Our approach

Our Approach

The solution required us to leverage both our strategy expertise and analytical modelling capabilities. Our specific activities included:  

1. Demographic Assessment: We leverage publicly available data sets to build a look-alike model, comparing demographics in neighbourhoods to existing, high-value customer segments.
2. Scoring Neighbourhoods: Using demographic and market data, we scored neighbourhoods across various cities to identify which zip code would provide the optimal location to open in.
3. Expansion Blueprint: A tailored rollout plan was proposed, specifying the number of sites per city and a phased timeline for their opening. The plan included a capacity analysis to ensure sustainable growth while leaving room for future expansions.

This approach resulted in an analysis tool displaying heatmaps of key statistics within US cities which highlighted optimal locations to open up in. This tool also identified potential cannibalisation risks as the expansion progressed.

"An excellent session yesterday and a really useful analysis report. It will help the business greatly."

Portfolio Company CEO

For more information on how our consulting expertise can help your business navigate strategic challenges, contact us at contact@datavisionservices.co.uk to see how we can best work together.

The post Building a strategic expansion plan for a leisure chain looking to enter new markets appeared first on DVS - Data Vision Services.

Natural Language Interface to Databases (NLIDB) or Natural Language Querying (NLQ) system. These systems use Natural Language Processing (NLP) techniques to understand and interpret the user’s queries and then translate them into SQL or other query languages that the database can understand. Then this data can be summarized in graphs, tables, and written reports by interacting with the system in natural language.

There are 3 ways this will affect PE Businesses:

1. Centralized Analytics for Owned Businesses:

NLIDB systems can serve as a powerful tool for centralizing and streamlining analytics across a portfolio of owned businesses. By connecting to multiple databases and data sources, these systems enable PE professionals to access critical financial and operational data from various portfolio companies in a unified and coherent manner.

This centralized analytics approach offers several benefits:

• Consolidated Insights: PE firms can effortlessly gather data from different businesses and create consolidated reports and dashboards. This provides a comprehensive view of the portfolio’s overall performance and financial health.
• Benchmarking: NLIDB systems allow for easy benchmarking of key performance indicators (KPIs) across portfolio companies. This aids in identifying underperforming or overperforming businesses and implementing strategies accordingly.
• Risk Assessment: PE professionals can quickly assess the financial risks associated with each portfolio company by comparing financial data and market trends. Early detection of potential issues can lead to proactive risk mitigation.
• Efficiency: Centralized analytics reduce the time and effort required to compile and analyse data from multiple sources, enabling faster decision-making and more efficient allocation of resources.

2. Democratization of Analytics Across the Business:

NLIDB systems empower not only data analysts but also non-technical professionals within the PE firm to access and interpret data. This democratization of analytics has several advantages:

• Self-Service Analytics: PE professionals, including investment managers, can formulate complex data queries in plain language without relying on data experts or IT support. This promotes self-service analytics.
• Data-Driven Decision-Making: With easy access to data and insights, decision-makers throughout the organization can make more informed and data-driven investment decisions.
• Reduced Bottlenecks: By reducing the dependence on a limited number of data experts, NLIDB systems eliminate bottlenecks in data access and analysis, leading to faster response times.
• Cross-Functional Collaboration: Democratized analytics encourages collaboration across different departments, as professionals from legal, finance, and operations can all engage with data to contribute to investment strategies.

3. Portfolio Monitoring:

NLIDB systems significantly enhance portfolio monitoring capabilities for PE firms:

• Real-Time Insights: These systems can provide real-time updates on portfolio company performance, financial metrics, and key events. PE professionals can quickly react to changes and make timely adjustments to their investment strategies.
• Alerts and Notifications: NLIDB systems can be configured to send alerts and notifications based on predefined triggers. For example, deviations from financial targets or significant market developments can trigger automatic alerts, ensuring timely attention to critical issues.
• Predictive Analytics: By leveraging historical data and predictive analytics models, NLIDB systems can help forecast future performance and potential challenges within the portfolio. This foresight enables proactive decision-making.
• Customized Reporting: PE firms can create customized reports and dashboards for each portfolio company, tailoring analytics to specific business needs and objectives.

In summary, NLIDB systems offer private equity businesses centralized analytics capabilities, democratized access to data and insights, and advanced portfolio monitoring features. These benefits collectively contribute to more effective decision-making, risk management, and overall portfolio performance.

When are NLIDB systems joining the workplace and what are the challenges?

There are challenges in the implementation of AI systems in the workplace. Data focused AI systems are going to be mostly constrained by long standing problems in data quality, modelling, resource scalability and cataloguing that plague businesses.

At DVS, our capabilities lie both in the identification of business applications or NLIDBs and the development stages of the infrastructure pieces needed to deliver them. Natural Language Annotations, Metadata and Semantic Enrichment will be key to the integration of NLIDB systems to existing databases, enabling AI to find and understand how the underlying data can be used to answer our questions.

References:

• Alshammari, A., & Alshammari, M. (2021). A Review of NLIDB with Deep Learning: Findings, Challenges and Open Issues. In S. K. Singh & S. K. Singh (Eds.), Advances in Data Science and Management (pp. 217–228). Springer Singapore¹
• Alshammari, M., & Alshammari, A. (2021). A Survey of Natural Language Interfaces to Databases: Challenges and Opportunities. In 2021 IEEE International Conference on Big Data and Smart Computing (BigComp) (pp. 1–8). IEEE²
• Alshammari, M., & Alshammari, A. (2021). Natural Language Interfaces to Databases: A Review of the State-of-the-Art Techniques and Open Challenges. IEEE Access, 9, 101156–101178³

The post Optimizing Portfolio Performance: The Power of NLIDB Systems in Private Equity appeared first on DVS - Data Vision Services.

To automate easy decisions so SMEs can focus on the tough calls

The ask

DVS were asked to produce a data-assisted decision support application, which would calculate KPIs with RAG scores to present to decision-makers. This would help improve decision making on which supplier to use. Rather than simply visualising these on a new dashboard, our outputs would be integrated with existing tools so that the insight was in front of users at the point that decisions are made.

What does good look like?

As we started the process, we quickly discovered that KPIs were not consistently defined across the business.  This was a concern as this product would have a significant impact on decisions at the core of the business.

To overcome this, we set up meetings to understand departmental needs. Then, we ran these through workshops with managers and executives to align and define a set of KPIs. We scrutinised these definitions so the business could agree on how edge cases would work, asking probing questions such as:

• How should we present the data where you have newer suppliers with less history?
• How should we weight the score for different metrics?
• What is the balance between quality and cost? How much more are you willing to pay for better ‘quality’ suppliers?

Alongside this, we understood that the way the data and returned RAG scores were interpreted would be crucial to the success of the project. As well as this, it was important to register that different metrics should be weighted differently to each other.

Technical pragmatism

At DVS, one of our key priorities is always to find the right balance between cost and performance. Our app has the proven ability to respond to requests in milliseconds, but we had to ask the question: “Is it good value to have additional compute on standby, to respond to concurrent requests instantly?”

In the end, we aligned technical and business sides of the organisation to agree on a balance of cost and performance, preferring consecutive requests over concurrency, which could reduce compute cost by as much as 90%.

The build

We wanted to build an application which could be owned, maintained and updated by the client, so naturally we developed and deployed into their existing Azure environment. Three Azure Functions were created, which could be run and monitored independently. The core Function was our ‘RAG Scoring’ application, which could be triggered by an HTTP request, and returned a JSON structure containing the KPIs and RAG scores.

Security & best practice

The applications were able to be secured behind Azure AD (which required any user to be registered with the client on Azure, as well as using multi-factor authentication).

Beyond that, we deployed our standard good practice in technical development, including storing secrets in Azure Key Vault and continuous integration through GitHub Actions.

Given the impact that our application could have on business decisions, it was critical that a record of KPIs and RAG scores were maintained. This was accomplished through triggering the API with a scheduled Function, which would store the output on a data warehouse.

The outcome

The new process has delivered better, faster decisions for the client. The capacity of the team has increased and allows the business to grow. The quality of the selections has increased too – next is to track the impact on their client NPS!

The future

This is the first step on a machine learning roadmap. We’ve built a behind-the-scenes ‘recommendation tool’ and are assessing where the model and the human process agree and disagree. The plan is to build trust in machine learning and automate as many easy decisions as possible.

The post Finding the Sweet Spot: Automated Decision Models and Human-led Decisions appeared first on DVS - Data Vision Services.

In recent months, generative AI (GenAI) has garnered significant attention in the media, generating seemingly human-like content within seconds. GenAI models such as ChatGPT (GPT3 and GPT4) or DALL-E have the potential of bringing unprecedented automation and efficiency to a wide array of sectors. It seems as though businesses must race to unlock the power of generative AI and its nuances in order to stay ahead in this fast-paced era. In this blog post, we offer a pragmatic view on these developments, focusing on three key guiding principles: the importance of understanding business needs when organisations start using AI; the importance of starting small and building iteratively, and the role that “traditional” ML models will continue to play.

Our AI & ML Adoption Guiding Principles:

• The first guiding principle emphasises the importance of understanding business needs when organisations start using AI. As generative AI models like ChatGPT and DALL-E continue to monopolise headlines, it is vital for businesses to identify their specific challenges and objectives to ensure successful AI adoption. The number of potential use cases that can be addressed through AI is seemingly endless. As such, conducting a through needs assessments that reflects the needs of stakeholders and defining “what success looks like” will be key to success. We recommend coming together in functional groups within your business to identify the potential use cases, then focusing on a small subset, with a clear view on the rationale and benefits of embarking on each.
• Secondly, we believe starting small and building iteratively when adopting AI solutions is again a key success factor. Embarking on multiple massive AI projects without a clear roadmap or sufficient experience can be an easy route to wasted resources. Instead, organisations should focus on small-scale proof of concept projects. These allow their AI applications to be tested, validated, and refined. Businesses can therefore learn from their successes and setbacks, scale gradually, and gain business stakeholder buy-in along the way. This iterative and incremental process should be accompanied by a clear understanding of the change management needed for success. One of our recent projects involved building a decision-support system for a technology-enabled logistics provider. While on the project, we spent most of our time working with our client to define business needs and what success looked like, and critically analysing the output of each iteration. Only a small proportion of time actually building the system. We are confident that this, rather than our technical skills, was the key to the success of the project. After all, an AI project shares many elements with traditional technology implementation. In these projects, people and process, rather than the technology itself, are the ultimate driver of success.
• Finally, we would like to stress the relevance of “traditional” machine learning models. While generative AI has been making headlines, it is crucial to recognise that traditional machine learning approaches will continue to play a vital role in various applications. Predictive analytics, pattern recognition, and classification tasks can still be effectively handled by traditional ML models and applied to a variety of business contexts. Oftentimes, these approaches also have a higher degree of “traceability”. This allows the steps involved in the decision-making process to be documented and critically analysed, which we believe is a key benefit over the traceability of today’s GenAI models. There is still plenty of headroom for traditional machine learning approaches and they should not be forgotten in the AI gold rush.

The post Our three guiding principles for successful AI and ML adoption appeared first on DVS - Data Vision Services.