Five Steps to Build your Real-Time Data Ingestion Business Case

by Equalum

Hero Image
June 25, 2019 2:56pm

Blog

Technical leaders juggle dozens of mission-critical priorities daily. System health and performance. Maintenance and upgrades. Security. Scaling a high-performing team that can navigate complex systems.

With so many priorities, it can be difficult to build a clear and compelling case for real-time data ingestion. After all, there’s a lot of buzz about the need to become real-time: but does it really make a difference?

A glance across key use cases reveals that there are indeed critical applications for real-time insights. Even so, it can be challenging to build a case for prioritizing those amongst others, perhaps more immediately visible, priorities.

Here’s a proven five-step approach to building a compelling business case:


Step 1: Coordinate with business owners to scope the opportunity:

Coordinate with business stakeholders, ideally those with P&L ownership, to understand the levers that would move the needle on business performance. For financial services, for example, this might involve real-time analytics to power the loan approval workflow (higher conversion rate and more loan origination) or fraud detection (loss prevention). For industrial manufacturing, this may involve analytics to optimize preventative maintenance (higher throughput and lower capital expenditure). In each case, there’s a clearly hypothesized link between an output of real-time analytics – and a critical business lever that would unlock growth opportunities.


Step 2: Articulate the state of current ETL and other infrastructure efforts:

Take stock of your organization’s current approach to data ingestion. Are you currently transporting data throughout the enterprise with homegrown scripts or a legacy ETL platform? If so, what is the cost associated with the maintenance of your current ETL systems? Has your organization invested in any exploratory big data projects so far – perhaps working with open source frameworks like Apache Spark or Kafka? Considering efforts that have already been undertaken will be crucial to gain stakeholder buy-in and avoid duplication of effort. For example, it would be difficult to convince an organization that has invested years in exploring open-source frameworks to work with a data ingestion vendor that doesn’t leverage those investments.


Step 3: Identify a partner who can help address key opportunities:

Consider both your organization’s short-term and long-term needs. In the short-term, it’s important that you be able to solve for the particular business problem you’ve identified. In the long-term, your organization is most likely confronting several key trends: an explosion of data volume; the “democratization” of big data, which places a premium on usability by business stakeholders; and greater incorporation of real-time insights into operational processes, creating a need for performance, flexibility, robustness, and fault-tolerance. Consider these future needs as you identify a prospective partner to accelerate real-time data efforts.


Step 4: Prioritize a single use case for a proof of concept:

It can be tempting when working with a new partner to boil the ocean: to test the most challenging data flows or transformation logic in order to dazzle the organization. Resist the urge. Instead, target low-hanging fruit by chiseling down the problem into a use case that has immediate value – and can help evangelize the applications of real-time analytics. For example, a retailer looking to leverage customer insights to personalize a website in real-time (in order to boost conversion rate) might consider focusing on just a single part of the site rather than the whole experience.


Step 5: Measure, optimize, and scale:

Benchmark performance and key business metrics before and after the implementation of a real-time data ingestion solution. Are conversion rates improving? Equipment costs declining? What about non-tangible costs (like maintenance costs associated with legacy systems?) Starting small also offers the opportunity to make implementation optimizations – to increase performance and ease of use, for example – before blowing a successful solution out to scale. Once results are clear and well-understood by the organization, it will be increasingly easy to scale past simple use cases to tackle the full breadth of the business problem.

Ultimately, following this roadmap – with its emphasis on needs analysis and progressive wins – can help create a business case for real-time data ingestion that will resonate with skeptical stakeholders.


Equalum can help you

Feel free to contact Equalum to help you to accelerate and radically simplify the deployment of real-time data ingestion. Contact us for a demonstration.

Ready to Get Started?

Experience Enterprise-Grade Data Ingestion at Infinite Speed.