Microsoft Purview Data Governance Reviews

I was working for a nonprofit here in Kansas called Children International. I was a global data engineering manager, and I was tasked specifically with developing a data governance program. And at the same time, that's right when Purview came out. So I spent a year in that realm. I've been very plugged into Purview ever since.

The main use is data governance. Now, not every firm is necessarily at that maturity level when it comes to understanding data as an asset. So, a lot of companies can use it just as a data dictionary, a business glossary, which is nice, but it doesn't give the entire picture of what Purview can actually do, and so, actually, there were a dozen of implementations of Purview that I have been a part of in the past. I've led as a consultant and an architect. I'd say maybe I have used twenty-five percent of every feature that it has to offer. I am still kind of catching up with the rest of the features.

Microsoft Purview breaks down barriers to sharing information, accessing data, understanding what a piece of data means, how it is defined, how it's interpreted, and then how the end user uses it. So, with Purview, when it's best used, and it is at its best when the end users are the ones that are very involved because they're the ones who are going to be using it. Its users can see the support, and it grows year over year, you know, a kind of revenue measure. I don't know how exactly this is created. Also, I don't know what fields it uses. I don't know exactly what this means. We have a different department in the company that may need this. So they feel the need to go out and then hop to Purview. So it definitely knocks down a lot of barriers. It eliminates a lot of needless communication between different departments, and it just centralizes all information about an organization's data assets.

Speaking of the valuable features, I would speak on behalf of my clients, considering how it seemed when it was implemented. So, the sensitivity labels are big. The insights are also very nice as it gives the visual representation of being able to look at what or how the data storage is performing, especially monitoring if they are tagging technical assets or if they are adding business glossary terms. So, that's big, along with the basic features of Microsoft Purview. I'd say the last one that is very big with most clients is data lineage and being able to integrate that end-to-end flow component kind of so they can see that this is data from a Power BI dataset, and then they can follow where that dataset is coming from like it's from a SQL query in a Power BI report itself and data coming from a SQL Server, etcetera, etcetera.

The out-of-the-box connectors and being able to connect to third-party apps and data storage were some of the areas where they started off strong, but that's something that I think needs to continuously be at the forefront of the developers and the engineers for Purview to make sure that it is constantly up to date and can talk with any data source that it may need within a data state. So I would say that's not necessarily a con, but it's a possible weak point that could be strengthened.

I've been implementing Microsoft Purview with my clients for three and a half years. I'm a consultant for a Microsoft partner.

I think Purview, in terms of its stability, a lot of people think Purview itself is for data governance. Whereas data governance is something that is more like how we now view, let's say, Toyota six sigma or Agile, where it's kind of embedded in an organization and kind of culturally and throughout the work that's being done, it's kind of embedded in the network. I think Purview is kind of the same way. So, data governance operates in the same way. It first needs to be adopted and agreed upon by those that will be using it. Purview is the tool, and that's, like, the actual interface, and that's the actual, you know, that's the tool that allows it to be done. But in terms of stability, I think Purview does as good a job as it can of providing a foundation to make stability with data governance as easy as possible for those that manage it, but it does come down to those that are in charge of Purview itself to make sure that it stays stable. Well, I won't just say stable. I'll say that it is as stable as the data governance maturity that exists within an organization. It can't be more stable than that.

Scalability is fantastic considering their pricing since it addresses their pricing where it can have your incremental scans, which can cut down on your computing costs. That's great for scalability. So, if you have more data and you need to scan more and more, you don't need to scan the exact same data again and again and again. You may be okay only scanning the entire database maybe once a week, and then, let's say daily, you have incremental refreshers that help very much, or that help a lot with scalability. I think as more an organization scales and the more that it uses Purview with that scaling, the more you can get out of it, and the more it'll end up helping the organization, the more the stewards that are in their tagging things, the more that data is suggested, and that triggers are set up for ingestion runs. The more it scales, the better it is.

I have spoken with Purview's product team consisting of engineers who created Purview that writes the documentation that will work on bugs hands-on. Also, I've been a part of a lot of conversations, and I provided a lot of feedback in terms of improving Purview itself as a product and how we use it with our clients. So, there's been a lot of collaboration on that.

Considering Purview, I rate technical support a ten out of ten.

There was not one solution. There were multiple solutions like Profisee involved. We also had Atlas and kind of a lot of commingling of those two and kind of piecemeal together solutions between a lot of those to do a Purview does. So not a whole lot, I guess, is what I'm trying to say.

The deployment of the solution was done on Azure. A majority of our deployments for our clients are through a private cloud, and they're locked down with private DNS zones and custom DNS records on the back end if they're on their own servers and private endpoints as well. So fairly locked down, especially with where we talked about a lot of the data being sensitive and a need for it to be protected and monitored. Well, if data needs to be protected and monitored, then it likely is going to be needing something to be secure. So it's all been on a private cloud so far.

I was involved in the initial deployment of Purview. The initial deployments were actually straightforward because we were figuring out about the features, and found that it was a little less feature-rich, so there was less of a question. And it hasn't given us the time to figure out where it works well or where it doesn't work well, and if we need to bring in consultants or can we do this in-house. As Microsoft Purview got a little more mature, the deployments got a little more complex, as we were kind of seeing that there are a lot of ways, like there were a hundred ways to do one thing or less. I think that applies to anything in software. But it applies to Purview as well and where we control access to a given data source while figuring out if we should use the policies and governance features within Purview. We also thought if we should do it in Azure portal using RBAC or should it be elsewhere. So I think we passed that part, and then now we're going to get back to some smooth deployments, and a lot of them will be templatized, you know, whether it be in ARM templates or Bicep like Terraform. Overall, at first, the setup process was fairly smooth to moderate.

I have not been able to conduct any actual ROI analysis on Purview just because it is hard to measure those kinds of intangibles of how those results show up. But they absolutely show up if we consider data to be an asset. Also then, it can also be a liability, just like something that would be like a can of soup on the shelf in a restaurant. That's an asset as it exists then, but also there can be hazards. So, if it's a large pallet, it could fall and cause damage. So, data cost would be a liability. I think that it is not something that currently appears on the balance sheet and can be cut to draw that line, but I think it'll be made much clearer in the near future, especially as we've become more and more data-saturated and we'll be more conscious of the value of data. Regarding ROI, I have not done any analysis in that realm, but I have full confidence that there are real ROIs that do exist and that will show up one way or another in the future.

Price-wise, I think it's very generous and accessible to not just enterprises but small to medium-sized companies as well. I think it's very fair in terms of how they break apart the storage and the actual computing, and that makes it very accessible. So, that is a very big plus for Purview. I think a lot of clients and organizations would feel that way as well.

On the time front, I can absolutely speak about the cost of the solution since time is money. Saving analysts or developers, or engineers time can add or translate into actual savings if you can do the right calculation. But just to stick on that time front, I've seen it cut down, let's say, a given team, or a data governance team, I'd say, in a reduction of fifty to sixty percent of the time that would be spent answering maybe questions for other parts of the business about what a certain data means or where this data lives, they can just point them into certain directions, and they can point them to Purview and have those answers or have or have this questions answered. So that's a big part. And then also being able to have those ties from the business to IT and being able to bridge that gap and being able to have a technical asset ingested from the data side or, from the technical side, and then being able to get the business to interact with that as well. I would say that total savings in terms of on a team-to-team basis, I could see anywhere from fifty to seventy-five percent reduction in time spent on those activities just by having Purview.

After looking through other options, we decided for our clients that in these situations where they're already in a Microsoft data environment in terms of their data, whether it's living in Azure or on-prem, but the analytics that they're doing, whether it be with Power BI, with Synapse and Data Factory, it integrated so well.

Speaking about the importance that Purview delivers data protection across multi-cloud and multi-platform environments, including AWS and Google Cloud Platform, I would say that it's very important, and I am going to lean on the side of this that has to do with data protection in terms of international data protection and various laws that kind of force data to stay in certain places. For example, I have one of my clients, which is a Fortune 100 company, and they have three different main regions around the world, including Asia Pacific, the EU, and the US. They had a lot of financial data that were in these three regions, but they couldn't cross each other even though it was the same data that was being used in the same applications, but it was difficult due to various barriers and regulations for the actual data control. Hence, they couldn't pass between each other. So that was kind of an issue. We had to actually go and deploy multiple instances of Purview. We had a dev and a prod instance in these three regions. That was definitely very important, and that was because of not only regulations but because of their desire to keep their data private, secure, and kind of locked down.

Speaking of how important it is for me that Purview can connect to iOS, Mac, and Android devices and data in other SaaS apps, it is not very important as it's typically with legacy databases and data warehouses.

Regarding Purview's data connector platform for supporting ingestion from non-Microsoft data sources, I think it's very robust, and out of the gate, they included a data connector that you can plug in straight to Amazon and SAP. I believe now that they have some connectors for Salesforce and things like that and Snowflake. So they are absolutely staying right on or ahead of the curve when it comes to making this kind of making sure that the interoperability between other vendors as well is taken into account, which is awesome.

Speaking of how important is Purview's natively integrated compliance across Azure Dynamics and Office, I think it's very important, and it does it very well. I mean, literally without a problem. It's completely seamless, and so it does that perfectly.

Speaking of how important it is to me that Purview was built taking into account critical regulations from around the world, I think that it is very important. So, that's what kind of gave birth to the original classifications for 200 or so out-of-the-box classifications for a lot of data for possible credit card numbers, passport data, and ZIP codes in different countries. They were definitely aware of that, and you could tell from their design process they keep updating those. And so that was definitely built with that in mind for sure.

I have never used Purview for data loss protection. Typically, in these solutions, when we're deploying Purview, they're usually at a stage in their data platform maturity where they already kind of have that taken care of elsewhere.

Speaking of whether Purview helped reduce the number of solutions I need to interact with, I would say that I don't think so because I think data governance exists kind of outside the solution's architecture for these kinds of things. So, I'll say no, but that's not in a bad way. It's just not its job, in my opinion.

In terms of Purview's visibility, I think visibility is the most significant part since that's kind of what it does as long as it's set up correctly and maintained by the data stewards and the business experts. The domain experts that actually know how to interpret these business glossary terms and can link them to data assets, then it is the most useful tool out there.

I haven't used Microsoft Purview's automation. However, I have used Purview's AI since we have had a few situations. While not many organizations have had to do this, we had two instances of Purview in a given tenant, which would be dev and prod. So, we would use some APIs and some Purview CLI. We automated a lot of those processes. Also, it's workflow automation in terms of approvals and sign-offs from a given definition being changed or a term being added and being able to send that to the right people, especially by sending an email to the right group for approval. So the automation piece is huge. I have not yet been a part of, or I have not, used any of the AI-related features.

Speaking of how automation affected my speed and accuracy of risk detection, I would say it went very well, especially when dealing with financial data or when dealing with possible credit card data, personal information, and health data. In PII or PHI, being able to have those classifications kind of flag that data if it comes through, and then being able to mask it or identify those is something huge. So, that's probably one of my favorite features considering how it's integrated into the various protection services that Microsoft and Azure have to offer.

Speaking of how automation affected the quality of the insights I have, I would say it has more to do with speed on that front. Also, it's going to be as good as the developers, data stewards, and experts working on it. As you know, it'll be as good as the effort that they put in to constantly go through and curate or update a certain classification, including what the threshold may be to make sure it's catching all of where it needs to cache for risk detection or risk mitigation. And there's definitely a one-to-one relationship that is a very linear relationship. So the more end users and stewards are involved, the more you get out of it in terms of every feature that it has to offer.

Regarding Purview enabling me to show my compliance in real-time, I would say that I believe that there are some integrations that I have not personally, which I think came out in Purview the last time when I was very heavily involved with Purview, which is maybe four or five months back. So, I have not used those yet. But in terms of real-time, I don't think we have used it yet. But I'm sure we'll do that eventually.

Regarding whether Purview helped to reduce the time to action needed for insider threats, I would say it has, especially in the context of early detection for possible sensitive data, like health data, credit card information, and things like that, and being able to alert the right parties. So, they can make decisions on whether they need a mask or not or if there was a problem with the actual data source, whether the data kind of slipped through, and if it should have been taken out. So, definitely, it works in that kind of early detection there and in terms of the speed of making those decisions.

I would say that Purview has helped save time and money for my clients. So, it's definitely a long road for data governance, and a lot of the benefits aren't necessarily very tangible, so they might not show up on the balance sheet. However, data is an asset and needs to be treated as such. Hence, those who implement these data governance programs and who use Purview to do that are absolutely gonna be ahead of the curve, and we'll save countless dollars and time. I mean, the savings start immediately. So, down the road is when you really see those effects come up.

I would agree that Purview has affected my ability to stay on top of compliance.

Regarding others who are looking into evaluating the solution, I think they should prepare by making sure that the individuals or the teams that internally exist that are in charge of or that have a very large say, and how data governance is or functions within that organization, they need to be on board and they need to be willing to kind of to go through the motions and be okay with having an iterative process once Purview is deployed. It's not all going to be solved on day one, and they're not going to have everything that they need right away. Again, it's a tool that their data storage and domain experts can use, I would say.

As I am trying to be as unbiased as possible, I would suggest those planning to use the solution contact Microsoft since there are a lot of financial incentives and reimbursements, and discounts that exist right now, also in the past, and in the future, I would assume that for Purview's demos and PoC engagements. So, trying to find or just having the first part, which is the internal understanding of what data governance is and what they want out of it, while the second part is Microsoft does very well by tucking in partners that can show them ropes.

Since there were some features recently that maybe I haven't been able to test, I rate the overall solution a nine out of ten.

We use it for the creation of enterprise PSA, as well as for our enterprise data catalog and data governance.

This is a type of pilot project or proof of concept. Our organization wants to see how it works, how complex the maintenance side will be, and how effective and efficient it will be.

It enables us to show compliance in real-time. However, because it is a pilot project for us, we are not showing the information to other parties yet.

It has definitely helped us reduce our time to action on insider threats. Finding proper data is always challenging in a big organization and this is a really positive aspect of Purview.

The first thing that we were particularly interested in is data lineage. It makes it pretty easy to connect and scan data sources, and do any kind of additional steps like Synapse or SSIS processing.

Overall, I would laud its ease of use, intuitive interface, and easy navigation. It's a very user-friendly product that is easy to work with.

In addition, we considered the issue of multi-cloud and multi-platform environments when we were trying to identify the best tools that would serve all the requirements of our organization. Currently, we are not directly pulling data from AWS. We are working with Azure. But we have the potential in the future to consider such options. This feature is very valuable because we are planning to establish automated data-pulling from AWS.

And the solution's data connector platform for the ingestion of data from non-Microsoft data sources is very easy to use. I investigated what data sources it can work with. It's certainly easy to work with all Microsoft data sources like SQL Server, Synapse, and data lakes, but it also has great functionality working with Oracle. And of particular interest to us is the ability to pull data from Excel, CVS files, and other types of flat files.

Another particularly nice feature is its integration with other services. This is one of the biggest advantages of the solution. For example, if you need to store a key for database access, it's easy to integrate with key vault services.

And a feature that is very important to us is the fact that Purview was built taking into account critical regulations from around the world. Before we started working with Purview, we put great effort into evaluating its capabilities and functionality. We created a comparative matrix with other tools and this was one of the factors we included. We considered it a big positive for Purview.

The scanning of data sources can be scheduled to execute automatically. That is the only automation we are using within the solution.

One area for improvement is the detection of data types. This is really important. It has some of that functionality, but I consider it very limited. Maybe they can add some custom programming, or machine learning could be particularly useful for the detection of the nature of the data.

If it could tell us, for example, when a given type of data is a social security number, that would be helpful. Currently, we need to open it and determine what the format is. We would like to know if a given type of data is PII data.

I have been using Microsoft Purview for almost two years.

I found some defects, through my interaction with the solution, in the reporting system. Purview has its own reporting and I found a function that apparently is not working. There are some bugs in that section, but other than that, everything looks okay. It works well.

We haven't experimented with scaling it, but my perception is that it's very scalable.

We haven't had to contact Microsoft's tech support.

We didn't have a previous solution.

The setup was pretty straightforward. It took us about one week, but a lot of the dependencies are not directly related to Purview. We had to obtain a password and user and that takes some time.

Microsoft takes care of patching, so I haven't seen any need for maintenance by us. It's on the cloud and they take care of it. We don't need to worry about that part at all and that is one of the big advantages of a cloud system.

We did everything in-house. I had architecture and administrative support, but I did the actual work with Azure alone.

I'm handling the proof of concept work. No other organizations have access to it. We don't want people to start using something that is not really complete yet.

Because this is a proof of concept, we haven't seen direct cost savings, but we see it having a positive impact on development work, particularly for BI folks who are interested in self-service and building reports for themselves. This is a very good exploration tool.

Microsoft Purview is the best option I have encountered when it comes to price. Maybe some of my information is outdated, but Microsoft offered it so that you could use it almost without paying. This was a really nice aspect. 

I don't know how Microsoft's pricing stands today, but for us, it was very cheap and effective. We are still paying some money but it's not significant. 

They should look to keep the pricing moderate in the future.

We compared Purview with other solutions, but not on a practical level. We only compared it with the information we were able to find on the internet, including the documentation and the functionality, but we did not install or do anything with the other solutions.

For a large enterprise, it's very important to have tools like this. It's especially useful for IT people who are dealing with data science or who need to find or investigate data or find out what data an organization has. It's also useful for those who maintain a central data warehouse and for people who want self-servicing business intelligence.

At this point, Purview has not reduced the number of solutions we interact with. In the future, we will consider that option. At this point, it operates as a standalone for us, because it is a pilot project in our organization.

Microsoft Purview is not yet a leader in data cataloging or data governance, but I'm pretty sure it will be, given how the tools are evolving and the functionality it already has. At the same time, I'm certainly looking for some of its functionality to improve, including easier integration with Microsoft Power BI.

Our IT team, data engineering team, and data science team all rely on Microsoft Purview for a unified data repository. We maintain a table for storing sensitive data with segregated access. To readily understand the data types within, anyone can simply leverage Microsoft Purview's classification system. By selecting a specific classification, users gain access to relevant table details.

If we create keys for non-Microsoft data sources, we can leverage Purview's data connector platform to connect and ingest the data.

It is good that Purview takes into account critical regulations from around the world.

Microsoft Purview has greatly improved our organization's ability to view and share permissions with minimal human intervention. Searching for desired information is now a breeze. However, integrating permissions for access in large environments can be a stumbling block. With ten departments, granting access to a single table for one department is straightforward. But when access to multiple tables is required, they must go through the main administrator for verification.

Microsoft Purview shows us compliance in real-time.

Data segregation is the most valuable feature. This provides clear visibility into the hierarchy of data flow, including its destinations, loading points, and table updates.

While Microsoft Purview offers data protection across multi-cloud and multi-platform environments, granting access in such complex settings can be lengthy and expensive.

Every scan we perform incurs a charge, making exploration quite costly.

I would like to have complete video documentation for training.

I have been using Microsoft Purview for six months.

In my experience with Microsoft Purview for our data governance project, I'd rate its stability at a seven out of ten. While it excels during exploration phases with use cases and demos, we encountered some challenges during implementation.

Each time we contact Microsoft support, the support person seems new and has to look into the issue because they've never encountered it before.

Neutral

Deploying initially is simple. It only takes as long as creating a storage account in the Azure portal, which is quick.

Microsoft Purview is expensive.

I would rate Microsoft Purview a seven out of ten.

Our current clients are medium-sized businesses.

In the six months we have been using Microsoft Purview it has not required any maintenance.

Microsoft Purview has several built-in solutions, including data loss prevention, e-discovery, life cycle management, and information protection. It functions as a DLP tool and includes a compliance portal that enables integration with various other solutions to ensure compliance. Therefore, it provides readily available integrations.

The solution deployed as a SaaS.

It is crucial for Microsoft Purview to offer data protection across various cloud platforms because many customers are now utilizing cloud technology. It is not always the case that everyone will use Azure exclusively, as many customers may have multiple cloud vendors. Therefore, it is essential to support multifunctional or multi-vendor flows to meet the needs of these customers.

To ensure proper functionality and data security, we need to accommodate a wide range of operating systems. Therefore, our solution must support multiple vendors or be multi-tenant, allowing us to have visibility over all devices within our network.

It is important that Microsoft Purview is globally compliant. The solution comes with pre-defined rules and policies, providing us with a wide range of capabilities to manage multiple geolocations simultaneously.

Microsoft Purview's data loss protection feature is helpful in remediating policy violations as it provides extensive forensic data. This includes information such as the user, the activity performed, the starting point, and the flow path of the activity. Additionally, the tool has forensic analytics capabilities that enable us to identify and prioritize policy violations effectively.

Educating users on data loss protection can be straightforward. For example, Microsoft's product includes integrated guides on the console. Whether the user is an individual or an administrator, if they are unsure how to use the product, there are supported guides and links available to assist them. This makes the experience easier and ensures that best practices are followed when managing data loss protection.

It was important that Microsoft extended Purview's data loss prevention to Mac OS endpoints because it enables us to manage all devices on our network through a single console.

Microsoft Purview has simplified my work with its effortless deployment.

Microsoft Purview has decreased the need for multiple solutions to communicate with each other. If we were to discuss other DLP solutions, an additional agent would need to be installed. However, Purview utilizes Defender, which enables it to capture data and obtain all necessary information without requiring the deployment of any other agents. Therefore, there is no need for any additional agents to be installed.

The solution provides visibility into the state of our organization by giving us inventory details and maintaining an updated version or list of the inventories that are currently in use. This allows us to have clear visibility of the devices we are using, specifically end devices. It also enables us to easily manage non-active or disconnected devices.

Microsoft Purview enables us to show our compliance in real-time. 

Microsoft Purview includes multiple predefined sets of compliance rules that can be compared to our specific compliance requirements. We can then map these rules to our policies.

The agent saves us the time and effort of manual reporting. When the device is active, the agents also become active and start reporting. The reporting process takes the same amount of time regardless of which policy we use.

Microsoft Purview helps us to maintain compliance by providing a checklist of tasks that need to be completed, along with mapping.

The ease of integration with other Microsoft solutions is the most valuable feature. 

The availability of pre-designed policies tailored to specific geolocations and customer requirements is also a valuable feature.

The Microsoft Purview data connector platform, which supports ingestion from non-Microsoft data sources, can be somewhat complex. For instance, when using Linux or Mac OS, additional agents are required. However, deploying these agents can lead to high resource consumption, such as increased CPU, hard disk, and RAM usage.

The performance has room for improvement. 

I have been using the solution for three months.

Microsoft Purview is a SaaS platform, so its stability is very good.

The scalability is all based on the license. 

The initial setup is straightforward. For our implementation strategy, we began by collecting inventory details. We then identified the supported devices by their operating systems, separating them into Windows, Mac OS, and Linux. For Linux devices, we created software packages offline and deployed them to the IT assistance team. After the agents started reporting on the console, we restarted the Linux devices. For Windows devices, we were able to connect directly. First, we were in the POC phase, testing a few devices and developing the rules and policies, followed by applying the rules to all the devices.

The number of people required for the deployment depends on our infrastructure and the size of our environment.

The implementation was completed in-house.

I give the solution a seven out of ten.

As a consulting firm, we utilize several Data Loss Prevention solutions such as Symantec DLP, Forcepoint DLP, and Microsoft Purview. Our recommendations to clients are based on their specific needs and financial plan.

Microsoft Purview comprises various solutions, and I recommend acquainting oneself with all of its aspects to make the most of the tool.

The main use case is DLP, for alerts and insider risk management.

It is very important for our organization because ours is a very data-heavy organization. We process a lot of data on a daily basis, and we need to check where the data is coming from, how it is being classified, where it is being sent, and whether it is being used by the intended user. We need all the security controls.

We have clarity and visibility at a granular level. We can check data at rest or in transit—where data is coming from and where it is going—on-premises as well as in cloud solutions. That's how it is helpful.

The attack surface has been reduced significantly. Earlier, we didn't have any visibility into it.Now, we review every log to know where data is coming from and where it is going, as well as who is handling it. And all our employees also know that if they are doing anything mischievous, someone is watching them.

In terms of time saved, we write a policy, get an alert, and work on it. It has reduced a lot of human intervention while checking all the logs. It gives us the log source that we can check directly.

The best part is that I can create classifications per my requirements. I use it to classify multiple platforms like AWS, GCP, Azure, and different file sharing systems.

I also like the auto-labeling and the encryption of data when it is being sent out of the organization.

We also need complete coverage for every device that is connecting in our environment, so it's important that Purview can connect to iOS, Mac, and Android devices.

We just started using Purview DLP for macOS endpoints. We deployed it and
we are fine-tuning the policies. The macOS support is a very important factor because most of the teams handle data. We need to check whether they are sending any source code and how they are handling the data. We need to know whether they are using a key management system to handle secrets.

The fact that Purview was built to account for regulations around the world is important for us. It has to keep updated regarding all regulations because our users are across the globe. If we don't know the regulation data, retention policies, and other policies regarding the data, we may not be in compliance with those policies, and our company will have to pay a hefty fine.

I have been using Microsoft Purview for about two years.

It is a stable product. The stability is an eight out of 10.

Support should be improved in the form of good documentation and video lessons where a person can check things out. There is a community, but it takes a lot of time if we want to get an answer to a question. There should be an easy-to-reach place where, if you're stuck, the support system is there.

Neutral

The deployment was straightforward. We got help from Microsoft support's FastTrack. They helped us deploy it.

As for maintenance, we have to update the policies.

Overall, the only problem is the support system, which is a bit costly. But the product is good.

Our process for potential policy violations is to first check whether something is a true positive or a false positive; we check the document that is being sent. We then check it from our team's side, the responder's side, and then we check it with the compliance team. We look at the policies that have been violated and also check the business use case where something is being sent outside the organization. It's a manual effort for one of our teams.

As for educating users on how to best handle sensitive data, that is done by our compliance team. They do awareness training for our end users and send them awareness emails as well as monthly connections where they give them awareness information about how to handle sensitive data.

We are only using Purview for DLP on the incident-response side and for data protection. So I'm not sure whether it has reduced the number of solutions in the stack we are using, but it is very useful for us.

And although I'm not involved with the compliance side, we can check the compliance dashboard to deal with compliance-related issues. We have a dedicated team that checks it.

We use it for the text of emails. We use it to keep track of groups as well. We also use it to go through the mailbox of a former employee to retrieve the mailbox for a new employee. We use it for e-discovery and content search in emails across our organization.

We use it internally for just a few use cases. I know there are a lot of connections between it and other Microsoft applications and multiple clouds, but we don't have those use cases currently.

We carry out implementations for customers on these particular solutions. For example, if we're selling a Microsoft 365 solution to them, we use it to secure their data, especially their emails through backup, and SharePoint.

It has improved our ability to retain data and retrieve the data that we need at a future date.

We are also able to see how sensitive data flows across the organization, so it has been very helpful in telling us where that data is originally from and where it is being sent to. We have the ability to track it and control and restrict it from going outside our organization. We've benefited from that a lot.

Purview also enables us to show compliance in real time. We can see what the requirements are and then we can apply them across the organization. That has been very helpful.

It has also helped us to stay updated and make sure that we are not out of compliance. It keeps us updated with any new policies that are required for organizations like ours. That's of great value to us.

Another advantage is that it has definitely reduced the time-to-action on insider threats, although we don't measure that at the moment. But from experience, we can tell how much it is saving us in investigations, compared to before we had the solution.

The most valuable feature is the information protection, the way we're able to retain emails.

Another aspect that is very important is that Purview has been built to take critical regulations from around the world into account. It gives us trust that all of the compliance requirements are being met and that we just have to take care of our data. We don't have to worry about whether the regulations are being met around the infrastructure and we can just focus on our data. It's very important to us to have that level of trust in our systems.

Also, Purview's data loss protection for remediating policy violations is very good. It gives you the opportunity to know your data and apply policies around it. If those policies are flouted, you can always track what's happening. You have options such as alerting the person who is committing that action, or you can take automatic action by blocking, for example, an email that is being sent externally. It's very useful.

I've been using it since as far back as when it was Security and Compliance Center, before there was a separation between compliance and security. That happened around 2019, so I've been using it since 2019.

The stability is great. Like every SaaS application, there are infrastructure issues, maybe once a year. Overall, it's good.

The scalability is also great.

The service level agreement is excellent and the support follow-up is also great. They have good knowledge.

Positive

In terms of ROI, my bosses take care of that calculation, but I know we are getting benefit and value from Purview.

The categorization within the licensing could be improved. There are a lot of solutions within Microsoft Purview. If the licensing could be a bit clearer and the solutions could be better categorized according to function and across multiple environments, that would be excellent. The licensing is very confusing.

The pricing, for the solutions and value being provided, is fair. But that ties back to what I said about the licensing. There are a lot of standalone solutions you can get, and there are different licensing options for them. Depending on what you need, you can have a cost-effective solution; you can figure out your cost and benefit. It's affordable.

We are still exploring whether Purview can help reduce the number of solutions that we have interacting with each other. There are a lot of solutions within Microsoft Purview, but we still have some data that is on-prem and we are still looking at how we can expand and connect to those areas. It's something that is in progress.

In terms of maintenance, it's a SaaS solution, so the applications are automatically updated. There's almost zero maintenance. We do have to take care of configuration and updating preferences. I am able to handle that myself.

My advice is to develop a clear use case and a roadmap, perhaps from a consultant if you don't have the time, or spend some time doing research on it, because there are a lot of great solutions within Microsoft Purview. You need to have a strategy for the way you combine the solutions together.

The primary purpose is to catalog all the different data sources. The idea is to get insight into what is available and, more importantly, to document and better understand the data quickly and easily. You're not doing any manual work, you're just scanning sources, which means you can automate it. It automates the majority of the documenting process.

The most important benefit is its data documentation and the data catalog. It saves a lot of time compared to doing those things manually. Normally, when you want to describe your sources and get an overview of what is available, it takes a lot of time for data engineers and other people in the company to document the data. That whole issue is eliminated by using Purview.

Also, the fact that you can easily view your metadata helps with data exploration. A lot of the time, there are so many sources at a company that at some point, most people don't even know what is available anymore. A really key feature is that not just one person but lots of people can access it for the same price, to do data exploration. They can all see what is available and decide what they want to see in a company report, for example.

The beauty of Purview is that it's all about a central location where everyone goes. I wouldn't recommend creating multiple Purview instances, although you might have one for production and non-production. But, ideally, you would just have one Purview for your entire organization and then provide access to multiple people to make use of it.

On the documenting side, in particular, it saves a lot of time, and time is money, especially when you are dealing with people entering data and information into Excel. That can be replaced by Purview and that saves a lot of time. Purview also gives you information that you can act upon. Instead of finding out too late, you can act earlier, and save money in that sense.

What I like the most about Purview is the fact that you can really easily connect to data sources and retrieve the metadata in a batch manner. Instead of having to manually write down which tables and columns exist and then describe them, you can do that process in one go, by simply connecting to a source. That's a huge time-saver and a great benefit of Purview.

The solution takes into account critical data compliance regulations from around the world and that is one of the most important aspects of Purview. New laws are being enforced for data compliance and a lot of companies have a great interest in this feature of Purview. I think Microsoft is going to be focusing on that for the next couple of years to help organizations improve on data compliance.

In terms of reducing time-to-action, if you set up a clever rule that gets applied to your scan—it would just take some time to create that rule—in theory, whenever you are scanning your data you could identify something that is going on and act upon it. But I haven't seen that in practice yet.

The fact that Purview delivers data protection across multiple platforms, including AWS and GCP, is really important, but I feel the tool can mature further in that area. You can set up rules and scan your data and then you can figure out whether your data is secure and compliant, but feel that Microsoft could improve on this and add more features to the tool. I think they will do so over time. The solution has only been generally available since last year, so it's still quite early in terms of maturity. The multiple platforms feature is very important and there is potential there.

A bit of a downside is that although you can explore the data, that creates a great interest in data lineage or the data flow. How does it go from a source to a platform to a Power BI report, for example? It is possible, to some extent, to see that with Purview, but the lineage feature requires some manual work on the development side or more work from Microsoft to improve on it.

The data lineage is effective and useful when you are using all Microsoft products, but as soon as there's any complexity or you have a different tool in between, like Databricks for data transformations in your platform, for example, the lineage isn't going to be added in Purview because there is no connection to it. On the lineage side, a lot more can be done, but there is a lot of potential.

An additional feature I would like to see is in the following scenario. Suppose you have your sources scanned and you have all the tables listed in Purview. Right now, to update and label them, or to group them, would take a lot of time because you have to manually click on the assets and the tables that you have. But given that a database can have hundreds of tables, it would be helpful if you could update the assets in batch and, possibly, multi-select them. That would be a nice addition.

I have been involved with Microsoft Purview from the private preview stage, which was about two or three years. At that stage, it was only being shared with certain companies and nothing could be shared externally. In that phase, I got to share what I learned from the tool with Microsoft.

I haven't used it all the time since then, but more recently, I got to work with it for a few months so I got to see the latest update and changes that were made.

It is quite stable in terms of the scans running and not failing. It's not going to be slow or not function when you do an action inside of Purview. The stability is great.

It's really scalable, just like most things in Azure. You can add to it but it gets more expensive. You can add as many sources as you want, and the scanning of sources goes quite quickly, even for really big databases. The reason is that you're not copying any actual data, you're only getting the metadata, meaning a description of the tables and the schema, et cetera.

I haven't had to use Microsoft's technical support for Purview.

From my experience with other Microsoft Azure tools, the support is not bad, but it might take some time. Once you get someone working on it, an issue always gets resolved, but it can take a bit of time to get the right person involved to help you out.

For documenting data I have used Excel. I've seen huge Excel files with lots of data descriptions, and they took a lot of time to create. 

Also, on the data quality side of things, I have used an Azure data platform: an SQL database and a Power BI report. For example, if you're scanning data and you apply a rule to check if a column is empty so that you can classify it as "empty column," that would be a data quality rule. Instead of using Purview, I have used Azure.

It's very straightforward. There are not that many fields to fill in initially. Connecting sources, the first step, didn't take a long time. You really quickly get to see things, especially when it comes to Azure sources. It's all integrated so you can connect really easily. You just need to have authentication rights assigned. So connecting is quite fast.

The deployment is all-cloud. It's all Azure, which makes it really easy to deploy the tool really quickly. And if you have other data in the cloud, you can really easily connect to it.

The second part that takes a bit longer is defining the data, just like you would normally describe your tables and your columns and all your data definitions in Excel. In Purview, that also takes a bit of time. You have to find the way to describe it most easily. You can use the rules while scanning your data and automatically label or classify the data. But creating those rules takes a bit of time: How are you going to scan the data and what rules do you apply?

Getting the resources going just takes a day or two. But to connect to them and make things functionally available takes more time.

It's a one-man job. Even for connecting to resources, all you need is an admin who can grant you all the rights that you need for those sources and you can really easily scan them. The part where you need more people is on the business side because you need to describe, understand, and classify your data. That takes a lot more people because one person might know something about the customer database and a different person might know something about the finance database.

There is an investment of time involved, but once you set up those rules and you have the sources to scan, it automatically checks your data. It takes time to set it all up, but over a longer period of time, you will actually save time and see a return on investment. How fast that happens depends on your organization and how many data sources you have, as well as on how many people are using Purview and how efficiently.

You pay a minimum amount every month for the data map. You scan your sources and the metadata gets saved and then you pay for what is stored, which makes sense. However because there is a minimum amount, in the beginning, you might pay for more than you are using. That's something that some of my clients didn't like. That's why I say it's quite pricey; you're always paying a certain amount.

It would be nice if it went to entirely pay-per-use. For example, on Azure, when you have storage accounts, you pay for exactly how much you store. That would be nice to see in Purview as well. And while it's pay-per-use, you pay for features as well. For example, you pay for the cataloging part, including describing your data and adding labels and classifications. I would like to see a standard price and exact pay-per-use. 

I understand, in practice, that might not happen, but the pricing may be a bit overwhelming for some clients. They will say, "Hey, I'm already paying this much and now something else comes with another cost? Why is that?" It raises questions.

I'm working at a company that is a Microsoft partner, so it's all Microsoft-first.

We did do a quick analysis for a few other companies and there is some competition out there, but the other solutions are quite expensive. They are enterprise tools that are a bit more mature but the license costs $100,000 for some of them. Purview is pay-per-use and a lot of companies are interested in that. It's still quite expensive compared to most Azure components, but compared to the alternatives it costs less. That may be because it's not that advanced yet.

My advice is to involve people from the business side who can describe data and describe business terms. That's what is most important. Otherwise, it's just going to stay a technical implementation and it won't be used, which would be a huge waste. From the start, involve the people with a mandate who can actually start using it in the future.

Regarding its data connector platform for ingestion from non-Microsoft data sources, there are so many sources in the world that they have a ways to go there, but I do feel, especially in the last year, that the solution has grown a lot in that area. All the big, and most-used data sources, like Amazon, SAP, and many other sources, have been added, which is a great step. But if you work with sources that are more unique, the kind that are not used by many other companies, those are not available and you would have to write code in Purview for them. You can use the API that is available and you could insert metadata and lineage information into Purview, but that is a manual process. You would have to develop that for specific sources.

Purview's natively integrated compliance across Azure, Dynamics 365, and Office 365 is also quite important, but I haven't worked on that myself.

In my experience, Purview hasn't come far enough yet to help us reduce the number of solutions that interact with each other. We use Purview right next to all the other tools, which is okay. It takes a lot of time for a company to adapt to using Purview. You can scan data quite easily and figure out how to apply rules and classify and document your data, but you still need people to adapt and make use of it. I haven't really seen that last part very much in practice yet. It takes a bit of change management to get people to make use of it properly. As a result, it hasn't replaced tools yet for me.

Purview doesn't enable you to show compliance in real-time, but you can schedule how often you scan your sources. When the sources are scanned and added to Purview, they become visible and you can see if you're compliant or not, but that's not real-time. You can schedule scans daily, for example, but then you have daily data sets rather than real-time data.

Overall, the potential for this solution is really large. Data management is extremely important and Microsoft is investing very heavily in Purview. Right now, it's not quite there yet.

Our client was looking to do data governance and we decided to implement Microsoft Purview for them. The client did not have very specific requirements, because they had no idea what to do with data governance and they wanted someone who could kick-start things. We implemented the basic functionalities for them.

Our client was impressed because they now understand how much sensitive data they hold. They now know they have to assign someone who can work as a data steward, someone who will manage and protect the sensitive data to make sure that no one else can access it.

It has also helped our client save time when compared with doing data governance manually. It could be up to 80 percent faster because, if we had to do it manually, it would take a lot of time to create a script to scan all of the data in the data repository. With Purview, we can configure the connectors and just hit the button and it will do the rest.

Overall, Purview is quite good for helping you stay on top of compliance.

One of the best features is the classification rules, especially the scan rule sets. They are really useful, especially when we need to understand the current data the company has to ensure that all the problematic data can be put under someone's responsibility.

Also, although it is not automatic, managing the business glossary terms is valuable. If we have to work with other people who are also data governors, it's important to understand how the current data catalog is going to work with the business. We can do it manually to flag the business terms that we would like to handle.

It is also important that Purview delivers data protection across multi-cloud and multi-platform environments. Microsoft has built-in tools that can scan risky data that could be sensitive. It also provides connectors to other data platforms or clouds. These abilities are important for anyone who wants to see whether their organization has sensitive data, and how much.

In my understanding, because most of the connectors only scan databases or data lakes on multiple clouds, it is certainly not going to provide direct protection on other platforms, like iOS or macOS.

Also, Purview has no data ingestion capability. Of course, it can scan, but it won't import any data into Purview, just metadata. That is still important but it doesn't do data ingestion.

Another area for improvement is in managing the business glossary terms. If they could provide the same type of method that we use to configure the scan rule sets, that would be helpful. Currently, there is no option like this, so we have to do it manually. Automatic detection would be great.

I would also like to see third-party plug-ins. For example, there are several data quality or data management options on the market. Because Purview is not a 100 percent data governance solution, it would help if we could pick some of the good capabilities of other products. If we could plug Purview into the AI option from Ataccama, it would be enhanced immeasurably.

We implemented Microsoft Purview for a customer's company about two or three months ago.

It's stable. There were no weird glitches.

I think it's scalable, although Microsoft hasn't mentioned this explicitly. But it's a serverless solution, so it would scale. It just depends on how much data we need to scan.

We haven't had a chance to talk with Microsoft support because we haven't encountered any problematic situations.

Our client did not have a previous solution. They had no idea which one to go with. They just wished to have data governance and that's why we came up with Purview for them.

The initial deployment might seem complicated at first, but Microsoft has a guide with basic documentation on how to start and it's quite clear. Newcomers would need to understand the documentation first before they could implement this.

For example, for the resources in Azure, it might require the subscription owner to turn on some features. And in terms of the connectors, they would require an additional resource to store the data connection credentials called Key Vault. We had to implement all the resources to work along with Microsoft Purview.

It took about a week to get things up and running, but that was for the pilot solution. We did not have any very complex data to scan. We had to gather the information from the client about how much of the data they wished to scan, and when we got all the information, we started with the resources. 

We had three people involved in deploying it. One was responsible for most of the Azure connections and permissions. Another was the data governance operator who decided what to scan. The third was a data curator who made sure that all data was in the correct classifications.

It took about two weeks, after we implemented Purview, to realize the benefits of the most important features. The way that we implemented it, it was natively Microsoft-based, so there were not many other solutions to consolidate.

Compliance is not shown in real time because it might take some time to scan, but it depends on the amount of data involved. It's going to take some time to compile everything and show compliance.

Purview itself is on the cloud, but there is an on-premises database server that we need to connect with. Currently, there is no maintenance involved. It is done automatically.

The pricing is moderate. It's not too expensive, but it's not the most competitive.

In terms of additional costs, there is an option for advanced data lineage for lineage visibility with Power BI, but because our client did not have Power BI we didn't use this.

Although I haven't had a chance to work with Dynamics 365, I can see that Purview could be connected to that too, and I think it is important. And as far as I know, the classification rules built-in by Microsoft only apply to EU GDPR, as well as American and Canadian sensitive data protection. For all other regions, there are no options just yet.

If an organization wishes to do data governance that is not too advanced or sophisticated, Microsoft Purview could be the perfect fit for them, but they should, of course, do a PoC first.