Datadog Reviews

The solution is basically used for servers and applications.

The UI, basically, is the most valuable aspect of the solution. I really like the look and feel of the solution. It's not very distinctive now since other players have caught up, however, they were the first in the market to present such an effective UI. 

The many dozens of integrations that the solution brings out of the box are excellent.

It's easy to set up.

Deploying the agents is still very manual. 

Network monitoring could be better or rolled into this solution so that you do not have to buy a different product.

Customization of the tool itself should be taken into account. At the moment, although what they provide out of the box is good, they don't offer many customization possibilities. I know it's difficult, however, it's something that they would need to look at. When the customer gets some customization, they want customized requirements. We cannot do it. 

I've been dealing with the solution for five years. 

It's quite stable. I have never had an issue in regard to reliability, so it's very stable.

It's very scalable. I have not reached the limits at any time, never in the solution. I've never seen any performance degradation in large environments. I would say it's very scalable.

Each client has its own instance. We do not share instances with multiple customers. There's usually between 20 and 30, depending on the customer.

I never use technical support, to be honest.

The initial setup for the solution itself is quite straightforward. You just set it up and that's it. However, when it comes to, for instance, deploying the agents to the servers, or at least the target machines, it's still a manual task. They still do not have centralized management of the FD agents, which basically delays the deployment of the solution. It's very manual still.

How long it takes to deploy is difficult to pin down. It will vary based on the environment size. Obviously, if it's ten servers, it will basically take half an hour or one hour. If it's 5,000, obviously, besides the number of notes, other considerations will need to be taken into account. If t's a large environment, it will take much longer. We would need to basically develop a solution, or an effective process to deploy the agent and configure them in a standardized manner. This is something that the tool itself or the tool provider does not offer out of the box. You need to build it. That's a drawback.

How many people you need for the deployment and maintenance processes depends on the environment's size and geographical area. On average,  I would usually require for every 500 notes, one resource for implementation. Then for overall support, I usually put one resource per 1500.

Before, the ROI was much higher as you would not have to compete with any kind of tool since they were very good in the space. However, with time, other companies have picked up the slack. Now, you have other tools which provide a higher ROI. I cannot give a specific ROI percentage since I don't use it for personal use with deployment. We deploy it on behalf of customers. Obviously, depending on the deal, depending on the size, and the ROI will vary. If people are looking for a global monitoring solution in the same tool as Datadog network monitoring, they are always hindered as Datadog does not provide an adequate solution for it. That kind of decreases the ROI since you still need to get another tool to do the network monitoring.

The licensing is a bit complicated. When you pay for it on a note basis, that's perfectly fine. However, when you put log analytics on top of it, it's based on traffic. This is actually an issue. It gets complicated.

I'm providing Datadog. I'm a retailer.

I would recommend the solution. 

I would suggest if their environment is in the cloud, companies have their environments in the public cloud, such as GCP, Azure, or AWS. Datadog is a very good candidate to provide an overview of the monitoring. If you want to consider a hybrid solution where systems and servers and applications also provide a good solution and have a lot of APM capabilities, the only drawback will be network monitoring. When you grab a tool that you want to basically monitor the entire environment at a single point of contact, with Datadog, it's possible, however, there's not an effective tool to do network monitoring.

I'd rate the solution seven out of ten.

My customers were using Datadog for monitoring purposes. They were using it only because the solution is running on AWS and it's a microservices-based solution. They were using an application called Dynatrace for their log.

Most of the features in the way Datadog does monitoring are commendable and that is the reason we choose it. We did some comparisons before picking Datadog. Datadog was recommended based on the features provided.

Most of the monitoring tools nowadays are have or are going to have embedded artificial intelligence and machine learning to make monitoring and logging more proactive and intelligent. Datadog has incorporated some artificial intelligence.

The solution does not require a lot of maintenance.

The solution had all the features we were looking for and we were able to create a central dashboard as per our requirements.

All solutions have some area to improve, and in Datadog they can improve their overall technology moving forward.

I have been using Datadog for approximately four months.

Datadog is a stable solution.

Datadog is a highly scalable solution because it is a SaaS solution. Having this solution be a SaaS is one of its most appealing attributes. When the vendor is going to manage data scaling and everything for you, you are only going to use the solution as per your requirements. Autoscaling is a great feature that they have.

The support from Datadog is exellent. If you're stuck on something or you are facing any issue, support from the vendor itself is available. You will receive a response instantly from the vendor on anything related to the requirement,  issues, or feature you are looking for. The responses have always been in a timely manner.

I rate the technical support from Datadog a five out of five.

I have used other similar solutions to Datadog and when I do a comparison between the other tools Datadog is on top, it is great.

Since Datadog is a SaaS solution we had not deployed the Datadog on-premise or in any Cloud. We were using the SaaS solution from the vendor itself. From the provisioning perspective or from the monitoring and dashboard perspective, we were using Terraform to create the typical monitoring as code. Everything was basically automated, we were not doing anything manually.

If someone wants to set up Datadog on-premise or in any of the Cloud machines, they have to consider a lot of things from the auto-scaling perspective.

My recommendation is Datadog is very good. Your team can mainly focus on the development rather than the solution itself. The vendor is going to take care of auto-scaling and maintenance and everything for you.

I rate Datadog a nine out of ten.

We are using the solution for migrating out of the data center. Old apps need to be re-architected. We are planning on moving to multi-cloud for disaster recovery and to avoid vendor lockouts. 

The migration is a mix between an MSP (Infosys) and in-house developers. The hard part is ensuring these apps run the same in the cloud as they do on-premises. Then we also need to ensure that we improve performance when possible. With deadlines approaching quickly it's important not to cut corners - which is why we needed observability

Using the product has caused a paradigm shift in how we deploy monitoring. Before, we had a one-to-one lookup in ServiceNow. This wouldn't scale, as teams wouldn't be able to create monitors on the fly and would have to wait on us to contact the ServiceNow team to create a custom lookup. Now, in real-time, as new instances are spun up and down, they are still guaranteed to be covered by monitoring. This used to require a change request, and now it is automatic.

For use, the most valuable features we have are infrastructure and APM metrics.

The seamless integration between Datadog and hundreds of apps makes onboarding new products and teams a breeze. 

We rely heavily on the API crawlers Datadog uses for cloud integrations. These allow us to pick up and leverage the tags teams have already deployed without having to also make them add it at the agent level. Then we use Datadog's conditionals in the monitor to dynamically alert hundreds of teams. 

With the ServiceNow integration, we can also assign tickets based on the environment. Now our top teams are using the APM/profiler to find bottlenecks and improve the speed of our apps

The real issue with this product is cost control. For example, when logs first came out they didn't have any index cuts. This caused runaway logs and exploding costs. 

It seems that admin cost control granularity is an afterthought. For example, synthetics have been out for over four years, yet there is no way to limit teams from creating tests that fire off every minute. If we could say you can't test more than once every five minutes, that would save us 5X on our bill.

I've used the solution for about three years. 

The solution is very stable. There are not too many outages, and they fix them fast.

It is easy to scale. That is why we adopted it.

Before premium support, I would avoid using them as it was so bad.

Neutral

We previously used AppDynamics. It isn't built for the cloud and is hard to deploy at scale.

The initial setup was not difficult. We just had to teach teams the concept of tags.

We did the implementation in-house. It was me. I am the SME for Datadog at the company.

The solution has saved months of time and reduced blindspots for all app teams.

I'd advise users to be careful with logs and the APM as those are the ones that can get expensive fast.

We looked into Dynatrace. However, we found the cost to be high.

We deploy various services for our main platform on AWS across multiple regions. We have a development environment, a staging environment, a QA environment, and a production environment. We deploy our many services across hundreds of instances. 

We have many server farms, all responsible for various services on our market intelligence platform. The deployment of each server farm or even individual instances varies depending on what stood up. We have instances built in three different ways, with two different pipelines and some even on user data scripts.

My team has a 24/7 on-call schedule where we need to be ready to handle and mitigate incidents with the platform at any moment. 

We have countless monitors set up on Datadog that alert directly to our queue using an email that generates a ticket. 

The actionable steps for each type of monitor and its associated incident are easily included in the alerts whenever something is triggered. We generate links to the Datadog monitors and can instantly drill down into what went wrong and for how long.

The features I have found most helpful are synthetic monitoring, APM, and alert features. The monitoring functionality, in general, and tagging infrastructure are great.

Synthetics have become bread and butter for us as we have migrated many tests over to Datadog. We have simplified and consolidated our synthetic tests while also making them more robust with the help of your tagging. 

A large portion of our monitoring is based on synthetics results, and alerts integrate seamlessly without an incident queue system. We use dashboards heavily. 

The metrics capabilities are extremely helpful, and we use virtually all of the widgets.

My main place of improvement for Datadog would be the documentation. While the tool is robust with many different capabilities, users would greatly benefit from more examples in the documentation. 

The number of current code snippets available in the docs is not enough, and some need to be updated even today. 

One function I would add would be a button to generate a report of the performance of a synthetic test and the performance of each of the steps in the test over time.

This timeline varies in terms of how long we've used the solution. We have one platform completely in the cloud and one still on-premises. We've had the solution for many years on AWS.

Datadog has flexibility.

The product needs to have more enterprise approach to configuration.

We use the tool to monitor our whole infrastructure. CPU, memory, and disk space are the types of things we use it for.

It is a stable solution.

It is a scalable solution.

The technical support team is good and responsive.

Positive

The initial setup is not very easy and the deployment took eight months.It took quite a few teams to get it all accomplished. I rate it a six out of ten.

I rate the solution eight out of ten.

We use Datadog for monitoring to get the traces and logs of all our applications. Datadog provides dashboard and alert capabilities to identify if something is wrong with various teams. More than 200 users, mostly software engineers, work with Datadog. 

Datadog provides tracing and logging, whereas Dynatrace focuses on tracing, and Splunk is more of a logging tool. Datadog's advantage is that we don't need two tools. 

Datadog isn't as mature as some of the established players like Dynatrace or Splunk. It's a new product, so they are constantly releasing new features, and I don't have much to complain about.

We have used Datadog for seven months.

We haven't issued any issues so far, so it's a highly stable platform. 

We are a unit within a much larger entity that is using Datadog. It can scale up to meet your needs. 

We have regular calls with the Datadog team. They take feedback and bring in the product managers to quickly answer questions and fix issues. They help you deal with some of the issues you have with any new product, but Datadog is one of the fastest-growing products in the monitoring space.

You don't need to install anything because it's a SaaS product with a web-based UI. They provide you the credentials to give you admin access. You only need to install the agents where you need monitoring. The time required to deploy the agent depends on what you're monitoring, but the solution itself works like Office 365 or any other SaaS product. 

We primarily use the solution for application monitoring (APM, logs, metrics, alerts).

It's useful for active monitoring (static monitors, threshold monitors). We get a lot of value out of anomaly detection as well. SLOs and monitoring of SLOs have been another value add.

In terms of metrics, the out-of-the-box infrastructure metrics that come with the Datadog agent installation are great. We have made use of both the custom metrics implementation as well as the log-based metrics which are extremely convenient.

We also leverage Datadog for use of RUM and want to explore session replay.

It is easy to implement and scale applications with standardized visibility, monitoring and alerting

We get a lot of value out of passive and active monitoring. While different teams across our organization have used different services (metrics, logs, APM, RUM), almost all teams have been able to use the dashboards to report and track high-level metrics and active monitoring. 

Active monitoring (static monitors, threshold monitors) is great. We get a lot of value out of anomaly detection as well. SLOs and monitoring of SLOs have been another value add for our organization.

The APM and tracing provide visibility and the ability to get right to root cause issues while being able to deploy new services without much need for custom instrumentation quickly

The active monitoring (static monitors, threshold monitors) has been very helpful. We get a lot of value out of anomaly detection. SLOs and monitoring of SLOs have been extremely valuable.

The metrics and out-of-the-box infrastructure metrics that come with the Datadog agent installation are quite helpful to the organization. We have made use of both the custom metric implementation as well as the log-based metrics which are extremely convenient.

Datadog is so feature-rich that it is often hard to onboard new folks and tough to decide where to invest time. 

The APM is a perfect example of this. This feature alone has so much (profiling, tracing, span summary, flame graphs). I would love to see more of the insight and automation-focused features, such as the log patterns, where I can spend time more efficiently.

The cost of Datadog at scale can get very expensive very quickly. I would like to see a better usage/cost dashboard with breakdowns like the AWS cost explorer.

I've used the solution for three years.

We use Datadog for three main use cases, including:

• Infrastructure and application monitoring. It is ensuring that our services are available and performant at all times. This allows us to proactively address incidents and outages without customers contacting us. This includes monitoring of cloud resources (databases, load balancers, CPU usage, etc.), high-level application monitoring (response times, failure rates, etc.), and low-level application monitoring (business-oriented metrics and functional exceptions to customer experience.
• Analyzing application behavior, especially around performance. We often use Datadog's application performance monitoring on non-production environments to evaluate the impact of newly introduced features and gain confidence in changes.
• End-to-end regression testing for APIs and browser-based experiences. Using Datadog's synthetic testing checks periodically that the system behaves in the exact correct way. This is often used as a canary to detect issues even before users reach them organically.

Since we integrated Datadog, we have had increased confidence in the quality of our service, and we had an easier time increasing our delivery velocity. 

We have seen time after time that the monitors we have carefully created based on all ingested data are detecting issues quickly and accurately. 

This means we allow ourselves to manually test things less frequently. We have also had an easier time investigating application errors and slowness using Datadog's APM and log explorer products which allow us to introspect any part of the system, in its execution context.

The most valuable features include:

• Integrated observability data ingestions: All data that Datadog collects is connected. This allows easily connected logs with failed requests, and slow database questions with services and requests.
• Broad integrations allow us to monitor our entire production environment in a single place, not just cloud resources. Since all parts stream metrics, logs, and events to Datadog, we can have unified dashboards and manage monitors and incidents all from the same page.
• A high level of configuration. We can configure and modify many parts, from how data is collected from our applications to how Datadog parses and visualizes it. This means that we always get the best experience, and we don't need to find ten different products that do small things well or settle on one product that does everything badly.

Since the Datadog platform has so many separate features, solving so many use cases, there are often inconsistencies in feature availability and interoperability between products. 

Older, more mature products tend to be complete (many features, customization, broad integrations, etc.), while newer products will often be at a "just above minimum viable product" phase for a long time, doing what's intended yet missing valuable customizations and integrations.

We've used the solution for 12 months.

The solution scales very well on technical aspects, being able to ingest large quantities of data from many services. However, the pricing often doesn't scale naturally, and effort has to be put in to keep ongoing costs at a reasonable amount.

Customer service and support are generally very high-quality. In most cases, they reply very quickly and offer well-researched and relevant responses. This is contrasted with many vendors who take a long time to reply and send links to documentation instead of understanding the problem.

However, we had cases where support took several weeks to reply to a complicated request and sometimes eventually responded that the issue cannot be resolved. These are rare edge-case occurrences.

Positive

A large part of the initial setup was straightforward. We were able to collect about 80% of the relevant and 90% of the meaningful insights from just a couple of hours of connecting the AWS integration and the Datadog APM agent. 

Getting it to 100% and configuring and customizing things to our unique situation, took about two weeks. Datadog's documentation and support team were extremely helpful during both phases.

We handled the setup in-house.

From the number of outages stopped or shortened (which lead to lost revenue from non-renewals) and the number of hours saved on investigations (which correlates to engineering salaries), I estimate that the ROI of the implementation time and monthly charges to be between 10x and 20x.

We use the solution as a SaaS deployment.