We just raised a $30M Series A: Read our story
2018-10-02T19:05:00Z

What needs improvement with New Relic APM?

13

Please share with the community what you think needs improvement with New Relic APM.

What are its weaknesses? What would you like to see changed in a future version?

ITCS user
Guest
2222 Answers

author avatar
Top 10Real User

It offers transactions, but it does not offer an endpoint-level insight at the URL level. When we get a request, we want to know what the life cycle of that service is, and where the cycle is. This is what I am trying to locate with most of the solutions now. I am trying to research how to find a cycle per endpoint and not at the service level. It is very difficult to award the service level cycles at an endpoint level. It is important for us to get new insights to create better hygiene around the business use cases. At the endpoint level, the visibility is not that great, and metrics are not available. It gives you a full view of the entire function's execution and not from the context of the URL altogether. Also the response time, the latency contribution, and the throughput contribution are areas that need improvement. You can get the throughput contribution from New Relic, but not the latency contributions. You cannot get it at all. These are the major limitations. When working with AppDynamics, I did not find any limitations, but the same can not be said with New Relic. The way that it classifies the actual services is a bit ambiguous. It's not perfect. For example, I see there are certain solutions that are listed as extra services, as a dependency, and still I find that among load contribution, it tries to show that those services separately, which is confusing. With the transactions, when it tries to show a type of "bufferHandler" from inside, it doesn't show what the nature of the request is. Especially with Microservices, it doesn't show what kind of method is present, which makes finding data very difficult. Instead, you need to go to the raw data. I think that defeats the purpose of using this tool. The transactions do not show the time consumed by the request, from the metrics execution perspective. It was suggested that I did not know how to read it but I have done all that I could. It is very difficult to relate to and requires a lot of experience and time to read through, which it should not. It should not be difficult to find the latency and throughput for the entire system when requested. It should not be difficult to develop the data that relates to the various types of execution. It should have complete exposure around the endpoints. The services-to-service dependency is fine but most of the startups have only one or two services that are all cycled. It does not provide you with a lot of help when you are showing that the two services are dependent. What all of the dependent endpoints are and how are the cycles being formed is information that should be available in most tools, but not with New Relic and some other tools.

2021-05-29T16:41:57Z
author avatar
Top 10Real User

I haven't yet had too much of a chance to read up on the product or to compare it to other solutions to see if there are items that are standard elsewhere but lacking in this product. I need more time with the solution to really talk about any shortcomings if they exist at all. It would be nice if there were pre-made dashboards. There should be out-of-the-box options that you can have. It would be nice to have something already done, so that you don't need to reinvent the wheel again, to create your own dashboard. The pricing might be high if you are a company that goes through a lot of data.

2021-04-07T13:06:01Z
author avatar
Top 10Real User

There has been some problem with the agent, and it is just not working well. It is not able to record information with the application server. They have been able to fix the issue, but it took quite a long time. This is the main issue in the APM products and also in New Relic. The mobile application monitoring has been pretty difficult to set up and also quite expensive. It should be a little bit easier and cheaper. Because it is pretty difficult and expensive, many customers don't take it.

2021-03-01T13:57:33Z
author avatar
Top 10Real User

There really is nothing that stands out with New Relic. With the insight, I think it will be found lacking for its report aggregation capabilities. How granular I could go down at looking at certain data, especially related to the operations, is limited. The API integrations that they have for us to automate our configuration was fine, but I think for some of these tools, it was over-engineering for us to try and automate any of that. So, we just use the user interfaces.

2021-01-30T06:40:07Z
author avatar
Top 10Real User

The older view is much better than the new view that they have. We'd like to go back to that previous version. The user interface just isn't as nice as it used to be. The solution does not seem to be lacking any features. I can't say that I recommend any specific features for the next release.

2020-12-24T19:24:06Z
author avatar
Top 5LeaderboardReal User

The main reason we switched away from New Relic is that the cost is too high. They should bring the pricing down to be more competitive. I would like to see the capacity planning improved. The security standard and compliance are areas that should be improved.

2020-11-25T20:11:58Z
author avatar
Top 10Real User

The solution needs some sort of improvement on the synthetic monitoring site. The product has good documentation for Linux, however, their documentation for Windows is lacking substantially. It's something they need to develop and mature over time. It would be great if it had out of the box integration with AWS, GCP, and other clouds. For example, if I'm using BigQuery in GCP, I want the data that I am executing in BigQuery to be reflected in New Relic.

2020-07-26T08:19:05Z
author avatar
Top 10Real User

Data retention should be increased, advanced RCA and all possible and best recommendations should be provided. Pricing should be revised

2020-06-18T05:17:54Z
author avatar
Real User

I would like to see the company implement the AI auto-baseline feature which Dynatrace has.

2019-03-12T07:26:00Z
author avatar
Real User

I would like if it could have predictive analysis. Today, we only have the option to configure thresholds. In addition, it would be nice to have centralized log management, like Datadog does. As New Relic already has all of the application information and traces, it could compare them with application logs and do better analysis. Thus, it could be cheaper, have predictive analysis and log management.

2019-01-14T13:16:00Z
author avatar
Real User

They could improve the education process and how people understand that these tools are very technical. I understand everything very quickly and where it all comes in because I grew up with the product, but right now if someone was to pick it up from day one, it is a very steep learning curve. The monitoring is only as good as the alerts that it produces. By having it set up fine grain alerting, it is a bit of a pain. They already have all these other companies that use their system, so they should easily be able do alerts based on deviations that we don't need to program on a per instance or artifact basis.

2018-12-11T08:31:00Z
author avatar
Real User

There is room for improvement on the alerting, insights, and infrastructure monitoring. It provides basic level alerting, and I would like more details. I would also like an infrastructure network that provides real-time views, showing the issues. Compared to their competitors, they are missing some features at the moment.

2018-12-11T08:31:00Z
author avatar
Real User

The APIs could be better. I would also like more APIs and features to integrate with streaming solutions, like Kinesis or Kafka.

2018-12-11T08:31:00Z
author avatar
Real User

There is a picture which goes to your browser and it monitors requests from other users. However, it's impossible to use now because the price is very high. The feature is very nice, but I tried it during the trial period, and the current price makes it impossible to use.

2018-12-11T08:31:00Z
author avatar
Real User

It is complicated, especially in how you interpret the data that it provides. For someone who works in it every day, I can figure out what I want. For the general, every day developer who uses it once a month, there is large learning curve to figure out exactly the information that they want from it. If it had a bit more canned, out-of-the-box features, especially some of the reporting features, that would be more useful. Sometimes, it is difficult to work through and figure out. Some things are difficult to work through which is why I haven't done them yet, because it will take me six hours to figure out how to set them up, e.g., the dashboard. How I want it to look and how the developers might want to interpret that data, but I don't have six hours to go figure this out, and it takes a long time to do this stuff. They have this alerting capability where I can set up an alert policy, then within that alert policy, I can set up as many alerts as I want. I can set up one or I can set up a 1000. I would like a feature where I can turn off alerting at a policy level. Thus, when a policy is inactive, I can shut down all of my alerts within the policy. Right now, I have to go through them manually and deactivate each one that I don't want to use.

2018-12-11T08:31:00Z
author avatar
Real User

I would like to have storage monitoring. E.g., being able to monitor SANS, specifically protocols, like NFS and CIFS metrics. I have not seen the ability for it to do this and having some way to do this would be awesome.

2018-12-11T08:31:00Z
author avatar
Real User

We are sharing different AWS accounts, and if a Lambda has with the same name but a different AWS account, it is a little hard to understand whether AWS or Lamda belongs to that account. Also, we have multiple accounts on the drop down to filter by Lambda, but we see two Lambda with the same name, then we don't know which one to choose. So, it needs to improve its filtering. We would like a more sophisticated filtering for the Serverless AWS pieces.

2018-12-11T08:30:00Z
author avatar
Real User

We have had issues with our agents going offline.

2018-12-11T08:30:00Z
author avatar
Real User

The price needs improvement.

2018-12-11T08:30:00Z
author avatar
Real User

* I would like to be able to invest less time in IT and ad hocs. We should be concentrating on other issues. * I would like more deep dive monitoring into services and being able to install it on some apps.

2018-12-11T08:30:00Z
author avatar
Real User

We would like to receive more AWS-specific details from the New Relic Dashboard, like EC2 health.

2018-12-04T07:57:00Z
author avatar
Real User

I think the APM mode can be improved. In addition, its difficult to have a predictive tool to see how the application would behave in the future when it basically only shows the historical data.

2018-10-02T19:05:00Z
Learn what your peers think about New Relic APM. Get advice and tips from experienced pros sharing their opinions. Updated: October 2021.
542,267 professionals have used our research since 2012.