SiQueries - The story of Cameroonian Big Data startup


BIG DATA is coming at us at the speed of light in petabytes, exabytes, zettabytes, very soon in yottabyte (1,000,000,000,000,000,000,000,000 bytes). The amount of data that is being generated is exploding like never before. And within that data lies valuable insights, new ideas that will could transform and solve business problems and provide new business opportunities.The potential of big Data can be unleashed for business users by condensing it and intelligently presenting only what is relevant and contextual to the problem at hand. That is where SiQueries comes in.

With SiQueries, you can easily tap multiple structured and unstructured data points to capture, explore and manipulate data from numerous sources, and then share insights for critical business decisions to stay ahead of the curve.

SiQueries is a SaaS(Software-as-a-Service) data analysis and visualization startup founded in 2013 in Cameroon. It provides massively scalable data warehousing service, allowing anyone with data analytics needs to collect and carry out analysis on data sources (web, mobile, server, sensors, etc.) in real time. Also, you can use data analytics API and build analytics that directly integrate into your app. They are currently testing their solution in a closed beta version with a couple of clients operating in the banking and telecom domain.

SiQueries - Dashboard

The Founder Story

So far, the world recognized Africa as an agricultural economy, but over the last decade, many tech developers are working tirelessly to solve the continent’s problems proving that Africa is also a techie continent. That is against the odds of not having robust programming education available.

Like most others, who enroll in African universities, Ebot Tabi's plan was to get a computer science or engineering degree. But he was more than disappointed with the poor curriculum. There were no practical sessions. Programming exercises had to be written on papers and submitted. The majority of students studying mathematics and computer sciences at his university could not even code, develop or use algorithms to solve common programming tasks even in their final year of studies. Students were told to memorize and reproduce a code. Study materials were so outdated that Ebot had no choice but to switch to a programme that gave him more time to study by himself. Most of what he learnt was online.

Literally, a self thought programmer, Ebot considers himself fortunate to have met some awesome people who he learnt much from: his mentor Noah Pepper and a friend Rebecca Enonchong.

Before SiQueries, Ebot founded a geo-location startup. One of the challenges he faced there was data. At that time, he was new to tools like Hadoop and ended up developing more interest in data, infrastructure and high performance backend services. Later, he noticed, “Business intelligence is mostly just too hard -- too hard to use, too hard to manage, too hard to buy and too hard to get right. Often small and medium enterprises cannot afford to shell out hundreds of thousands to a million dollars for a license." That is why he started up SiQueries. If small and medium enterprises lagged in business intelligence, that would make them less competitive in a market where decision making driven by data has become a key factor. SiQueries addresses this need.

The team

Currently, the SiQueries has a 6-member team. As they had a considerable amount of experience in telecom and banking, they didn’t have too many challenges in terms of technology infrastructure. Ebot feels "extremely privileged" to work with his current team. This year, SiQueries would hire more hands. The main challenge before them presently is how to run the service in a cost effective way -- attain high throughput while the costs of running stay low. The startup hasn't been funded externally yet. They might start looking out for Series A soon though.

The product

SiQueries is SaaS platform with a highly interactive dashboard for data analysis and visualization for small and medium businesses. It lets you connect various data sources such as MySQL, Postgres, Amazon Redshift, Google BigQuery and Oracle and run real-time queries over these data sources to gain valuable business insights. It also helps you aggregate, create data model, analyse and visualize market data in a more interactive approach to drive more values in your decision making. It makes it easy to draw metrics, line charts and pie charts on chart-creating dashboard.

For example, the Social Connectors feature lets you monitor any keyword in real-time and receive filtered, organized and prioritized results from your social layer into your ERP, and then to a single chart. You can also use your SiQueries data with any charting library to customize the look and functionality as you see fit. The team owns all of the IP for the product. They are in private beta mode currently. So you can contact the team to arrange a demo for your business. The offer is open for mobile app creators, web developers and others who are interested in building analytics into their apps or services. As of now, their pricing information isn't publicly available. But it is known that they will monetize the business intelligence service via subscription fee.

SiQueries uses Apache Cassandra to store its time series data. Why SiQueris chose Apache Cassandra? YourStoryAfrica asked Ebot. He replied: “Apache Cassandra is our Swiss knife for data storage. We started with the solution MongoDB (which still use), but we got to a point where we needed to write faster and scale. Apache Cassandra gave us some options not available with MongoDB. We believe MongoDB will definitely reach that level very soon. With Cassandra, we were able to combine tools such as Storm, Druidio and Kafka to build a better, scalable and high performance infrastructure which writes between 9,000 and 10,000 per second depending on the load.”

On the platform, data collection is quick and painless. Huge amount of the data is pushed into SiQueries infrastructure via REST API, which are being processed is represented internally in time series analysis. Vast amount of data is ingested and consumed in real-time via a few Storm topologies that are inherently parallel and run across a Kafka cluster of machines to generate alerts or to be displayed on interactive graphs. By the looks of their team and dedication, so far we are convinced they will be the first African startup to achieve 1,000,000 write byte messages per second like Netflix.

Business expansion plan

SiQueries is anticipating a rapid expansion across the continent in correlation with the recent rapid growth in mobile data in the region as there is a need for better and affordable business intelligence solutions.

YourStoryAfrica asked Ebot what the future holds in terms of making data work for Africa in a mobile era? “What we are certain of is that the amount of data generated from mobile will definitely grow faster than what it is now. It will be more unstructured and definitely the challenge of managing, carrying out analysis will grow and demands for better and deeper analytics will rise,” he said.

The competition is huge. To name some of the companies we get as feedback on what clients are use for their BI Analyitcs: Oracle, SAP, SAS, GoodData. The good thing happening now is that the extremely large amount of unstructured data is being generated daily from different data sources. How to mine and drive business values is definitely challenging as you have to keep in mind the nature and data formats, which are completely different from what we used to know a few years ago. We believe this will increase in the subsequent years as we are seeing a steady growth in data around Africa.

As long as SiQueries continue to provide an easy yet powerful service to connect to and analyze data for SMBs and SMEs, the odds will be in their favor to join the league of a few African tech companies. Like Mxit and Ushahidi whose services are being used in the world outside Africa.