively by Oracle’s own in-house team.

Prominent users of MariaDB include Google, Mozilla, and Wikipedia.

Popularity of MySQL and MariaDB

Whilst today MySQL is still the second most popular database engine in the world, its popularity has been slowly declining over the years. At the same time, interest in MariaDB (currently the 11th most popular database) has been steadily increasing.

Global interest in MySQL since 2004
MySQL global interest since 2004 – Source: Google Trends
Global interest in MariaDB since 2004
MariaDB global interest since 2004 – Source: Google Trends

A brief history of database evolution in MIDAS

Our web based room booking and resource scheduling software, MIDAS, has been in active development for over 15 years now.

To help ensure maximum server compatibility, in the early days of our software, data was stored in simple CSV (Comma Separated Value) files.

In 2010, we migrated data storage from CSV to XML (Extensible Markup Language) files.

Two years later, in 2012 – and to coincide with the release of MIDAS v4 – we made the switch to a MySQL database.

MIDAS Database Evolution
MIDAS Database Evolution

Read more: MIDAS v4 – Why the database change?

Between 2012 to 2019 we continued to develop MIDAS exclusively using MySQL as its back-end database.

In 2019 we started receiving queries from a couple of customers asking if we’d consider supporting other database engines, as they weren’t big fans of MySQL.

By this time, MariaDB was gaining in popularity, and as it was billed as a “drop in replacement” for MySQL, it seemed a logical choice to explore.

In 2020, we released MIDAS v4.24, with preliminary and experimental support for MariaDB.

In our own development environment here at MIDAS HQ, we ran both MySQL and MariaDB in parallel during 2020. During this time, MySQL remained as the “preferred” database engine we use for development and testing of our software.

At the beginning of 2021, we made the decision to switch our “preferred” development database engine over to MariaDB. (We still continue to run MySQL, but this is mainly just for testing purposes these days).

Why we moved to MariaDB?

MariaDB has a number of advantages over MySQL, including:

  • MariaDB offers improved performance in many scenarios.
  • MariaDB is community-driven.
  • MariaDB has 268 contributors vs 83 contributors to MySQL *
  • MariaDB is in more active development (MariaDB has 193,330 code commits – the latest was today, MySQL has 163,534 – the latest was 3 months ago *).
  • MariaDB is gaining in popularity.

* correct at time of writing

All these factors led us to decide to migrate our development environment over to MariaDB in 2021.

In July 2021, we also seamlessly migrated our cloud-hosted customers over to MariaDB.

Should I choose MySQL or MariaDB?

If you’re considering a self-hosted edition of MIDAS (remember that we also offer a cloud-hosted edition too!), our software currently supports both MySQL and MariaDB databases.

At the end of the day, it will come down to personal preference, but we like to give self-hosted customers a choice. So if you’re looking for a room booking system that’s compatible with either MySQL or MariaDB, choose MIDAS!


Strawberry Perl vs ActivePerl

Perl” is the coding language we develop our web based room booking and resource scheduling software, MIDAS, in.

Most Linux and Mac OS based operating systems come with Perl pre-installed, yet, Windows operating systems do not.

We test MIDAS on a range of operating systems, servers and platforms. Our in-house development of MIDAS is primarily within a Windows-based environment. This means that we needed to install a Perl distribution on Windows.

ActivePerl ActivePerl

When MIDAS development started back in 2005, there was really only one mainstream solution for running Perl on Windows. This was a Perl distribution named “ActivePerl“, produced by ActiveState.

The reason we liked ActivePerl was two-fold; firstly, a completely free “Community Edition” was available. Secondly ActivePerl came with a handy tool called the “Perl Package Manager” (PPM). This made installing and updating Perl modules easy. It provided a graphical interface where modules could be quickly installed, updated, or uninstalled with just a few clicks:

ActivePerl's Perl Package Manager

ActivePerl included a number of “default” Perl modules. MIDAS requires some additional modules not included within the standard ActivePerl distribution. The PPM tool allowed easy and quick installation of any such modules as required.

Many of our “self hosted” customers intended to install our MIDAS booking software on their Windows-based server. Therefore, we would recommend ActivePerl due to its availability, regular updates, and ease of use.

ActivePerl Strawberry Perl

Since 2005, other Perl distributions built for Windows have come along. Perhaps the most notable of these being “Strawberry Perl“, which first appeared in 2008.

Back then we explored what Strawberry Perl had to offer when compared to ActivePerl. After evaluating Strawberry Perl, we decided ActivePerl would continue to be the Perl distribution we developed under and would recommend to our Windows-based customers.

What initially made ActivePerl better than Strawberry Perl?

When we first evaluated the newcomer Strawberry Perl in 2008 against the more established ActivePerl, differences became clear from an ease of installation and use perspective.

Firstly, Strawberry Perl didn’t include a visual “Perl Package Manager”-type tool for installing and maintaining Perl modules. Rather, Perl modules required installation via the command line. On Linux-based servers, installing modules via the command line is the norm, but many of Windows-based users were less familiar with command line use. Consequently, a graphical Windows application which allowed easy installation of Perl modules was preferable.

Another difference was that ActivePerl was established and more stable. Strawberry Perl was still the newcomer and felt a bit “rough around the edges”. Some Perl modules were also not fully supported or failed to install easily/correctly in Strawberry Perl.

As such, because we continued to only recommend ActivePerl to Windows customers, it was logical to continue to develop under ActivePerl ourselves. We would however keep an open mind and keen interest in the ongoing development of Strawberry Perl.

For the most part, our Windows-based customers continued to opt for our recommendation of ActivePerl. A few chose Strawberry Perl instead and were able to initially do so successfully.

Are we POSIXtive?(!)

However, around June 2010, Strawberry Perl suddenly removed a key component from their distribution which MIDAS relied on; the ability to natively work with and set Timezones. This resulted in those running MIDAS under Strawberry Perl seeing “POSIX::tzset not implemented on this architecture” errors. We had no idea why Strawberry Perl removed this functionality, or whether it was just an unintentional bug/glitch in their software. The reason for the removal of this functionality wasn’t forthcoming, or even acknowledged, by the Strawberry Perl team. This led us to initially suspect that perhaps it may have just been a bug.

This wasn’t a major problem for us, as we’d never officially recommended or supported MIDAS running under Strawberry Perl. It was of course though an inconvenience for the handful of customers who had been running under Strawberry Perl.

As a fix wasn’t forthcoming from Strawberry Perl, the solution for affected customers was either to install an older version of Strawberry Perl, or switch to ActivePerl.

By late 2010, it became clear that the developers of Strawberry Perl weren’t going to address/fix this issue. So we re-engineered our MIDAS software to work around this issue. Our next release in January 2011 once again ran under Strawberry Perl without issue.

What we learnt from all this was that Strawberry Perl still felt in its infancy and in a state of flux. We still didn’t consider it “stable” enough for use in production server environments.

We continued to recommend ActivePerl for all our Windows-based customers.

ActivePerl was in continual development, with regular releases which reasonably closely tracked the latest versions of Perl available for Linux-based servers.

Something changed…

In late 2016, we felt things began to shift and change with ActivePerl.

We began to notice that the latest versions of Perl modules stopped being offered via ActivePerl’s Perl Package Manager (PPM). Initially, this wasn’t a great cause for concern. MIDAS didn’t require the very latest version of any Perl module.

The released of ActivePerl 5.26 saw things further decline…

As you may be aware, MIDAS uses MySQL as its database. Perl therefore has to be able to connect to a MySQL database in order for MIDAS to function. The critical Perl module for doing this (DBD::MySQL) wasn’t made available for ActivePerl 5.26 via the Perl Package Manager.

This meant that MIDAS wouldn’t run for customers under ActivePerl 5.26. Customers would instead have to install the previous ActivePerl 5.24 build, which was still available for download from ActiveState.

Now, in the past following a new release of ActivePerl, it could take several weeks for Perl module updates to become available though the PPM.

So we waited… and waited.. yet still no DBD::MySQL module appeared for ActivePerl 5.26. ActivePerl 5.26 became useless for any application like MIDAS which need to connect to a MySQL database!

In our view, ActivePerl declined from there; in order to download ActivePerl an account was now required on their website. Additionally, they discontinued Perl Package Manager. Instead users had to “custom build” their own ActivePerl package in the cloud with the modules they needed, or use a command line tool to add in modules.

ActivePerl’s development started to lag behind Perl itself. For example, at time of writing, the latest official version of Perl available is 5.30.2. The latest version of Strawberry Perl available is also 5.30.2. However, the latest version of ActivePerl available today is 5.28.1 – nearly 2 years behind where Perl is currently at!

Then in 2019 ActiveState’s website was reportedly hacked.

Doubts began to arise over ActiveState’s commitment to continuing to continue to provide Perl and a free “community edition”. Their focus seems to have shifted more towards monetization and on their Python language products instead (as evident from the majority of their recent posts on Twitter)

Why we moved to Strawberry Perl?

These developments were a worrying trend for us. This is why last year we began equally promoting and recommending Strawberry Perl alongside ActivePerl on our server requirements page. We also provided a helpful step-by-step guide for installing Strawberry Perl, to complement our previous guide for installing ActivePerl.

Strawberry Perl has certainly come a long way since its first release. It’s now very stable, is passionately developed, and closely tracks the official version of Perl with frequent releases. Best of all, it remains completely free!

Whilst there’s no “visual” tool to install Perl modules as there was with the PPM under ActivePerl, installing modules under Strawberry Perl is still straight forward. We’ve found that the latest modules are always available (including DBD::MySQL!)

Many previous ActivePerl users around the world have already made the switch over to Strawberry Perl. At the start of 2020, we also moved all our development from using ActivePerl’s distribution or Perl to Strawberry Perl.

Should I choose ActivePerl or Strawberry Perl?

If you’re considering a self-hosted edition of MIDAS (remember that we also offer a cloud-hosted edition too!) for installation on a Windows-based server, whilst we still presently support both ActivePerl and Strawberry Perl on our website we would strongly recommend you choose Strawberry Perl.

If you must use ActivePerl, then we’d suggest getting your hands on 5.24, although this is now four years old, and we’re big advocates for keeping server software up to date. So going forward Strawberry Perl would be our preferred option on Windows.

UPDATE: September 2021

Since we originally published this article in April 2020, we’ve since revisited ActivePerl.