What needs improvement with Mule Anypoint Platform?

The product is expensive.

With GenAI coming into the picture, it can be used to focus on specific setups and use cases. If you want to have an end-to-end solution, more AI infusion into the product will be beneficial. For example, when it comes to an industry-based solution, and someone approaches our company and states that they want to focus on healthcare or some financial services, we have some specific end-to-end solutions where Anypoint can be used as a backbone. If you are able to enable some AI on top of the functionalities provided by the product, it would be beneficial for users. The inclusion of GenAI in the tool can be good since it is an area that is currently unavailable in the solution. Mule Anypoint Platform can serve as a backbone when trying to build up an end-to-end solution using AI.

Anypoint Studio for developing integrations can be described as a moderate to complex process, and from an improvement perspective, it needs to be made easier. Mule Anypoint Platform has developed a new development studio based on Visual Studio Code, but it is still in the beta stage, so my company hasn't used it yet. I hope in the future, it will be easier to use the tool because the current Anypoint Studio is based on scripts, which are quite heavy. My company hopes that in the future, with Visual Studio Code, the development studio would be easier to use. The high price of the product is an area of concern where improvements are required.

It would be better if we had a clearer view of the solution's future releases. Mule Anypoint Platform is migrating to a new JDK, and the old JDK won't be supported within a year. Currently, we do not have a calendar or scheduler about when all the connectors will support the new JDK 17.

Mule Anypoint Platform is complex for beginners. Users without programming skills will find it complex. It should also improve its pricing.

The main area where improvements are required in the product revolves around budgeting. The cost of the product is an area of concern where improvements are required, especially when compared to other tools, like Dell Boomi or Oracle. Mule Anypoint Platform is made available with many components, and whether users use them or not depends on their choices. For example, though there is a tool called Anypoint MQ, our organization prefers to pay and use Kafka, as we don't want to use MuleSoft Anypoint MQ. In Mule Anypoint Platform, there needs to be proper segregation to help users identify what they need and don't need, making it an area where users need to be careful when opting for MuleSoft. Mule Anypoint Platform is an integration tool and not an ETL solution. When a user has to deal with a huge number of data, then Mule Anypoint Platform should not be a preferred choice since it can only be used for lightweight purposes revolving around APIs. There is a need to have a clear architectural decision made before opting for Mule Anypoint Platform. When there is a requirement for heavy data transformation, then users need to decide whether to go with Mule Anypoint Platform or any other platform available in the market. Though the tool comes with many useful components, it depends on whether the user plans to use all of its features. Recently, the product has come up with a new feature that is similar to an API catalog. MuleSoft had come up with the UAPIM feature almost eight months to a year ago, which had some concerns in the area of budgeting. Whether you build APIs using Javacore or any other tools, they can be cataloged using MuleSoft's UAPIM feature, for which users need to pay an extra amount, making it an area of concern for users. It would be great if MuleSoft's support team could provide help with the area of Kubernetes.

The solution's licensing methodologies could be improved.

In terms of improvement, the deployment time could be a little shorter. Other platforms such as Boomi or SnapLogic are much faster to develop. Anypoint Platform has added the option to generate flow online, but it still needs some improvement. It should give better control over account management. Mule Anypoint Platform includes an API Manager and it has a strong ESB. However, the API management capabilities of Apigee offer more advanced and dynamic features for managing APIs. Additionally, Mule should work on reducing the complexity of their platform and providing more flexibility for users. Those who are not well-versed in Java will have a hard time with the development of Anypoint Platform. While it offers connectors, the connectors have limitations. It should offer features that can be customized by users, allowing for greater adaptability, and offer customization options within these connectors, to make it easier for users to tailor them to specific needs. Platforms such as Dell Omni and SnapLogic already provide solutions for these problems and allow users to easily plug them in and start working.

One area we'd like to see improvement in is the error logging and troubleshooting process. Handling errors can sometimes be challenging, but if Mule Anypoint Platform could allow defining user-defined errors and provide better management for them, that would greatly enhance the platform. Another aspect is the support from MuleSoft. We don't find it easy to get support from their product support team. I'm not sure if it's an issue with our specific team, but the responsiveness from MuleSoft support should be better. We rely on their product for our work, so that's something we hope could be improved.

I think Mule Anypoint Platform is one of the market leaders, so our customers prefer it, which is why I also use it. The solution's user interface has shortcomings and can be made much better. Mule Anypoint Platform can be made more user-friendly so that people with lesser technical knowledge can understand the product. Mule Anypoint Platform can get too complex for non-technical people.

The negative side of MuleSoft's Anypoint Platform is its cost. The solution's pricing model is very strange, and it will be really expensive if you use APIs a lot.

The compatibility with vendors can be improved. Microsoft Azure heavily uses single software. Compatibility with other services or cloud providers like AWS or GCP is not very well defined. This is an area they need to be a little more cautious about. Nowadays, everyone is talking about Generative AI, how intelligence can be brought into different systems that are ready. If we talk about Mule Anypoint Platform, the system and data are there, but the integration layers are not very advanced. They should bring in Generative AI and other AI platforms because even it will need a lot of self-building intelligence and decision-making. They have good competition from cloud providers nowadays. There is a lot of scope and innovation competing in that area.

In order to set up a storefront, we currently rely on a third-party solution. It would greatly enhance our operations if this feature was integrated into their existing solution. Some of the functionalities are hosted in the cloud, which occasionally poses challenges when integrating with on-premises applications. Improving this aspect could be beneficial for a smoother and more seamless experience.

The transforming message is an area in the solution that needs to be improved. There is not much information on the internet, and since MuleSoft is new, there is nothing to research or find in Stack Overflow. They are probably trying to test two layers at once. I don't know how that would be possible. I've been trying to work between something another company left recently. So while testing it, it's been hard because there are a lot of layers, and some of them are not okay. If I could test two locally, that would be great.

I think in the next release if they could work with the low-code platform that developers are struggling with that would be helpful. I would like to see some data integration and automation.

1. Clarity w.r.t. Licensing Model: Mule is a complex licensing model -- There are 4 types of connectors -- Select, Premium, MuleSoft Certified, Community

2. Unclear/incomplete documentation: Like for example in the case of Mongo DB there is no way you can understand the significance of "connection string" based connectivity like LDAP based Authentication or X509 kind of connectivity.

3. High price / vCore; you are an enterprise and have enough money then they can afford Mule; ETL sort of transformations if not understood properly will increase the Memory and core count.

4. Community support and Open Source camaraderie can for sure be improved.

5. Limited configurability on cloud and RTF deployments.

6. Dataweave is another language to learn and transformations are CPU intensive in nature. There is no portability for your scripts or interoperability if I want to port from other languages into Dataweave if I am newly coming onto the cloud;

6. Auto mapping features like in the case of Salesforce connector to other connectors as well.

Mule Anypoint Platform could improve by having better integration with MDM. There are challenges when doing the interaction at this time.

What I hear from my customers is that it's very expensive compared to the cost of other integration suites. The cost keeps increasing. MuleSoft should come up with customization factors by using a different way of getting the cost-related stuff to attract customers. That is, they should come up with some cost optimizations. In the next release, I would like to see deployment automations. I would also like MuleSoft to come up with a central platform because with on-premises and hybrid environments, where some are cloud-based applications and others are on premises applications, a DPC setup has to be done. It would be good if they can implement one centralized user interface that provides all of the facilities.

Code quality, Code Security, SaaS, and DaaS security, can all be improved.

The pricing should be improved, it's an expensive product.

One of the issues I have is that Jira tries to be more than an API-based integration platform and it encourages writing of Java-based applications underneath the cover, even though it's not meant to be an app service, so it's slow. That's how they try to sell it. The solution has three layers; the experience layer, the process layer and the system layer. The system layer is used to connect the systems; the process layer can be used to create processes like transformations, logic, etc.; and the experience layer is used to interface with whoever's going to consume it. The problem is that the process layer is not efficient. It acts like an interpreter and interpreters are slow, and if you give it some path like converting to a 100 kilobyte file, it takes 15 times more memory than if I just wrote a program When you try using it for the processing layer, it doesn't work well. We ended up pulling back to .NET for our own internal microservices that do a lot of processing. We use the .NET microservices as the process layer, because we know it's faster. Then we plug it into Mule to experience the API interface and that's when we need to make a choice, do we put our system APIs right through our microservice, or do we use Mule as a system API layer, but then let it interface with our process API. We just had to make choices.

The solution is very expensive. They need to work on the pricing. Technical information is not always easy to find. They should work to make it more accessible.

One should have hands-on experience on the DataWeave. If somebody knows MuleSoft and is planning to work closely with the solution, they should be comfortable with DataWeave. Pricing is one aspect of the solution that is troublesome. There should be a lot of improvement in general, as this is a premium product, I can say. Most of the customers like to go ahead with MuleSoft implementation, but at the end of the day what is happening is when they go ahead and see the pricing, they get scared. Performance-wise also they need to improve a lot. Basically, the number of ports is lacking. When you have ETL kind of transformation, ideally it will be supported, however, we end up using a large number of ports. That will drastically increase the cost to the customer, however. They need to be able to allow for an increase in ports without expanding the price. Everybody's now integrating with AI and ML kinds of features. They need to add more of these features into their offering as well. The scalability on offer could be better.

Its documentation needed a little bit of work to make it more usable. It is a platform that is used mainly by developers and other people for connecting systems. Its documentation was confusing in some areas and was not very helpful in other areas. I had to go to a consultant to get some work done, which ideally shouldn't be required.

Better documentation, in particular with respect to the initial setup, would be helpful.

I don't know if there are any real pain points per se. The terms of use and how it's priced has become very difficult to manage, which is forcing us to look for alternatives already - and we haven't even been using it for that long just yet. The initial setup is rather complex. There should be more templates associated with a standard to large ERP. There are 18 interfaces required between HighJump to SAP, our warehouse management system. Those templates are standard. Everybody that has HighJump and that has SAP would have to do those 18 interfaces and they should be in a template - yet those templates don't exist. Due to that, there's too much API building from scratch. They should be able to build an API library.

It has different types of subscriptions. For platinum or lower subscriptions, there are not too many things that can be done. We don't see many features. They should release a basic version that has logging and monitoring features. These features should come with Mule Anypoint Platform for free instead of making customers pay separately for these features. Its dashboard can be improved to have a lot of charts so that it is easy to visualize information. The utilization part can be improved. The dashboard is good currently, but it can be better. Other solutions like Elastic have a good dashboard, and they allow you to administer the product from the UI. Currently, for RTF, there is a different dashboard or utility. It would be good to include the same utility in the cloud solution. It would be good if there is a centralized repository that includes the links to the information about various troubleshooting issues. The documentation is there currently, and it is good, but the troubleshooting information is too scattered. We have to go to different links to find troubleshooting information. This kind of centralized repository would be helpful for new customers who are implementing this solution. It will be helpful to see different kinds of issues that can occur.

With regard to the integrations, I'd like to see something similar to the features you find in Dell Boomi related to the intelligent management data and intelligent mappings. Those would be helpful improvements that would help developers. For beginner users with little technical knowledge, some low code would help. The solution is Java oriented but doesn't require Java knowledge which is good. But when you are debugging something or when you have other issues, it's important to know a little bit about that. I'd like to see some simplification. Anything that can be automated in the Data V area would also be helpful.

In terms of improvement, customer support is a bit slow. I emailed the sales team for a pricing comparison, still no response. The developer community is far more generous on the other hand. Pricing model is very expensive, Mule should consider Pay as You go model. It will help on-boarding new clients in Asia region. Mule should have better B2B integration. Mule supports the idea of API Led Architecture and system of records approach. When it comes to broad perspectives such as B2B, competitive tools have better EIPs.

Overall, I think they're keeping pace with the changes that are happening. I don't have any complaints about anything. They could just improve implementing the integration. I would also like to see some kind of monetization of A-P-S. That's important.

As far as improvements, the runtime management in MuleSoft can be much better. It has some issues, or we have issues with it. As for additions, the first thing that I want is for them to create their own connectors. That is very important. Other things beyond that are hard for me to imagine right now especially because they have a lot of features already that we are not yet using.

Technical support needs to be improved, especially when you need help with more technical aspects of the solution.

MuleSoft's release calendar is rather conventional with two major, two minor releases and hotfixes in between. The competition sometimes offer more rapid release cycles and provides improved and new functionality with a faster time-to-market.

Some improvements are required in the batch component in case of errors in the batch steps and when processing a large volume of data.