Industry Update
Opinion Article10 January 2020

Experience Feedback – Adopting An Automated Revenue Management System

By Aymeric Erulin, Multi-Property Revenue Manager

share this article
1 minComments

More and more automated RMS are available on the market. Each of them presents different qualities, specificities, strengths, and weaknesses. But apart from those technicalities, the biggest challenge for the decision maker or user is to buy an expensive solution that will make him lose direct control over his revenue management strategy. If you are currently considering this kind of solution, this article is for you. Here are the steps, challenges or milestones I experienced when deploying an automated RMS.


  • Excitement

The new RMS was chosen by our management. For the RM team that was great news, less administrative tasks, more efficiency, and better optimization (a great sales speech indeed). This new system would save the team precious time that could then be wisely used to dig deeper into the data, focus on other revenue streams and develop our skills on the latest available technology.

  • Impatience

For the system to properly work, it required daily data and booking pace. As either our old RMS was not able to provide this information either the new RMS could not read it, we had to wait for the new RMS to be plugged into our distribution system to capture six months of data.

  • Wavering

First pain point: the system configuration. While continuing with our daily tasks we had on top of it to do the initial setup and online training for the whole system. Both of those tasks happened to be time-consuming.

  • The online training was long (approx. 30 hours), with a lot of videos to watch, practice modules and of course, no way to skip any of them. This was mandatory: if the training was not completed, no access to the helpline would be granted.
  • The initial system configuration was even heavier. The UX was not friendly (it has been improved) and the setup required a lot of manual work to construct forecast groups, configure market segments, set planning restriction rules depending on each rate code, configure the price strategy, competitors and many more.

For one hotel, this process would take between one and two weeks on top of the daily tasks.

  • Disillusion

Finally, after succeeding in this first long process, we went through a review phase to make sure that the decision (pricing & planning) the system would decide on, were more or less in line with our expectations or at least an acceptable range. In this phase, we also got our first automatic system forecast. For most of us, it was very disappointing. Let's not forget that everything we heard until them about our new system was only the positive aspects (sales speech) that lead to too high expectations in term of results that the RMS could simply not deliver on its first run.

  • Uncertainty

Autonomous RMS still requires a lot of interaction and inputs from the revenue management team. We learned it the hard way after believing for so long that everything would run smoothly and independently from day one. There are many parameters that RMs need to steer to help the system to take the right (or at least acceptable) decisions. After the first forecast and decision review, we had to go back to the system configuration to fine-tune all the information we already inputted. As an employee, the system has a learning curve that requires RM expertise and guidance to function properly. A lot of tests, reviews and corrections followed before we could finally plug the new RMS to our distribution system. This period is the cornerstone of the whole project. Trust in the system is the base of accepting and implementing autonomous RMS and if either the RM or the hotel start doubting the system too much, you will end up with a very costly system and old fashioned working methods (manual). It is key that as a project manager or software vendor you provide the team with a lot of support during this process.

  • Recovery

Configuring the system is not rocket science, but it requires good training, follow-up, and knowledge of your hotel/market. With the necessary support, the team will finally get to the bottom of it by understanding better how the RMS works and what levers are the right ones for a given situation. Slowly but surely, the forecast steering will improve, the pricing and planning decisions are going more in line with what's acceptable and the trust in the system will increase again.

  • Balance

If the project manager succeeded in keeping his team onboard and confident during the whole process, a new satisfying balance will finally be reached. The daily tasks of our RM team will evolve and they will become more efficient and able to endorse more tasks to develop the revenue.

  • Acceptance

The final challenge will be to keep all revenue management stakeholders onboard. It is not because "we always did it [RM] this way" that what you did was right! You can be very surprised by the decisions of the system but most of the time, even as surprising as they could be, they are the right one. Questioning your strategy and beliefs first is required to fully benefit from the system capacity. At this final stage, you will realize that the future of revenue management is not yet very clear for everyone and that adjustment in the responsibilities and structure are needed. [cf. Revenue Management: Have You Been Doing It Wrong? ]

Main takeaways

  • An autonomous RMS is not a "plug-and-play" software and not fully autonomous. The revenue managers will always have an important responsibility to steer and challenge the system.
  • A good project manager and great support from the system vendor are required throughout the whole process.
  • The success of the projects relies mainly on being able to keep the team convinced that this new system will help them in their daily tasks. Do not oversell the system but remain pragmatic about the pain and gains about the project.
  • It is not because the system's decisions are not what we believe is acceptable that the system is wrong.
  • RMS vendors will never tell you exactly how the system works (it is this knowledge that makes the system profitable). The user needs to accept that sometimes the RMS could be a black-box.

From the beginning to the end of the process, it took us more than a year to reach a new status-quo. Considering the team always remained confident, it seems that a year would be a minimum time frame needed to achieve such a project. For most of us, the biggest challenge has been point 8 when it came to convincing Director of Operations or Hotel Manager that the way we see revenue management and our working methods should completely change.

Aymeric Erulin

    More from Aymeric Erulin
    Aymeric Erulin
    Multi-Property Revenue Manager
    Send email
    Latest News