1. First, if the evaluation is for in-house product building with a view to market the product or for customer needs?
2. In either case, It is very important to understand what processes that you are looking at to automate, based on that you need to select the right product, based on the available bots and the licensing model
3. It is very critical to have a global view of what are the most possible processes that your organization or customer is looking at to automate on a long term, as few products individual license might be cheaper, but as time passes and once you start adding the bots the cost goes up... hence the road map of what is the end goal is very important in selecting a product based on the respective licensing model.
4. Based on the identified processes, feature comparison of the products plays an important role in identifying the right product, as additional features might cost more if not available in the licensing model that you chose for a product
5. Product support and upgrade process is another factor to consider in choosing the product, this is in general with all the software, not very specific to RPA tool
My answer is going to a bit different as I am a business user with limited programming skills
1. Software costs including recurring fees are very important. Automation Anywhere wants to charge a per invoice fee for using IQ Bot (for using OCR), which to some extent defeats the purpose of automation.
2. Training and community support is also critical especially if you have users that do not have programming skills
3. Consulting support - especially if you can have a third party sell the software as a service, where the build and host the bot and you pay them a fee based on usage
4. Troubleshooting and change management especially if you do not have in-house resources to monitor the bots on a regular basis.
5. Cloud based offering that limits your need to have dedicated IT infrastructure.
6. Customer support and the ability to assist users when they are stuck or need some help. Like the 20 minute rule another provider offered - call them if you cant resolve in 20 minutes.
7. Most of the RPA companies have Partners that owuld help you with development - just wish they would do it all. It gets very expensive when you start factoring in the costs of using external consultant to develop bots.
This is a great question and is not a one RPA solution fits all. You must first understand why type of business unit you want to work with first and look at what automation is the most important to the business overall. Implementing a COE is vital, each of the top vendors have a different way of methodology to implement both a COE and their licenses in the beginning, so consulting for a few months is the best way to assist a company in flushing out all of the "what-if" and "unknows" scenarios for a company. A company must have a plan to scale, because once RPA becomes evangelized, RPA will be a viable commodity in every business unit-and many business units rely on each other. A company must look at the customer support they will receive from the company; including technical and business line support. If there is not a consulting firm that is standing by to assist the company in all aspects of their growth, and RPA project is bound to fail. There is change management, training, reskilling, and you have to look at which companies and consulting firms associated with those RPA companies do that best --for the company that is looking to develop an RPA practice. I would also look at whether or not this is a commercial or federal customer. Both have different requirements and engage in RPA and the cloud with security focuses very differently. Many ancillary products used by some RPA companies are not available in the federal space, so you need to know what the agency in the federal space truly wants to accomplish in their timeframe. Does the company work in an agile framework, can they sustain this on their own? Will the consulting firm be training a team? You will have to hire specific people to consult if you are as opposed to folks with a generic background in automation. If the company/agency is looking to scale, you might want to look at building an entire digital transformation team instead of just an RPA team. This is a marathon, not a sprint.
I can only comment on healthcare-related process automation projects.
So, within healthcare you have to define where the automation is needed such as "front office" (e.g. clinical) workflows or "back office" (e.g. coding, billing and RCM) tasks. Since back-office tasks are well managed by UiPath, Automation Anywhere and R1 RCM, let's looks at the front office workflows which are more complex and relate to patient care.
Clinical tasks can range from simple patient scheduling and tracking solutions to semi-complex de-noising tools that improve the quality of low-dose CT scans, or even to implementing AI algorithms to improve the quality and time to delivery of radiation oncology treatment plans.
The above workflows have to be precisely mapped and orchestrated step-by-step which often requires subject matter experts (SMEs) to provide the appropriate level of service. For example, in clinical medical imaging, DICOM, HL7, and FHIR-based digital workflows require serious expertise in order to ensure that mapping is done correctly. Mistakes in process mapping can mean errors in patient care. Not good.
So, let's use the 2021 PwC Automation toolbox as a guidepost, RPA solutions may or may not fit the requirements for healthcare or medical imaging automation. More "intelligent process automation" (IPA) solutions such as those incorporating advanced visualization, AI or machine learning tools may be required. Consequently, mapping the frontend or clinical process from start to finish with all relevant "actors" is critical to deciding what level of automation is required e.g. RPA or IPA.
In my opinion, the extend of the functionalities which the tool can support is the most important thing that we should be looking at when selecting a tool for RPA. When it comes to process automation through the selected RPA tool, the most important aspect that we should consider is the Automation Architecture of the system that we are implementing.
If the architecture goes wrong, the process will be impacted greatly no matter which tool that we have selected for the automation.
Also when selecting the processes to be included in the automation pipeline, the business should always start with a process that is simple and straightforward. After perfectly automating few such processes then the business can move into complex process automation projects.
I think 100% accuracy in the process is the key to nail any RPA integration. This followed by scalability and integration capability are the top 3 aspects I think should be first considered when evaluating RPA solutions.
1. It's important to understand the architecture of the RPA
2. It's critical to evaluate its suitability with respect to the operational or business processes of your organization.
3. It's important to have visibility of implementation beyond POC use cases.
4. It's critical to examine the licensing policies of the tool especially if you are considering its implementation enterprise-wide.
5. Cost of Licencing and the benefit it provides should be analyzed. In short, ROI should be taken into consideration.
6. The requirement of the infrastructure for the tool should be taken into consideration.
7. Availability of initial training on the functionalities as well as possible use cases should be ensured.
8. POC need to be conducted with multiple and dissimilar use cases prior to initiating the buying process.
9. Complete understanding about the support during the initial period as well as post that should be ascertained including the cost of support
10. Evaluate flexibility regarding licensing, upgrades, add-on costs, etc.
1)The functionality should be strong enough to meet the requirement for the business process and the operational activities.
2) The understanding of the tools from the RPA application its compatibility and requirements should be considered.
3) Sales, presales activities should be clear enough w.r.t licensing, purchasing of the bots and orchestrators.
4) Customer Support and Feedback. Training in regards to the difficulties are a plus.
1. Does it meet the requirement of the client?
2. Can it be designed to handle the exception that may error during the execution?
3. Option to monitor after the deployment.
4. Availability / Option to add extra feature after going live.
All the points noted below are valid and important considerations. Additional points I would add:
1. Pre-sales discovery and ROI - for contact centers ROI is difficult to accurately determine as the relationship to the process identified and savings are not linear. Process discovery can take time and the stop watch model is subjective - so having a capability that can help fast track and help determine processes suitable for automation helps with ROI prioritisation.
2. Business and IT need to collaborate. My preference is for Business to own the RPA project so selecting a tool that assists business owners and skills to manage makes sense
3. Whilst RPA is considered tactical in some quarters - tactical can often mean 5 to 10 years (my experience) so focus on agile sprints model.
4. Understand the licensing models - they can become complex
Hello Ariel, it completely depends on the automation requirement, if you need specific process automation, IT or business process automation, end-to-end automation, enterprise-wide automation, etc. Except pricing, we should really think of the technology bundled and scalability. A certain solution like moveworks.com or ayehu.com can only automate IT whereas solution like blueprism.com or uipath.com can only automate the business process. There are few solutions like www.automationedge.com which can do both.
Another aspect of consideration is technology integration. Automation in silos is automation in burden. You need a solution which has a huge set of technology integrations which is the key to scaling automation enterprise-wide.
Next can be adaptability. Not necessary product which is better now will be better and relevant tomorrow. Technology is changing with the blink of an eye and you need to find a solution which adapts to the changing market to stay relevant and provide you with the relevant solution to your problems.
* Through assessment of process that we plan to automate (need to weed out process which does not provide any business benefits)
* Business Objectives- Why do we need to automate?, What are the benefits that we are aiming for?
* Tool evaluation –To ensure that we select the right tool which meets the requirements
* Change Management
* Plan for Scalability
* Start Small
In my opinion, while selecting the right robotic process automation platform, it's important to consider that the process of creating the BOT should be user-friendly so we are not dependent on an external consultant for any minor changes in the process design.
Annual subscription cost as we need to incur it every year.
Each of the below points are valid. It should be ensured that complete functional/technical/commercial details should be obtained at the time of the evaluation. The main concern is licensing fees which need to be paid every year. i.e. your outgoing is fixed irrespective of whether you get the ROI for your initial investments in licensing. As reported only 3% of the organizations that have implemented RPA have reached a level of 50 robots. It seems in most of the organizations RPA implementation does not go beyond POC. However, in many cases these organizations themselves may be responsible as they did not evaluate RPA properly or were swayed away by the so-called USPs of RPA which were pitched by RPA vendors. Vendors should provide clear training and substantial handholding to the Organizations at least up to 25 robots and help these organizations to establish COE so that administering the robots is carried out effectively and risks and compliances of the organizations are appropriately addressed.
- Evaluate the product in the context of the processes you want to automate. Attended vs unattended automation, enterprise vs department level, will it interact well with the applications/systems involved in the process?
- Automation platforms are a gray area when it comes to IT vs Business ownership. Both groups need to understand and agree who is responsible for various aspects of the platform and infrastructure. Make sure your choice of RPA platform is compatible with the approach you're taking.
- If you work in a regulated environment, determine if/how RPA will be evaluated with respect to validation and audit. This can have a huge impact on inception to production timeframes
- Licensing costs and flexibility, especially if you intend to scale.
- Cost of training, support.
- Size of user community and activity. Are there community forums and groups you can join to get help and expand your knowledge?