Analytics

Enterprise Analytical Management is the "New Black"

“Analytics and big data are creating profound new opportunities, but not many businesses out there are able to combine the right resources to take advantage of them. Altair’s strategy experience and analytical expertise help equip our clients with knowledge and tools to unearth meaningful insights from data, enable better decisions and obtain competitive advantages”

Challenges

How to navigate the Ocean of Data?

“Data transforms how companies organize themselves, decide which technologies to use, and build ecosystems of partners and vendors”

Today, technologies are generating unprecedented amount and variety of data. As this growth continues, so do the potential to transform the landscape of business world and to capture tremendous value.

 

It transforms how companies organize themselves, decide which technologies to use, and build ecosystems of partners and vendors. The time has come to value data and analytics as a strategic asset that can help you win.

Big data isn’t a technology initiative to be left entirely to data scientists, mathematicians and IT departments, but a strategic business opportunity that requires technical savvy and organizational coordination.

 

To gain the upper hand, companies need to embed analytics deep into their value chains to ensure that information and insights are shared across business units and functions.

Insights

“Companies need to embed analytics deep into their value chains”

Our Approach

Embedding analytics in the value chain of an organization is no easy task. It requires a well-structured and carefully orchestrated approach that combines analytic excellence, technology and business expertise.

 

Here at Altair, we help companies develop models and methodologies to transform data sets into strategies and knowledge to align with their capabilities, teams and operations, with the objective of gaining a competitive edge.

Analytics Cases

A selection of our Analytics related cases

US Insurer Independent Agents Segmentation to Increase Market Share

The insurance market leader for auto and home had a declining market share over the last 5 years due to raising competition from online players. To gain ground in the market, a segmentation of the independent agents was done through supervised machine learning algorithms and the adaptation of a pillar from information theory, the Shannon-Weaver index. This index was adapted to measure the commercial presence and intensity of independent agents for their area of influence. Homogeneous groups of agents were defined, and tailor-made commercial strategies and value proposition were design for each segment. The segmentation allowed the insurance company to better allocate their resources and services to independent agents that brought the highest value in terms of writing premiums and market share.

US Corporate & Investment Bank Asset-Backed Securities Rating Model Development

An international bank acquired a tool from an external vendor to assess the risk from asset-backed securities (ABS) in its portfolio. The performance of this third-party technology did not meet the expected results because of its tendency to overestimate cashflow allocation. Our client decided to develop an internal tool and asked Altair to lead the development of the model, based on Monte Carlo simulations and implemented on VBA. The tool was successfully integrated within daily operations and helps to predict monthly asset values and the allocation of cashflow of asset-backed securities. After the validation of the tool, its capabilities were further improved, allowing for the inclusion of revolving asset classes in the analysis, thus improving the reach and prediction acumen of our client.

AI – Deep Learning Model for Smart Agriculture

Our client, an international agriculture firm with fields in Europe and Africa needed to ameliorate their production forecasting model to better adapt to market conditions. Altair was challenged to improve current forecasting model. Our approach was to train a computer vision neural network to identify and compute the number of fruits per tree and measure its size and maturity to forecast production at early stages. The model was also instructed to identify and suggest corrective measurements for general problems such as over and under watering and most common plagues and diseases. The result was an increase of 15% in forecasting accuracy and a better wholesale pricing strategy that allows to anticipate fruit harvesting and leverage arbitrage opportunities in the commodities spot/future prices.

Global Oil&Gas Company Key Accounts Credit Risk Model Definition

Major key account clients of an international energy company were demanding extended payment terms in their purchases of oil products. The company wanted to capture these opportunities, but its credit policy was conservative, and credit demands when through a detailed approval process. Altair developed a model to assess customer risk to margin mix levels to compensate for the increase in financing cost and client probability of default risk. The model allowed a faster evaluation of the commercial proposed conditions and suggested a threshold for each segment and country, and a minimum margin to cover a specific return on working capital employed in the operation. As a result, business volume in existing and new customer increased, risk exposure was optimized, and the credit approval cycle was reduced.

European Department Stores Pricing & Promotions Optimization Model

Our client, a leading department store that had a fashion apparel seasonal sales and promotional strategy instead of following “everyday low prices”, was concerned about low return on promotional investment during sales season and surplus stock. Altair reviewed their Mark Down Optimization (MDO) plan and devised a new strategy using Markowitz Portfolio Theory building a dynamic and standardized pricing model for sales season. Altair team oversaw the design of this new model, the development of new KPIs, and the implementation of the transformation plan; ensuring that price recommendations were followed during sales. The result was an increase of 8% in margin —exceeding the initial expectations of 3%— and a faster acquisition of recommendations by the team of planners and merchandisers.

Canadian Bank IFRS9 Model Risk Improvement

A Canadian bank had the highest level of provisions among its peer competitors because of the approaches used to estimate the different risk parameters. Altair designed a working plan based on the refining of existing model in order to minimize expected credit losses through the application of IFRS 9 accounting principles. The new model approach included the utilization of advanced time series forecasting based on macro data and default rates, training of the model with an amplified set of business and operating variables and use of combination of three different approaches: logit, AR and Supervised Machine Learning. Altair’s new model resulted in a reduction of 15% of provision levels and a better ratio between the receivable volumes and the provisions, reaching competitor’s ratio.

American Consumer Finance Network Footprint Optimization Using GIS Tools

The situation of an inherited branch network responded to a historical sketch and did not consider the demographic changes and cannibalization that have taken place in the last few years. The emergence of new competitors and products had affected the performance of its branches. Altair analyzed the location of the company’s and competitors’ branches, the addresses of its customers and the neighborhoods’ socioeconomic mapping. This analysis defines the areas of influence of each branch and look for potential gaps not covered by competitors. A geostatistical and Machine Learning methodology was developed to evaluate and categorize the situation of each branch, proposing which ones should remain open, close or relocated in growing areas. This model has optimized the office network performance by 35%.

Leading Retailer Weekly Sales, Stock & Intake (WSSI) Model Development

A major retailer with more than 500,000 SKUs wanted to fully understand its stock management with the aim of optimizing its product portfolio and improving all the warehouse logistics management. Altair carried out a deep client data analysis in order to understand its consumer behavior and developed an advanced weekly forecast of its demand based on its historical data and other macroeconomic variables. Altair was able to implement a warehouse restructuring plan based on the expected demand of purchases, along with Altair’s expertise of lean logistics methodologies. The implementation of a Weekly Sales, Stock & Intake Model (WSSI) and the efficient purchasing from suppliers helped our client to reduce the associated costs by 7% with a better stock management and the reduction of the current warehouse’s centers.

UK Bank ATM Network Customer Loyalty Case Model

Our client, a UK branch of a leading global banking group, was seeking to understand the ATM disloyalty patterns to reduce the high annual fees paid to third-party institutions and to improve customer services. Altair gathered insights through a customized dataset and implemented data-driven analysis, performing a segmentation based on average balance account, ATM usage, online banking activities, ATM locations, and operations at third party ATMs. The engineering process pointed out key findings to understanding ATM behavioral patterns of customers and developed a robust prediction model with learning capability. We also recommended actionable measures to optimize ATM footprint and minimize costs adapted for each customer group. Our client had a precise and robust prediction model which results in significant savings in Opex.

US Retail Bank Price-Balance Optimization Model for Regular Liability Products

The US branch of an international bank had an aggressive and non-targeted commercial strategy for its deposits. Interests were substantially above market average, but retention levels were below management expectations. Altair approach was to divide the project in three phases: (1) Segmentation of branches by its competitive pressure. This was achieved with the help of hierarchical clustering model. (2) Train an elasticity model based on historical data to locally maximize margin for each specific cluster and dynamically adapt pricing to new market conditions. (3) Multivariate regression model to identify churn drivers and address improvement areas. Altair achieved top-notch results that were: increment of balance of 2.2%, increased margin by 4.9% and reduction of 15.2% of churn rate within the first year.

We use cookies to give you the best online experience. By agreeing you accept the use of cookies in accordance with our cookie policy.

Privacy Settings saved!
Privacy Settings

When you visit any web site, it may store or retrieve information on your browser, mostly in the form of cookies. Control your personal Cookie Services here.

These cookies are necessary for the website to function and cannot be switched off in our systems.

In order to use this website we use the following technically required cookies.
  • wordpress_gdpr_allowed_services
  • wordpress_gdpr_cookies_allowed
  • wordpress_gdpr_cookies_declined
  • pll_language

Decline all Services
Accept all Services