Twilio Segment

I used Segment as a partner and a go-to-market partner for a customer data activation practice, and we were partnered together for two years very closely. Subsequent to that, I was still partnered with them, but we did less work together, and today Segment is a CDP that we would use, but my relationship with them is not as strong.

My main use case for Segment is for customer data activation.

I capture customer data, understand the patterns and trends going on, and test against those understandings in an omnichannel manner. The primary use case is for omnichannel orchestration and how we could personalize and make a better experience for an individual based on our understanding of who they were. Thus, the primary use case would have been personalization.

Segment is similar to most other CDPs in terms of its offerings, and where they differ.

At the time, one of the differentiators for Segment was the fact that they were part of Twilio and there were multiple technologies that would integrate, such as Auth and SendGrid, which created a whole omnichannel experience.

I really enjoyed working with Segment's partnerships team, as I did not implement Segment for myself but was a consultant that implemented it for others. Their partnerships group, when I was working closely with them, was very strong and very good. Plus there were strong market development funds to tap into, and we could jointly show up at conferences together and be thought leaders together, so I thought the program was good.

When we were at various conferences, we were able to show the specific use cases for customer data activation in retail and retail use cases and demonstrate how Segment connected to all the various Twilio components. This helped people see how it could come to life, how a call center could act on customer data, and how an individual could be authenticated so easily. Thus, us being able to demonstrate that and show the business value jointly worked really well.

Integration was hard. The integration of any CDP is hard, and the go-to-market messaging is that it is easy for any CDP, and it isn't. It would maybe be easy if a customer's data was in perfect order, but that is generally not the case.

The professional services were a little too simplistic. They had great people, but implementing a CDP is quite challenging, and I don't think that Segment teams thought of the whole architecture and the ecosystem and how Segment was one part of it. They became a little monomaniacal on what is the one use case and didn't recognize that data could live anywhere and needed to be harvested and transformed, leading to a hard time seeing the big picture. When we worked with customers who were using the professional services, they weren't always happy with the implementation because it was a little too simplistic and didn't live up to what the sales team said it could deliver.

I have been working in the marketing field for over 15 years.

I would rate customer service as 4 out of 10.

We deploy Segment for customers, so it is not specific to us. Therefore, it would be multiple different implementations.

I would say it is important to have a phased activation plan, think about your use cases, think about your data governance, and consider the environment that you are bringing Segment into. Having the technology is not a magic bullet, as there are many other factors that will make it successful. Really think about the customer experience you are trying to drive towards and work back from that. I would rate this review overall as a 7 out of 10.

My main use case for Segment is as a CDP to create a centralized view of the customer and help integrate the activation of audiences to downstream services.

A specific example of how I use Segment for this purpose is that we collect data from different data sources, and we then leverage the activation layer within Segment to populate data within Braze as the customer engagement platform. In practical terms, this could be somebody browsing our website or interacting with our loyalty system, and based on those user behaviors, we share those events and the data with those events with Braze, which is then used for personalized communications such as email, SMS, and push notifications.

That is the primary use case, and we have also implemented custom functions and transformation layers within Segment.

The best features Segment offers include the ease with which we can set up new data sources, the protocol structure they have for validating that the data is coming in the correct format, and the connection with back-end databases, mainly with Databricks where we both receive data via Databricks as well as push data to Databricks. The audience activation layer is also significant for our use case, and the ability to connect directly with the back-end database is very useful.

Segment's Databricks integration has helped our team significantly. In many cases from an audience standpoint, when building audiences, it often relates to customers as well as other data structures. With the database connection, I can create audiences, nested audiences, and other structures that connect customer data with many other types of data structures relating to loyalty, product, and purchases, and that has been a very powerful feature for us.

Segment has positively impacted my organization by making working with customer data a lot easier for us, as well as improving the speed at which we can ingest from new data sources and push to different activation layers. I have seen improvements with our time to market with Segment, which has improved dramatically due to the ease of use, allowing us to handle many use cases that were sidelined for years and now can be completed in days to weeks instead of months to years.

I think Segment can improve in terms of identity management, possibly in some of the protocols by adding more flexibility. I also see the need for more free-form searches such as regex or conditions that are not case-sensitive, which would be tremendously helpful.

I have been using Segment for about two years.

Segment has been stable in my experience.

The scalability of Segment for my organization's needs is very good.

Segment's scalability can be measured in terms of transaction volume, which it handled very well for us in terms of millions of records, and in terms of new use cases and functionality, we managed to find workarounds across various use cases by leveraging their functions and different activation layers, showing that from a scalable perspective, it performs quite well.

I have had experiences with customer support from Segment, particularly when we were learning the product and expanding our use cases. Their support team was extremely helpful in resolving issues, mainly when we struggled with the Databricks connectivity.

I did come from a previous system before Segment, but I am currently unable to recall what the system was, and it just was not adhering to all the needs we had.

The integration process with Segment was generally straightforward, although we faced a couple of challenges with Databricks, which I believe were due to corporate policies implemented within our Databricks environment that created the complexities.

I have seen a return on investment with Segment, primarily around time saved, and it also means we have had many more business-related users able to leverage it rather than needing development.

Before choosing Segment, we evaluated other options, including Azure Data, which was the next major competitor we considered, as well as Adobe Experience Platform.

For the basic functionality, the learning curve for new users getting started with Segment is very easy.

I find Segment relatively flexible within the defined functionality, and the ability to add custom functions helps extend that considerably.

Segment's documentation and resources provided are helpful.

Segment integrates very well with all our other tools.

I appreciate the simplicity in terms of getting into and leveraging Segment. Compared to some other CDPs, those options felt much more complex to reach the same stage, which is why I valued the simplicity of Segment.

I advise others looking into using Segment to understand who their intended users of the data are, what processing they want to conduct on that data, and how clean their input versus output will be. I would rate this product an eight out of ten.