The best thing about Amazon Personalize is how easily it helps deliver relevant recommendations using machine learning. It reduces the effort needed to build personalization from scratch and automatically improves suggestions based on user behavior. Review collected by and hosted on G2.com.

One downside of Amazon Personalize is that it can feel complex at first, especially when setting up datasets, schemas, and permissions. The pricing can also be unclear for smaller teams, making it harder to estimate costs early on. Review collected by and hosted on G2.com.

Amazon Personalize brings real-time, high-quality personalization and recommendation capabilities—similar to what Amazon.com uses—into applications, without requiring deep machine learning expertise. Review collected by and hosted on G2.com.

One drawback of Amazon Personalize is that it can sometimes feel like a black box. The recommendations are often good, but it isn’t always clear why a particular item was suggested. That lack of transparency makes it harder to troubleshoot issues or explain the results to others. Review collected by and hosted on G2.com.

It offers multiple functionalities for building applications delivered as software as a service, making it easier to create and manage SaaS-based application software. Review collected by and hosted on G2.com.

Because of the multiple functionalities, there are some navigation problems, and it can be a bit difficult to find what I need. Review collected by and hosted on G2.com.

What I like about Amazon Personalize is how quickly it lets you go from data to real, production-grade recommendations, without needing to be a machine-learning expert. Review collected by and hosted on G2.com.

Model transparency is limited—it can feel like a black box. There’s also less room for customization compared with building your own models. Costs can add up quickly at scale, and it really needs clean, well-structured data to perform well. Review collected by and hosted on G2.com.

What I like about Amazon Personalize is its ability to deliver highly accurate, real-time recommendations using machine learning, without requiring deep ML expertise. It also integrates smoothly with other AWS services, which makes it easier to improve user engagement and deliver personalization at scale. Review collected by and hosted on G2.com.

One drawback of Amazon Personalize is that the initial setup can be fairly complex, particularly for teams that don’t have prior machine learning experience. Review collected by and hosted on G2.com.

Real-Time Personalization

No Machine Learning Expertise Required

Integration with Existing AWS Stack

Flexible Recommendation Types Review collected by and hosted on G2.com.

Requires Structured Interaction Data

Cost Can Add Up

Cold Start Still Happens

Retraining Lifecycles Require Thoughtful Design Review collected by and hosted on G2.com.

Supports both real-time personalization (low-latency API) and batch recommendations for offline use cases. Handles data preparation, feature engineering, and missing value imputation automatically. Review collected by and hosted on G2.com.

While it starts affordably, the total cost of ownership (TCO) for high-throughput, real-time recommendation scenarios can become very expensive compared to a custom-built solution in the long run. Review collected by and hosted on G2.com.

It helps me to create individual recommendations for customers Review collected by and hosted on G2.com.

I sometimes dislike because it lacks user interface Review collected by and hosted on G2.com.

Amazon Personalize helps us deliver accurate, real-time recommendations that boost patient engagement and streamline healthcare workflows, while requiring minimal engineering effort on our side. Review collected by and hosted on G2.com.

Amazon Personalize can feel complex to fine-tune, and getting truly optimal results often takes deeper ML expertise along with careful, well-structured data preparation. Review collected by and hosted on G2.com.

with out expertise in ML we can use Amazon Personalize Review collected by and hosted on G2.com.

process timing , less customization and model approach is less Review collected by and hosted on G2.com.