Segment

What do you like best?

Easy to try / switch on any product I need

Unfortunately, you cannot set up functions to be part of an existing source

Sync it to data warehouse early

Make sure all analytics systems are synchronised, easy and consistent deploy of new analytics tools.

What do you like best?

Very easy to implement, with connections for lots of systems

The pricing for very big projects is quite high

It gives a perfect view of the customer journey: no info is lost and there's data to take whichever decision to create better customer experiences. You can easily understand which features of your product are more used and which neglected, you can segment your audience and do one-to-one marketing, you know exactly how valuable each user is! It saves time in coding (you code once and it's done for all the platforms/systems you have to connect). You can experiment new platforms very quickly (no additional code required, so no engineer involved and you speed up time!).

What do you like best?

Simple interface, easy setup and incredibly powerful features to capture, govern and protect data and empower the business to self-sufficiently generate actionable insights and audiences, and activate them to any channel - all automatically and in real-time.

My only desire would be to have a more visual experience - highlighting a summary of details and behaviours for each profile, and perhaps some actionable insights that may be of interest to assist in building audiences. This isn't it's primary purpose, but it would be nice to have this feature built in.

It is important to thoroughly and properly consider your tracking plan and define all key customer activity that is important to understanding customer journeys and adopt an intuitive and consistent naming convention to support it. To simplify this process, I strongly recommend getting the Protocols component as part of your package - as this can enforce tracking standards and avoid creating messy data.

I can provide sales, digital, product, marketing, customer service, customer experience, legal and compliance, and data and analytics teams access to a single view of all customer behaviour to create custom and highly targeted audiences for executing interventions to deliver highly personalised experiences to customers.

What do you like best?

Title says it all, the tool allows us to a pass and control data flows on a level outside of code, and is extremely valuable to our core business, but the team that we have been able to work with to this point have been phenomenal. Our CSMs, AMs, and Solution Engineers have all been extremely knowledgeable and incredibly responsive.

The only thing that I would say here is that if you set this tool up incorrectly, it has the potential to be difficult to fix. Make sure that you are documenting everything and have an idea and plan before implementation on what you need and want.

It is a great tool once you get comfortable with using it!

It is our CDP. All of our event streaming data flows through it and it is working so well for our product that we are expanding it to our marketing. It really allows visibility and customization on a level that is outside of code, which is incredibly useful for our non technical team.

What do you like best?

Setting new events is easy and the software is very intuitive.

I can't extract the list of events in order to review them offline (in order to manage which to keep and which to disable).

When working in the "debugger" view, it's impossible to search using "OR" statements, which makes searches more tedious since you need to manually search more than once for different cases and compare the results "outside of the product" (copying data to a notepad for example).

Don't hesitate - it's a great product

We use it mainly to learn about our users' engagement with our software.

Setting new events is easy and the software is very intuitive. The data is flowing in real time which makes debugging (using the collected data) super useful. Managing the events is also super intuitive (using the "schema" view) with quick abilities of disabling/enabling events on the fly, sorting by less used events (or last seen) and many more useful options.

The ability to have a "Testing" profile enables us to develop new events and test them against the current ones, before releasing them to our public platform.

Another advantage is the amount of services you can integrate this product to, having the data flow seamlessly and having "less technical" staff members use the products they are familiar with based on the data gathered by the segment product.

What do you like best?

Very easy to configure and add new sources and destinations. The integrations Catalog is impressive, many options.

The pricing model especially for anonymous users. They will also count on your MTU usage, if you add the JS on landing and marketing pages, this number can go up pretty easily.

We're solving the classic marketing stack problem. Marketing team wants to use and explore many tools and now they can do that easily through segment, since my dev team only needs to add the Segment JS to the website and from there it's pretty much self-service for the marketing team.

What do you like best?

The various speakers especially from IBM. A strong panel explaining exactly the process to implement, and why CDP are so important moving forward and scaling.

I'm absolutely gutted I didn't win a pelaton. jk. Fabulous event .

scaling a startup doesn't happen without strong analytics, segmentation and a CDP like Segment.

What do you like best?

It's one of the most important, but least sexy tools in the stack. If you've ever had to manage ETL in a large organization where there are constant demands for new downstream tools getting Segment instrumented is a lifesaver. They have a huge library of sources and destinations. The documentation is very good. Their customer success teams are highly responsive and often willing to set up dedicated slack channels.

Instrumentation is not a straight forward thing. It's a mix of data, product, and engineering expertise and typically organizations don't have those all in one place (or even more rarely one person). So for start-ups and the tech world, it's a no brainer, but for enterprise, Segment really needs a strong professional services team. Also, Segment could really use a better solution for the on-prem legacy systems that bottleneck big companies. If they're already a redshift/snowflake/react shop - awesome - but most big old companies aren't.

Spend a lot of time upfront on strategy. Understand the data model. It's very easy to slip into rework because of early implementation mistakes.

Helps solve the data teams need for a consistent approach to user behavioral data (product data). Helps solve the need to quickly warehouse and use the data. Helps solve the evolving ETL headaches of downstream tooling. If the marketing team wants Google Analytics, the engineers want DataDog, the product team wants Mixpanel, and the data team wants Looker? No problem.

What do you like best?

Segment provides one API that allows you to plug in with hundreds of different services, greatly increasing the amount of data and the ease at which you are able to collect data. Without having to spend hundreds of developer hours learning and implementing different APIs for each data, tracking, or marketing software, you can still integrate with so many services. Additionally, the payload of the JS loaded on the client is greatly reduced. Segment gives you the ability to playback data to a new service you want to integrate with as though you've been already sending that service data for 6 months. I think Segment is an invaluable part of a modern data and marketing stack.

The cost of Segment makes it hard to use as a startup.

It will save you so much time and money!

Segment makes it way easier to integrate and consistently track data.

What do you like best?

The ability clearly map, collect, and transform data in near realtime has been game changing. The fact we could slowly migrate our old marketing data flows has been helpful and productive. Having clean data among our sites that we can just combine in the same Schema has increased productivity.

The documentation for some connectors is lacking. We have a few third part vendors who we've struggled to migrate because there is almost no documentation other than making a connector.

Do the work ahead of time in your planning. Turn the business questions into measurement. Follow the specs and the onboarding forms. Ask lots of questions. Understand the call types and get a good grasp on how many users / events you'll need.

We have heavily reduced development costs related to data ETL and measurement events. Less outages.

The largest unexpected benefit was have seen is that we can test changes to third parties much faster than previous. Having a single set of semantic events has improved implementations times.