Matrix connects social media with business bottom line using SAS® Sentiment Analysis

How do online communities and social networks affect business? What is the best way to interact with clients online? Why is it crucial that businesses "get" sentiment analysis?

Alessandro Petrella, Sales Director at Matrix – the Telecom Italy-owned provider of Web and mobile business services – discusses the new business standards and models necessary for our digital world.

First, tell us about Matrix.
Matrix supports companies in defining digital strategies for online business and for the development of Web communications solutions through effective CMS (content management system) technologies, WebTV and software as a service.

Our offerings range from helping businesses define digital and brand strategies to developing communication solutions centered on more innovative technologies.

Our portal, Virgilio, with its almost 13 million monthly visitors, possesses a 50 percent share of Italian Internet users, and places all the typical interaction tools for Web 2.0 alongside information on channels by subject.

Niumidia Adv and iopubblicità obtain national and local advertising, respectively. Matrix Code, dedicated to the B2B market and Web technology, is a group of 200 people that coordinates and manages the technological components of the group, from the Internet world to digital advertising platforms. Matrix Custom is dedicated to tailor-made integrated communication solutions for our large clients.

Online communities, and social media in general, play an increasingly important role in shaping opinions and influencing decisions. Do you consider it important for a company to monitor online sentiment?
Social networking has revolutionized the way in which companies interact with consumers. The diffusion of social media is profoundly changing the way in which consumers form their own opinions about a brand and its products. As a result, sentiment analysis is a better way to discover, without intermediation, what the emerging trends are. And sentiment analysis is a better way to identify critical issues and to evaluate brand reputation.

What are some ways a business can benefit from sentiment analysis?
By reading blogs and forums, you may discover innovative uses for existing products. That happened with one of our multimedia products that was originally directed at a general target, but which became successful in the professional world.

Feedback from the Web gives us a foundation for taking targeted measures. The ability to see and analyze particular indicators allows us to pinpoint strengths and weaknesses and then make adjustments to keep us aligned with your customers' needs. And you can adjust future projects based on those needs.

Why did Matrix launch a sentiment analysis project?
Matrix's positioning no longer consists of merely supplying banner ads on the Web. We now offer all-around communication products for our clients, thereby capitalizing on the capacity of Virgilio to attract an audience and generate traffic.

Not only are we using SAS Sentiment Analysis to understand the experience of our customers, but we are even using it to help define new products that we offer. We will be the first users of the solution in that we will analyze the sentiment expressed by users in relation to Virgilio services – with particular regard to email, which is the principal driver for our community. One of the products we offer, called Virgilio Caring, is based on SAS Sentiment Analysis. This is a tool that our clients use to evaluate opinions circulating online about their products, create targeted communications transmitted by way of Virgilio, and interact with reference contacts within social networks.

Why did you choose SAS?
After a careful examination of other products, we chose SAS for its flexibility as well as its completeness in terms of functionality, precision of the algorithms, abundance of reporting and ease of implementation and use.

SAS has proven to be flexible in the design and service implementation phase and has given us the ability to experiment pragmatically on how to obtain useful information from the great variety of digital content present on the Internet.

It's bound to be a challenge to pull just the right bits of information you need out of that vast ocean of data we call the Internet. How did SAS make that job easier for you?
From the comparisons performed, we saw firsthand how reliable and concise SAS is. In particular, we increased by an order of magnitude the quantity of information retrieved from the Internet that has value in terms of sentiment expressed. This result is no doubt due to the efficiency of crawling, but it is especially due to the sophistication of the SAS algorithms, which are able to understand whether a certain, nonstructured document expresses a positive or negative sentiment. This is an extremely complicated characteristic, because it is difficult to clarify syntactic expressions in common use, which more than anything vary depending on the context.

How do the reports produced from your analyses support decisions?
The added value of the solution is the quantity of analysis and reporting it brings to the decision-making process. It offers summarized data that is easy to consult and truly meaningful.

In our case, the results of analyses are certainly of interest not just for people in marketing and communication, but also for our developers. Understanding how users perceive our applications allows us to take action in terms of product design and functionality.

Without a doubt, there is a learning phase in order to take advantage of all the potential of the solution, which is rich in functions and analytical and statistical models. But I have to say that, with regards to the past, SAS tools have greatly improved in terms of ease of use, as demonstrated by the graphical interfaces and wizards.

'Sentiment' seems to be a phenomenon that is difficult to evaluate and measure. Do we need new metrics?
The world of television has the gross rating point, a standard that measures the percentage of a target audience reached multiplied by frequency of showings. We in the world of the Internet need to apply that same standard to online campaigns. Instead of shooting in the dark, we need at least a preliminary definition of the targets we're aiming for.

Will SAS® Sentiment Analysis help you define these targets?
We began from a precise requirement, which was that of brand reputation and sentiment analysis. During the development process, we realized that SAS could even provide us with support in other areas, automatically. For example, the analytical tools of Virgilio Caring were useful in clustering clients by social-demographic segmentation and themed profiles. And this helped us to target advertising and to differentiate the published products, thus supplying our clients with information of genuine interest to them in the format they prefer.

This story is a translation of a story that originally appeared on ita.sas.com.

The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Copyright © SAS Institute Inc. All Rights Reserved.