What needs improvement with Alteryx?

I haven't had much experience with the tool, however, it could improve its documentation. It's something I need to understand better. The tool could include more native connectors, such as for global ERPs, instead of requiring additional fees for these connections.

When configuring target tables, it is difficult to see the full text when deciding on load operations. Performance can be slow when updating existing records. Documentation could be improved for recurring issues.

The product could benefit from improved integration with visualization tools or even the inclusion of built-in visualization features. It would reduce the need to use multiple tools and pay for additional licenses.

The product's pricing needs improvement.

I think sometimes the solution doesn't load properly or takes so much time for the workflows. Though the workflow runs and completes the file in Excel, if you use the same formula, it's a bit slow. Also, the image processing is not so good because I tried to do some image processing and they were like, sometimes they put two to eight. In the document, it was two, but the OCR predicted it as eight.

There have been some issues with licensing, particularly with the increased prices. This has led some companies, including ours, to consider reducing the number of licenses or potentially discontinuing their use of Alteryx. Since most of our team members work with Alteryx, deploying workflows to clients who may not have licenses became a hurdle. Every user needed a license to run the workflows, which could pose difficulties, especially for clients who only needed occasional access for small data analyses. Unlike other software like Microsoft Power BI, which offers free trials or user licenses for basic functionality, Alteryx doesn't provide such options, potentially limiting its accessibility for occasional users. It should improve scalability.

In terms of improvement, the transition from the desktop to the cloud can be challenging. While the desktop piece works seamlessly within the company's security architecture, when you load workflows onto the cloud, the same connections don't translate due to firewall issues. It would be beneficial if this transition could be made more seamless and straightforward for users. The solution is already feature-rich and doesn't require any improvements. Their current versions are ahead of the curve. Their machine learning and predictive modeling features place them at the forefront. Some advanced features are more than what's needed for our regular business purposes.

Alteryx's predictive data models are pretty average and can be improved. If I'm using it on a day-to-day basis, the predictive model should keep improving based on data gathering and data preparation.

I don't see a huge downside. Because if there is a strong leader with some good architecture knowledge, they should be able to cut down most of the downsides. However, there are a few limitations. For example, if you just leave the workflows to the developer, they may not properly document or organize the workflows. And then, if they're not organized, it's very difficult for another person to pick them up. Another downside is that two professionals can't work on one workflow at the same time. Let's say I have a huge workflow, and someone needs to support it. All I can do is virtually connect and share my screen, or I have to be with them together to look at the workflow and make updates. So the downsides are very few. But if you're more organized and have a dedicated architecture for your own solution, that will reduce a lot of downsides. But there are a few minor problems. For example, we can't browse multiple files. When we deploy a solution on a gallery, let's say I have ten different files, and I have to upload them all at once. This is something that's difficult in the gallery. So case by case, I see some downsides, but often we do something alternative. And then, recently, I came across one biggest downside. There's something called the Download Tool, which is used to push datasets over an API. If you have API credentials, you can configure them in this Download Tool. But I noticed that this Download Tool has a limited capacity. That means it can only push, like, X amount of values or X amount of strings, but it cannot push a huge amount of dataset. For that, I have to deal with the Python port again. So, that's a bigger downside, which I recently came across. In future releases, I would like to see Alteryx improve the connectors. Since Alteryx is a low-code platform and even our developers might not have extensive coding knowledge. If there's a possibility to enhance the connectors, it would be truly valuable. By connectors, I mean that I have come across some instances where these connectors incorporate a sort of programming language that could augment Alteryx's capabilities. This way, we could utilize these connectors instead of having to create a separate solution. For instance, if I intend to extract or transmit data from SFTP to Alteryx, it would be beneficial to have a pre-developed connector accessible within the community. Take the Power BI connector, for example. It's important for this connector to be consistently maintained. I'm aware that these types of connectors are already accessible within the community, but there are instances when they function as intended and times when they do not. Would it be possible to have a centralized repository where these connectors are regularly updated to align with the most recent versions? This would indeed be highly advantageous. I understand that it entails a significant amount of maintenance work. Nonetheless, given that Alteryx operates as a low-code platform and we frequently encounter a variety of challenges, we aim to minimize the need to engage additional resources such as Python or Java developers. Relying on such external dependencies can introduce complexities. Therefore, if Alteryx could introduce more connectors of this nature, it would be a substantial improvement. APIs and connectors play a pivotal role. In particular, we have numerous scenarios where I had to leverage APIs and SSEPs.

The price of the solution could improve by being lower.

To use Alteryx, you need to think like the people who write software codes. The formula we currently use in Alteryx can be automated. This would be helpful for people using Alteryx who are not from a coding background and can write the formula differently. Whoever is using Alteryx should feel some convenience using it. We found most of the useful features built into Alteryx, but they all are on a paid basis.

From a commercial standpoint, more options should be available for pricing. The licensing model needs to be specific.

They should make the solution user-friendly for nontechnical people by giving specific names to the options. That will be helpful for them to understand.

I'd like it to be easier to work with PDF. When you want it to read PDFs, it can be a bit difficult. That may be a PDF issue, however. Maybe if I was doing more spatial or geographical tasks, I might need a bit more; however, for my purposes, I have all the features we need.

The solution can be made more affordable.

Sometimes, there are performance constraints. Especially when a large file has to be ingested, the system slows down a bit. Its performance is the only thing that can be improved.

There is a community for Alteryx, but it is so spread out. For example, if I face an issue with Informatica, I have Informatica forums that keep on adding to that. They have a very cohesive structure when it comes to the issues and their error messages. They are very easy to use for Informatica issues, but in the case of Alteryx, the information is too spread out, and it is difficult to find the answer to a particular issue that someone faces. If Alteryx can organize its forum and the issues in a better way, it will be easy for developers. I mostly used it for flat files, but I have many colleagues who reported that to tune a query, in case they want to directly connect to the database, there is no option to optimize the performance of the query, as we have in Informatica. With Informatica, we can use hints and other things and they directly hit the database, but with Alteryx, that is not the case. So, when it comes to database connectivity, the performance is not as good as a flat file as a source.

Configuration is very low and is not good for the customer end. They should be focused more on regulation. They need to come onto the cloud as soon as possible.

I have not yet been exposed to Alteryx as much to comment on negative experiences. I'm taking a class this semester that has to do with Alteryx. I'm pretty sure that once I'm learning the projects and utilizing Alteryx more, then I can give back more feedback. The usability could use some improvements. There are icons that you choose in terms of inputting the tool or even for filtering and then running different functions, there are different ways of designing your input tools and output tools, and that could get confusing. I would suggest that if there's any way of simplifying it so that instead of tools or instead of understanding how it works, we could focus more on its usability and the output of it. For me, it was a bit challenging to figure out how to work with Alteryx. If there is any way to make the learning curve less steep, that would be ideal.

Alteryx can improve the model management and deployment processing of large workloads. In a future release, there should be more cloud integration with GCP. The ability to connect to GCP components as well as deploy things on the cloud would be beneficial.

My understanding is that they will include a visualization feature. All of the reports are migrated or exported in an Excel file, and most of the time, a business intelligence tool is required. They could have better reporting. The aesthetic could be improved. We have limitations on the number of fields. It doesn't really get everything if we have to add more columns.

Alteryx can improve in data science. They have to have more features and components in the data science aspect because they claim to be a data science tool. However, in order to be more competitive, they have to improve on their data science propositions. Thre are other solutions on the market, such as other players in the market, Data2Go or DataIQ, and Alteryx needs to catch up.

It would be nice if they can provide Alteryx with more options for In-DB connectivity. That functionality is there, but it doesn't include all software we are connecting. In-DB is a concept in Alteryx where we don't need to export data into the Alteryx server to process. The data remains in the source system, so Alteryx can directly manipulate it. That feature is available only for a few tools with limited options.

In some ways, I believe it is not yet as integrated as it could be. I believe that the data integration inquiring component could be improved slightly. The data integration component could most likely be improved to increase enterprise scalability.

Alteryx's development environment could be improved as it requires installation locally and can't be developed in the cloud. You have to push the model into the cloud from your local machine, which is a pain compared to other tools like DataIQ, which you can develop fully from scratch in the cloud.

The biggest limitation is that the solution doesn't allow you to do sophisticated things in terms of fiddling with their implementations of machine learning and deep learning components. I think they should really work on integrating or have a capacity to integrate tailor-made algorithmic code. I think that's one of the most important things they need to be doing.

A colleague of mind mentioned that the solution should have more options for the visualization of data. The visualization filtering could stand improvement. The documentation could be improved upon. The price could be better. Though we are satisfied with the solution, we feel the price to be a bit on the expensive side. We make use of another ETL solution, the PowerCenter. We would like to improve on and escalate Alteryx, but find it too expensive at the moment to do so.

There could be a bit of improvement related to performance. Sometimes it demands a lot of resources for running it, like memory and search. One thing that I would like to see is the ability to get all the versions. Currently we have just the current version and the previous version. Sometimes they have no old workflows. They will run and they are in previous versions but we don't have access to it. It would be helpful to have an opportunity to download, maybe not the first one, but a couple past versions that have been useful.

I still haven't had any issues with the Alteryx designer. I don't have anything to speak negatively of now, however, I'm sure there are some features that need improvement. The technical support could have a little bit of improvement. They should work to make the solution less expensive. The product needs some more tools for the image processing part. That would be pretty good. There's an image processing suite, however, it could use a little bit of improvement as well as the visualization aspect. They should maybe add some capabilities for more interactive dashboarding. That would be great.

The workflow and pipeline need to improve. For example, if you have a job running, and you need to run another, the scheduling could be easier in Alteryx. In the next release, I would like to see the completed pipeline, improved workflow with data, easier, and better scheduling to allow tasks to run one after the other.

This is a proprietary tool and maybe the open source part of it could be improved. If there was an open source edition it would be an advantage; at least the training part of it could be freely available and then people would be more likely to adopt it.

The only weakness I would say is on the visualization side of having that dynamic incapability. I know that it is dependent upon having the on-premise and cloud server environments, but even there, the product functionality is somewhat limited. This would reduce the need to integrate it with other products.

The solution could improve in the visualization. There is a reporting palette available, but I do not think organizations would rely just on that one pallet. We need a BI tool for larger implementation reporting. The visualization cannot just be the basic visualization but immediate reporting. When comparing this solutions' BI tool to major competitors, such as Tableau and Power BI, the competitors both massively use visualization and dashboarding whereas this solution does not. However, this solution is much further ahead in other areas, for example, deployment. When it comes to just the visualization it not fair to compare these solutions.

The deep learning models are not currently supported by Alteryx. The world is moving towards more advanced machine learning with deep learning models and integration, and Alteryx should do the same. The advanced features are not there, given that most companies are constantly increasing the size of data. Alteryx is not very robust when it comes to working on large sets of data. The new platforms that are coming onto the market, even the open source technologies, are moving really fast. Alteryx needs to find an edge.

The principal problem is the pricing. They're expensive products. They don't have an option to scale the product while the customer is starting to use the product. There are no options for customers that want to move from smaller scale and expand the solution to the rest oth the organización. The Designer (pleople who develope models and flows) cost 5,195 USD each one, but if you want to collaborate, you have to buy a server which cost 80,000 USD not matter how many users you have, so you cannot scale from a small group of users to a large one. It's only available as an on-premises solution. It would be ideal if the product offered some sot of option for smaller companies.

The solution does have a bit of a learning curve. I initially managed the product completely. Then, when the user base grew, another group just took it over. What they're struggling with is it's not as mature as Tableau in the user management area. It was tougher to manage the server part of it right away, especially since the user base has grown. When you get hundreds of users in there, trying to manage access to the product is a little tougher than Tableau.

Our IT team has a lot of restrictions. Therefore, we cannot extend all the functions and all the features that Alteryx has, unfortunately. That's due to the fact that the internal restrictions we have are from a security standpoint. The desktop license is expensive. The screen when you are looking into your workflows and your ETL processes needs to be improved. You cannot manage it very well. For example, the column size, and items like that should be adjustable.

They can provide some pre-built tools for predictive analytics instead of us having to build all the tools. It should also be improved from the visualization aspect. It should have better visualization capabilities. There are tools out there that have better visualization capabilities, which Alteryx is lacking currently.

It should have Linux support. It currently supports only Windows. It does not support any other platform. Its price should also be lower. Their US partner management program is actually unresponsive. One of the reasons why we don't have a formalized partnership plan with them is because their partner management team is atrociously unresponsive. This is something that they need to change.

We've had a lot of issues within the Alteryx Server. It really could be improved. There are memory management issues in terms of the Mongo database that they manage. There could be a better user interface for the gallery, which is where we automate the workflows. Sometimes workflows tend to queue up, and they tend to get canceled for some reason that we don't know sometimes.

I can't recall a feature that may be lacking. Overall, we've been happy with the solution. It's a technical product and those that don't have proper training will have to deal with a steep learning curve.

The interface could be improved. Currently, when you are looking at the information on the screen it is broken down into three parts of the screen. For example, when doing algorithms and you want to see the results on the screen this is when it gets really difficult to view. However, this is fine visually to work with on a big screen but when you are working with a laptop screen it is not very comfortable. Everything is very small. You have to know to use it. You have to have some training and it's hard. This could improve.

Alteryx brands itself as a tool that you can use to avoid having to code. In reality, however, as soon as you start with your task, it gets just a little complex. Alteryx is just as complicated as coding, in my opinion. It is great for relatively simple data mining and data extraction type tasks, and I sure that it can do very advanced things, in particular when you look at the tool installation. That said, as soon as it becomes advanced, it is not very easy to use. In fact, it's quite complicated, so when persons with a coding background attempt to use it, they fall back on trying to code something, instead. The biggest complaint that I have is that the online help is not very comprehensive. It is very formal, but there are not a lot of examples and use cases, and there are many signs there are difficult to understand. I think that it's a challenge. In the future, I would like to see improvements made to the machine learning capabilities such that they are comparable with Python.

I honestly can't think of anything that needs to be improved.

With some of the features in the database, they don't offer certain things, like Keynote or transposing. I'm not sure, if, with the latest update, they've added these capabilities in or not. The solution can be quite complex in some aspects. For some specific coders, it's still is a kind of concern that they don't have coding options available. We're moving to Dataiku due to the fact that it caters to both coders and clickers (those who prefer drag and drop).

There are a few imputation techniques which they really need to include. Things you'd find in Python. And there are advanced techniques coming in every day there. Alteryx needs to update more quickly. There are also a few things from geospatial or updating perspectives that are still better in open-source software than what you find in Alteryx.

For the area in Alteryx, such as the server the prices are high and it would be better to reduce the fees. Some of the new tools in the intelligence Suite are the Optical Character Recognition (OCR) and the Natural Language Processing (NLP) language for English characters. We are hoping that the NLP features will also support Chinese characters.

For the most part, the solution is very complete, and there isn't much missing. The designer license lacks a few features that the server has. There's a big jump in terms of pricing between license tiers. I'm not sure I understand why the price jumps are so high.

The solution just lacks in terms of data visualization. That is why we use Tableau and Qlik in our organization. They help to pick up the slack. If data visualization was added, Alteryx would be a very good tool, and much more complete. The solution could maybe use a connector as a direct input for MySQL or something like that. Right now, we use ODBC connectors for MySQL. If a direct connector was available on this product that would be very helpful, especially for our organization.

There are no ready models to use in analytics. There is no content. You have the tools but you don't have the content. In the next release, I would like to see Auto Machine Learning.

The server is too expensive for what you get and it is really a designer desktop on a server. It is not worth the money.

The visualization could be improved, as could the BID bots. The solution could work on the BI side of the tool to make it a bit better.

I think the only area where the product lags is documentation and videos on the analytical app and the batch macro. We have learned the hard way, but I see people struggling there. Secondly, the tools for data science are not as user-friendly as KNIME. Additional features that could be included would be improved documentation, more user-friendly data science, and the pricing should be more competitive.

I'm sure there is always development happening with the product, but I think there should be both cloud and on-premises options. We have not seen a cloud deployment and have not heard they will be making one available. I think if there is the cloud deployment available, it would definitely be more convenient for some users to engage with and use while lowering the implementation costs. Cloud deployment would be easier for us and our clients because a lot of services that we push to the clients are through the cloud only. It is expensive to buy additional licensing, and it is time and resource consuming to install onto our own hardware — which itself requires maintenance and administrative attention. On-premises can raise the cost of deployment over just using a cloud enrollment option.

Alteryx is very good - I don't see any room for improvement. I can't think of any downside to Alteryx. I use a lot of third-party tools with Alteryx. It would be great if Alteryx could take third party tools and incorporate them.

There are a few ways the solution could be made better. The graphical user interface could be made more friendly. I mean from the GUI point of view, it is a little dated. But from a functionality point of view, I think even for people who do not know the coding part and who are more business users than programmers, they would not have a very hard time using Alteryx for creating machine learning models. It is quite good and quite useful already without improvement to the GUI. It could just stand to be more modern-looking. I would like if they could make some type of free license for the product. Not a free full license but a limited edition or limited capability free version. The limited-time version is a little restrictive. A free, open-term desktop version like Power BI and Tableau and KNIME might help users more with discovery and utility.

The event handling, such that the file system watcher, is in need of improvement.

More emphasis on extensible R and data augmentation interface design.

Add inbuilt interactive visualization in Alteryx.

I think better visualization would be helpful. In terms of data extraction, during the sequel queries of servers as an example, you can structure your sequel query and it's fine, very readable, but once you run it, it crashes into one complex single line, and makes it very difficult. Then you you almost have to go back to management studio, and run it through management studio and put it through restructuring, and it just seems to me that that's not very user friendly.

* Alteryx could be improved in the area of analytics and central governance. * The next feature release should include easier reporting. * Improvement in the coexistence of In-Database, Python and R tools.

Even when it already includes some AI models, this area could be improved. The another area which could be improved is the visualizations. It would be great to create the final users' visualization within Alteryx.