Site Search

Menu

Topics

Loading Results

Top-down, bottom-up or side-to-side? Sample strategies for upgrading your data platform

CEO Survey hero
  • Three sample strategies and guidance on how to upgrade your data platform
  • The type of Minimum Viable Product (MVP) you should build depends on what you need to achieve
  • The best strategies demonstrate how sustainable value delivery will be achieved

Upgrading a data platform for your transactional or analytical tasks can seem daunting, but it is often necessary. You may be looking for improved performance, cost savings, better integration, enhanced security, or incorporation of modern capabilities - for instance, streaming data processing, real time analytics or artificial intelligence. All have the potential to significantly benefit your organisation. 

Choosing the right strategy to implement it is essential. This article defines three sample strategies: top-down, bottom-up, and side-to-side, and offers guidance on how to use them to build a future-ready data platform.

Top-down approach: Building trust from the top

The top-down approach begins with a well-known component, like a dashboard or API, and builds all the underlying assets from there. This method is effective for quickly establishing trust and showing immediate value to stakeholders.

Advantages of the top-down approach

  1. Immediate visibility: By starting with a dashboard, stakeholders can quickly see the value of the data platform. This visibility helps secure buy-in and funding for further development.

  2. Clear objectives: The top-down approach often begins with well-defined business objectives, ensuring that the development aligns with strategic goals.

  3. Stakeholder engagement: Engaging stakeholders early in the process can lead to better alignment and support for the project.

Challenges of the top-down approach

  1. Complexity: Developing all the underlying assets to support a dashboard can be complex and time-consuming.

  2. Dependency management: Ensuring that all components are correctly integrated and functioning can be challenging.

  3. Risk of over-engineering: There is a risk of building more than what is necessary, especially when multiple independent projects commence at the same time, leading to increased costs and time.

Best practices for top-down implementation

  1. Start with a clear vision: Define the key data elements for the platform along with the associated metrics and KPIs.

  2. Iterative development: Use an iterative approach to develop and refine the dashboard and underlying assets, taking care to avoid overlapping domains.

  3. Engage stakeholders: Regularly involve stakeholders to gather feedback and ensure alignment with business goals.

Bottom-up approach: enabling exploratory use cases

The bottom-up approach focuses on working within a domain to identify adjacent data sets. This method is particularly useful for enabling more exploratory use cases using machine learning and other advanced analytics techniques.

Advantages of the bottom-up approach

  1. Flexibility: This approach allows for more flexibility in exploring and experimenting with different data sets.

  2. Innovation: By focusing on domain-specific data, teams can uncover new insights and opportunities for innovation.

  3. Scalability: The bottom-up approach can be scaled incrementally, reducing the risk of large-scale failures.

Challenges of the bottom-up approach

  1. Lack of immediate visibility: Unlike the top-down approach, the bottom-up method may not provide immediate visible results to stakeholders.

  2. Potential for siloes: There is a risk of creating data silos if not managed properly.

  3. Alignment with business goals: Ensuring that the implementation aligns with broader business objectives and priorities can be challenging.

Best practices for bottom-up implementation

  1. Start small: Begin with a small, manageable domain and gradually expand.

  2. Encourage experimentation: Foster a culture of experimentation and innovation to show value being delivered through the platform.

  3. Foster collaboration: Develop mechanisms to share domain-specific insights to strengthen the broader data platform.

Side-to-side approach: Migrating and uplifting legacy systems

The side-to-side approach involves finding sets of adjacent dashboards and datasets, making it particularly useful for performing migration and uplift from legacy technology stacks.

Advantages of the side-to-side approach

  1. Efficiency: This approach can streamline the migration process by focusing on related datasets and dashboards.

  2. Reduced disruption: By migrating adjacent systems together, the impact on business operations can be minimised.

  3. Cost-effectiveness: This method can help identify and eliminate redundant systems, reducing overall costs.

Challenges of the side-to-side approach

  1. Complexity of migration: Migrating legacy systems can be complex and require careful planning.

  2. Data consistency: Ensuring data consistency and integrity during migration can be challenging.

  3. Stakeholder management: Managing expectations and securing buy-in from stakeholders can be difficult.

Best practices for side-to-side implementation

  1. Thorough planning: Develop a detailed migration plan that includes timelines, resources, and risk management strategies.

  2. Incremental migration: Migrate systems incrementally to reduce risk and ensure stability.

  3. Engage stakeholders: Communicate regularly with stakeholders to manage expectations and gather feedback.

Minimum viable product (MVP) for data platforms

The concept of an MVP for data platforms can be subjective, as different approaches demonstrate value in different ways. It may be:

  • Top-Down: A functional dashboard with key metrics.

  • Bottom-Up: A set of exploratory insights or a machine learning model.

  • Side-to-Side: A successfully migrated and integrated set of dashboards and datasets for a domain.

Key considerations for MVP development

  1. Define clear objectives: Establish what constitutes an MVP for your specific context.

  2. Focus on value: Ensure that the MVP delivers tangible value to stakeholders.

  3. Iterate and improve: Use feedback to refine and enhance the MVP.

Demonstrating sustainable value delivery

Regardless of the chosen approach, the best strategies ensure sustainable value delivery. 

Key elements of sustainable value delivery

  1. Continuous improvement: Implement, test and improve processes for continuous innovation.

  2. Stakeholder engagement: Maintain regular communication with stakeholders to ensure ongoing support and alignment.

  3. Performance metrics: Develop and track performance metrics to measure the impact and value of the data platform.

Best practices for sustainable value delivery

  1. Iterative development: Use an iterative approach to continuously refine and enhance the data platform.
  2. Feedback loops: Establish feedback loops to gather insights and make data-driven decisions.
  3. Scalability: Ensure that the data platform is scalable and adaptable to anticipated future needs and challenges.

In conclusion, choosing the best approach for your data platform depends on your specific needs and goals. A fit-for-purpose Data Strategy can help identify the drivers and inform your decision-making. 

Each of the sample strategies — top-down, bottom-up, or side-to-side — has its own benefits and challenges, and defining one to address your specific circumstances is essential. By understanding the data ecosystem, defining a clear MVP, and focusing on sustainable value, organisations can successfully develop their data platforms and achieve their strategic goals.

Contact the authors

Mo Bashir

Partner, Advisory, PwC Australia

Contact form

Chris Westhorpe

Director - AI, Data and Digital, Melbourne, PwC Australia

+61 402 072 972

Contact form

No search results

{{filterContent.facetedTitle}}

Related Articles
heading
Digital Pulse shares insights from PwC experts to empower your digital journey. By exploring stories at the intersection of humanity and technology we help you solve today’s business challenges to unlock the possibilities of tomorrow.
sitelinks Home Topics About
policylinks PwC Consulting Privacy Legal disclaimer
sociallinks Connect with us Email Twitter
subscribe Get the latest in your inbox weekly. Sign up for the Digital Pulse newsletter.

© 2021 - 2024 PwC. Digital Pulse shares expertise from PwC consultants and wider industry networks to provide thought leadership and actionable insights to empower your digital journey. We focus on transformational change and how this impacts consumers and business alike.

Industries
Agribusiness and Food Banking and Capital Markets Construction and Transportation Education and Skills Entertainment and Media Government Insurance Power & Utilities Retail and Consumer Real Estate Telecommunications
Asset Management Charities and Not-for-profit Defence Energy (Oil & Gas) Financial Services Healthcare Mining Infrastructure Superannuation Technology
Capabilities
Asia Practice Assurance Business Align & Connect Consulting Cybersecurity and Digital Trust Deals Digital Transformation Energy Transformation Environment, Social and Governance (ESG)
Future of Work Indigenous Consulting Integrated Infrastructure Legal People & Organisation PwC Private Strategy Tax
About Us Alumni Governance Board Diversity and Inclusion Executive Board Non-executive Director Centre Social Impact Suppliers of PwC The New Equation
Media
Insights
Careers Student careers Graduate careers Experienced careers
Contact Us Local Offices Sitemap
Digital Pulse Topics About us Subscribe

© 2017 - 2024 PwC. All rights reserved. PwC refers to the PwC network and/or one or more of its member firms, each of which is a separate legal entity. Please see www.pwc.com/structure for further details. Liability limited by a scheme approved under Professional Standards Legislation.