Self-serve analytics? Can you please define? blog

I have been living in the UK for now more than 15 years and, when I first arrived, I was struggling to understand or be understood (some say that is still the case). I owe it to many people that I now control (vaguely) this beautiful language though I still recollect when someone told me: "So, you are actually French". In French, the word "actuellement" stands for "currently". A so called faux-ami. So my answer was: "Well, I've been French for a long time".

Bemused look ensued...

Now that I work in the advanced analytics space, I am starting to hear a word, well an aggregation of words, which is also open to misinterpretation. The expression "self-serve analytics". Many organisations are using this definition in marketing literature and it seems to be gaining a lot of popularity. But as we (PredicSis) seem to define self-service in a different way than many other definitions I came across and, as I've learned my lesson and want to avoid sounding foolish again, let me share some perspective.


I am seeing a few great companies for which "self-serve analytics" seems to serve one part of the market which I would qualify as the "advanced" part. The professional who do know their algorithm language. People proficient with statistical languages, R and other complex Python "things" (note: I am not falling in this advanced segment so "apologies for the technicality of "thing"). There are data science studios or pre-packaged R algorithms and many other very useful solutions that suit a growing market of data scientists.


So what does PredicSis call self-serve analytics?

Good question dear reader. Well, what we call "self-serve advanced analytics" is to give analytics capability to these advanced users (more on this below) but also, importantly, to people who know nothing about data science jargon, to whom the letter R is simply between the letter Q and the letter S or those who are starting to lose track when the algorithm word crops up in a conversation. The extent of the technical knowledge of these people is loading up a csv file into a browser interface and running an SQL query before hand. Mention the words overfitting, grid search, fine tuning parameters and they get lost. Mention even the word "feature" or "aggregate" and they are walking away. BUT, and it is a big but, these people know their data and their business very well. These are the BI analysts dealing with data daily. Or even the CMO who know their strut.

What PredicSis enables them to do is not to get distracted by jargon but simply inject non perfect data into Predicsis and they can get, in minutes, not in weeks or in months what the churn risk at the population level or at the individual level. And the underlying reason that drives the risk of churning. So they can get clear insights in minutes: "oh, so my customers who have a big car AND have clicked on the shampoo page are the ones with the highest risk to churn? I did not realise that, let's get some activities in place right away". Ok, slightly over-streched example but we hear of CMOs who use PredicSis on a daily basis and telling us they had preconceived ideas, insert data in PredicSis and realise their views of the market was flawed. That is the true power of AI in corporate environment (I had to put AI somewhere...). A picture or even a video is worth a hundred words so this can be seen in a 5 minutes short video:


We believe that both "self serve analytics" approaches are needed, both those who know about advanced data science and those who don't. Needless to say that, as Predicsis falls in the so-called auto Machine Learning category, we also have very advanced data science teams using it. Very large teams of data scientists use PredicSis as they want to do far more in far less time, they want to increase their interaction with the business as a whole. What they particularly use is what is known as the automated feature engineering and selection (sorry...). No need to get into the technicalities but in short, this helps them find combinations of variables which us, mere humans full of biases would not naturally do without a sheer amount of work and data combinations that would take a large amount of time and yes, money (the shampoo and big car example above would be one of these features).


This is what the Predicsis team calls self-serve advanced analytics. A mechanism to enable the drive towards a true data centric organisation, where not only data science teams but also CMO, BI analysts, product managers, marketing manager, HR managers can use advance analytics on a daily basis with the ability to get data centric answers in minutes, not weeks or months.


After 15 years in the UK, I've learned the hard way that words can have different meanings. Hope that this little post will prevent some bemused looks when we talk about self-serve analytics. And if you don't believe me (fair enough), if you are really serious about predicting your customer behaviours, in minutes, and have some projects lying around that you haven't started because "it is too complicated (actually it's not), the data isn't perfect (actually it doesn't need to, the machine deals with it) and it is too expensive (actually it's not, if you are on AWS it starts at $3ph)", contact me. What works for Top tier US payment companies, EDF, Orange and others might not work out for you but you'll be able to test what we define as self-serve advanced analytics.




The PredicSis Team Ready for AWS Summit Berlin

PredicSis is a proud sponsor of AWS Summit Berlin.

We are delighted to be participating in another AWS Summit. This time the PredicSis team will be in Berlin.
AWS in Las Vegas was an amazing event and we had a lot of great feedback.

The Berlin Summit on May 18th brings us just as much excitement as we are now on the AWS Marketplace enabling AWS users to start predictive BI in 2 minutes without any hurdles. No need of complex data science knowledge and PredicSis handles imperfect, unflatten data.

We will show you how to quickly start your predictive analytics project and get results in minutes and inject it in prod for real world feedback.

If you wonder how to start predictive BI to complement your existing BI stack, come and meet us, we will be located in the Marketplace sponsor zone. We work with large and small organisations, from AMEX all the way to the small start-up.

Want to schedule a time to meet with the team? Simply fill out the form on the contact page, and we’ll do the rest.

Can’t make it to Berlin? Don’t worry; you can follow live updates of everything that is going on by following on Twitter – @predicsis

New office in San Francisco

PredicSis in San Francisco

To support its growth, PredicSis opens up its first office in the United States. After great success in France and Europe, PredicSis is now accelerating its sales & marketing in the US. meets the increasing demand on self-serve analytics software. It blends smart data discovery and predictive analytics in a unique insight discovery workflow geared towards business actions.

'Founding our office in the Bay Area is the next logical step to actively get involved in the ecosystem of partners and customers.  The AI awareness of the Bay Area is another reason which lead us to chose San Fransisco’ Jean Louis Fuccellaro, co-founder and CEO of PredicSis explains.


      Artificial Intelligence, Machine learning,Neural Networks…Let’s try to be simple!       






       Steven Spielberg is a visionary! He released, A.I. 15 years ago and unfortunately, he did not raise much attention at that time. Today it would have been probably very different as it’s hard to escape not only AI but some of the buzz words coming along–Machine Learning, Deep Learning.... AI, Artificial Intelligence, is even becoming a public debate mostly around the potential threat it could bring to humanity. On this topic, please read the excellent post   Will Machines eliminate us?         In the high tech world, AI has recently gained lots of traction as it seems to be a crucial way to deliver business value out of the big data as well described in   Is Big Data Still a Thing?   But recently, a day after Microsoft introduced an innocent AI chat robot to Twitter it had to be deleted after it transformed itself into  a very evil one . It shows that AI needs to be well trained otherwise it can easily go out of track. When doing business, this is the last thing you want to happen: it’s better to know where you’re going.  Our customers or prospects often have questions such as "what’s the difference between Machine Learning and Deep Learning?”, so I thought it would be interesting to put down some answers to these questions. It is definitely not academics answers but hopefully it could be helpful for true beginners in the field.      Buzz word #1: AI  AI (Artificial Intelligence) is an automated decision making system, which can in an autonomous way continuously learn, adapt, suggest and even take actions. Google car is a good example of AI.  Boston Dynamics  made real cool stuff too. Robots and AI systems cannot rely on computer programs explicitly integrating every possible cases and appropriated actions. At the very core, they require algorithms able to learn from experience. This is where Machine Learning comes in.     






        Buzz word #2: Machine Learning  Machine Learning, or ML for short, is the science of designing and applying algorithms able to learn from past cases (kind of a child within a computer). It is worth saying that you cannot predict something that has never happened though: no past cases, no prediction. If some behaviour exists in the past, you may predict if or when it will happen again. Sounds weird? Let’s take an example: You want to know which customers of yours are going to buy product A in the next seven days. You start from extracting past cases of customers who bought product A and who didn’t. Then comes the “learning” phase. It’s the phase when the machine creates profiles of customers who are likely to buy. Now, by looking at the current Customer data, the machine can compute the likelihood of an individual customer to buy product A within the next 7 days.  Questions you might want to ask can be very diverse. You could detect which sensor is going to breakdown, which transaction is going to fail… There is virtually no limit - stating the event occurred in the past and there is a link between past data and the question (unfortunately there seems to be no chance of winning the national lottery with Machine Learning; if the game is not rigged).     






     When training an ML algorithm, one has to deal with complex tasks such as optimizing learning speed, number of samples analysed at each training step, evaluating multiples sets of parameters for each algorithms… And some ML algorithms are more complex than some others, so, please, beware!     Buzz word #3: Neural Networks  Neural Networks are a type of Machine Learning algorithms. Like a biological brain, neural networks consist of interconnected units called neurons; they mimic the way the brain can learn, changing the local patterns of neural connections. Artificial Neural networks are often organized in layers, exchanging, transforming and transmitting information (data).  The training phase is usually time-consuming due to the complex structure and the large number of hyper-parameters. That’s why it is worth training such networks mainly in cases like voice recognition, pattern recognition within pictures, facial recognition, etc. where one does not have to re-train models quite often.     Buzz word #4: Deep learning  Let’s go deeper in the family... Deep learning is part of the Neural Networks field, which is itself part of the ML. It is the subfield dedicated to the study of stacks of Neural Networks and often involves many (many, many) layers. Increasing the number of layers and combining different networks allows to dramatically increase the predictive performance of the overall architecture. The bad news is the resulting skyrocketing time it then takes to learn such complex structures. This is a show stopper for many marketing applications as they require fast learning and adaptation cycles: one cannot afford to spend the tremendous amount of time (and money) for learning complex, and performing enough, deep learning nets.  For other applications, like the ones we mentioned before, it is worth investing in the computing infrastructure and human experts required. And, maybe, deep learning will be the key to understand the way we  dream .  Artificial Intelligence is a large and complex field, and getting a basic understanding might seem an overwhelming task. To summarize: it makes use of a family of techniques (ML) and sub-techniques (Neural Networks) giving powerful tools (Deep Learning) to extract value from Big Data, and as such brings deep changes for businesses using it.  What’s interesting nowadays is that AI is more and more used in the industry as every businesses, even the smaller ones, can now generate zillions of data record.     Now, you might be in such a position you’re sitting on top of a mount of data. And you want to ask your Big Data “what benefits could they bring to my business?” or “which techniques should I use?” That will be the next topic: insights to understand which problems benefit from Machine Learning!    

Steven Spielberg is a visionary! He released, A.I. 15 years ago and unfortunately, he did not raise much attention at that time. Today it would have been probably very different as it’s hard to escape not only AI but some of the buzz words coming along–Machine Learning, Deep Learning.... AI, Artificial Intelligence, is even becoming a public debate mostly around the potential threat it could bring to humanity.
Today, it’s hard to escape not only AI but some of the buzz words –Machine Learning, Deep Learning…- which come with it. AI is even becoming a public debate mostly around the potential threat it brings to humanity.

HDDs fail sometime (know which ones beforehand)

Did you notice how electronic equipment always happen to fail at the worst time? Well, does a ‘good’ moment for a failure exist anyway? This is why you, the latest data scientist unicorn, is often asked to find a solution to the following business problem: reduce failure cost related to the management of a pool of electronic devices by extracting early reliable failure signals from raw sensor data.

How to save precious time in data preparation

As a data scientist, up to 80% of your working time is dedicated to the data preparation and feature engineering tasks. Chatting with the data owners, building predictive models, delivering proved RoI to your stakeholders, eating, sleeping, and other basic functions fill in the remaining 20%. And it’s just plain sad.

So! Like myself, I know you wished you sometimes had a magic wand, transforming your raw data into informative and reliable features. 

Well guess what. This wand is already here

How to Take Advantage of Machine Learning and Kissmetrics to Reduce Churn

Oh you, young and motivated SaaS company. I know you very well. You have a disruptive way of seeing things, a cool product, reasonable prices (well…), and just want to make the world a better place. But you won’t be able to do that if you don’t try to understand the behavior of your hard-earned customers.

You know (well informed that you are) that the “Leaky Bucket theory” is a real thing. Acquiring a new customer will cost you more than trying to keep the ones you already have. And that’s where KISSmetrics in-app analytics and machine learning can help you make better-informed decisions for your customer retention campaigns.

How an API can help improve marketing campaigns - An interview of Adrian Lopez, from Path travel

On May, 21st 2015, PredicSis and ChurnSpotter participated in the PAPIs Connect conference, an international event on machine learning and APIs, intended for decision makers, and held in Paris. 

As a way to promote our API and to allow people to discover PredicSis' products and to promote PAPIs Connect, we thought of a fun way do introduce attendees to the world of predictive analytics. We launched a contest.

The response was truly great and we gained invaluable remarks and feedbacks on the API. Therefore we thought that it would be interesting to make a little interview with our winner.