Cube Cloud

We needed Cube  in order to have a robust semantic layer on top of our ClickHouse  database to avoid exposing our projection database directly in our app, and we needed to have sub-second latency metrics for our users.

We directly embed the queries generated by Cube  in our app.

Our software engineering team provides the data team with events coming from ClickHouse . We ingest this data and enrich it with other sources, which allows us to create new dimensions and measures that should be displayed in our app. We did not want to expose our data warehouse or our production database directly in the app, so we use Cube to generate JavaScript queries and put them directly in our customer's app. A specific use case is for our deliverability team, which provides our clients with metrics about email deliveries.

Cube is a robust semantic layer that is really helpful in converting SQL queries into JavaScript functions, and it integrates smoothly in our single page framework application.

Among others, I would highlight the ability to convert SQL queries into JavaScript functions easily, as well as the cache feature that is really helpful in managing our resources.

For the deliverability team use case, we needed to have almost real-time data displaying at sub-second latency. We are using the cache feature that stores the results every five minutes. This allows us to have almost real-time metrics while managing our resources efficiently at scale.

There is something that should be improved. We are providing metrics on email, and in the email industry we have both transactional emails and marketing emails. We have different models for these, but the metrics are actually the same: open rates, deliverability rates, soft bounce rates, and other metrics. There is no way to create a real template that is not exposed directly in the UI. We basically customized it by creating a file for all the metrics, and then we extended our previous views with this template. However, this template is exposed directly in the UI, which is not relevant for us. We do not want people using the UI and selecting metrics from this template.

For the UI, our use case is more for back-end engineering, so not everyone using it is using the UI. Something that could be really helpful when using the UI is the ability to make it nicer and more intuitive. To illustrate what I am saying, we cannot order the fields in the UI. We cannot say that we want organization ID to be on top. It is going to be sorted alphabetically, and I do not think that is the most practical way to manage everything, especially when we have views with roughly one hundred dimensions and of course some measures as well.

We have been using Cube for roughly one year and a half.

Cube is definitely stable.

Cube is really scalable. The one thing I did not test so far is the way it handles nested fields, but I am sure it is doing so properly. If it handles this kind of thing, in the future we could get rid of Omni  and maybe switch to Cube fully.

We did not really have to deal with customer support since we implemented it internally with on-premises.

We do not have previous solutions for this specific use case, but we have plenty of different solutions. We have ClickHouse, and we have data exposed through another semantic layer called OmniVision and OmniAnalytics, which is directly plugged into our data warehouse on BigQuery . In app, we have three different sources exposed: one directly from ClickHouse for main dashboards, one from Cube directly, and one from Omni  semantic layer for self-service analytics.

We considered using the semantic layer of Omni, but it is actually really expensive. We needed to separate our use cases. One thing that would be really helpful using Cube would be to have the ability to generate charts directly and embed them inside our app. I would say we could quit Omni for this kind of feature.

Implementation was super smooth. Within two weeks, we were up and running and the metrics were exposed in our app. We also really enabled a team within the company that was not able to play with data and expose it to the client, especially since this is a very niche team inside the company. We could not measure the ROI per se, but our ideal customer profile and targeted clients were really amazed by this because it was providing them with the way our server is running for them to handle their marketing campaigns. It is more for advanced users, but it is really important for them because they need to see the ROI in everything they pay for.

There is something that should be improved. We are providing metrics on email, and in the email industry we have both transactional emails and marketing emails. We have different models for these, but the metrics are actually the same: open rates, deliverability rates, soft bounce rates, and other metrics. There is no way to create a real template that is not exposed directly in the UI. We basically customized it by creating a file for all the metrics, and then we extended our previous views with this template. However, this template is exposed directly in the UI, which is not relevant for us. We do not want people using the UI and selecting metrics from this template.

One thing that would be really helpful using Cube would be to have the ability to generate charts directly and embed them inside our app. I would say we could quit Omni for this kind of feature.

I would recommend starting with a use case directly, using Docker , and putting it up and running really quickly. Then plug a source. Do not plug many sources at the very beginning. Just try it, check the value proposition, and I am pretty sure you will be amazed in no time. Identify a pain point and try to tackle it with Cube. Once you have done this step, you are pretty much committed to the solution because it works. I would rate my overall experience with this solution as nine out of ten.

On-premises

Amazon Web Services  (AWS )

Cube  is the best absolute best FP&A software, dollar for dollar out there. My organization looked at a few different tools and none of them came close to Cube  in terms of the value that we get from it now. We really wanted three different things for our organization: automated financial reporting, ease of financial review, and assistance with budget and flux models. Cube was the only software that really let a bunch of us non-technical users at my organization accomplish all of our goals without sacrificing anything.

Cube easily integrates into Excel and makes it simple for us to plug it right into our template and roll it forward. Our FP&A team has been able to utilize this software exceptionally well. Their forecasting and budgeting has been top-notch and faster.

Regarding how Cube fits into my workflow, it is extremely simple to set up and easy to run. The website portal is very clean and well-organized, making it possible to create forecast or budgeting scenarios with just a click of a button.

A specific example of how my team uses Cube in our day-to-day work is that above all, Cube has vastly enhanced our ability to get financial reporting done quickly and free up our time to really dig deep into various accounts. This has greatly improved the accuracy of our financial results beyond what you would even believe.

The clean portal and organization help my team by making it easy to navigate and the data collected is very clean and managed in an understandable manner, hence making it very easy to make data-driven decisions.

Regarding the features, customer service is great, customization of financial reports, ease of integration with other tools seamlessly, continuous system testing and upgrades, and easy creation of monthly and P&L variance analysis. Data import and export is smooth and efficient. Monthly reporting and analysis is easy to pull and update.

The positive impact Cube has had on my organization includes additional time for analysis, less than budgeted spend, and more accurate financial results resulting in better decisions. The error rate has reduced from 40 to 50%.

The reduction in errors has affected my team and the business overall by improving speed and efficiency for month-end close processes. Better consolidation of data for long-term trend analysis is evident, and easy P&L creation and variance analysis has been great.

Cube can be improved by enhancing data refresh over multiple tabs. The speed at which data is imported can also be improved.

Additionally, Cube needs to add functionality for headcount planning.

I have been using Cube for four years and a few months.

Cube is stable as I have not experienced any downtime or logging issues.

Cube's scalability is very good because it handles my organization with great efficiency.

Customer support for Cube is very responsive and solution-oriented.

I previously used  Workday  Adaptive Planning and I switched to Cube because it lacks a lot of the features that Cube provides.

I chose a nine out of 10 for Cube because it is very easy to use even for us non-technical users. It has a very intuitive, user-friendly interface. It has helped us improve speed and efficiency for month-end close processes. We have gotten better consolidation of data for long-term trend analysis. The setup is very easy and required almost no help from IT, which is a big plus.

Cube's ability to create custom reports easily on the fly is impressive. It is fully Excel-based and simple to use. Integration was extremely simple. It is simple to create custom reports on the fly, and easy to review financial performance by each department, which enables greater transparency in departmental-level budgeting.

I have seen a return on investment as Cube has streamlined the creation of monthly close packs, helped business partners better understand their monthly P&Ls, and allowed for more granular BVA reporting.

My experience with pricing, setup cost, and licensing is that the price is very cost-effective and licensing is very affordable, making it a great financial reporting tool for startups.

Before choosing Cube, I evaluated other options, including OnePlan.

Cube is well-suited to help save time on financial reporting if you need to refresh the same templates each month. It is also great for building templates and pulling in data directly from Excel or Google Sheets. However, it is less appropriate for companies that run into issues with uploading large sets of transaction data, and it does not have robust planning or forecasting features that you will find in other FP&A tools competitors. I would rate this product a 9 out of 10.

Hybrid Cloud

Amazon Web Services  (AWS )

What do you like best about the product?

I find Cube to be a very sophisticated product that allows me to design complex metrics and dimensions in a scalable and composable way. I also appreciate the useful caching (preaggregation) layer. Having a strong common semantic layer is key for us as a data product company serving dashboards and insights to our customers.

What do you dislike about the product?

The service provided by the cloud team is horrible. We are a paying customer, with an annual commitment, and are having big issues when scaling that are not being solved by them. We don't have visibility into what is happening.

What problems is the product solving and how is that benefiting you?

Cube closes the gap between our data model and business metrics, supporting complex metric design with a scalable semantic layer and a useful caching layer.