You wouldn’t take on a new employee, pay them a salary and then ask them to sit back and take it easy. Yet, many enterprises do this with their data. They invest in data management and storage to meet regulatory requirements, but it sits within the organization underutilized and becomes ‘lazy’ data. 

Every biotech, whether a CRO (Contract Research Organization) or CMO (Contract Manufacturing Organization), has a mountain of data that is full of potential (you can get an idea of how much potential here). But how can they go about tapping into it? How can they turn their ‘lazy’ data into a workhorse of business value?

What Makes Data Lazy?

Lazy data is simply data that has untapped potential – it is not being used to add value to the business; it just takes up space. One of the most common reasons that data is under-utilized is that large datasets can be unwieldy. It can be challenging to handle the volume of data your organization produces, so by default, data often sits ignored in log files, enabling it to be ‘lazy.’

Leveraging vast quantities of data in unstructured or weakly structured formats is daunting, and the skills and effort needed to implement a solid data strategy up in-house can be a barrier to entry. There is often an internal skills gap, where teams have the scientific or engineering knowledge to be successful in their roles, but lack the technical data infrastructure expertise to accomplish certain organizational goals. Teams don’t have the expertise to build the data pipelines and workflows needed to be effective. You could hire to fill the gap, but increasing headcount to achieve an unclear ROI is an impossible sell.

Identifying where to start is a secondary challenge. For biotech companies to see the benefits, they need a strategy that makes their data work hard and fast, through a series of ‘quick wins.’ Without the knowledge and experience to know where those opportunities sit within the organization, disappointment with the results will be almost inevitable.

Can Out-of-the-Box Work?

It is possible to run a solid data strategy with your existing team. Businesses can have under-utilized employees that could hit the ground running and drive value. But, to have the full skill set, with the bandwidth to run a data competency program that unlocks your data’s full potential, is somewhat unrealistic. What are the alternatives?

There are out-of-the-box solutions for data analytics, and there is certainly a place within some enterprises for a plug-and-play option. However, that place is not typically within a biotech firm. The challenge with an out-of-the-box or generic Data Pipeline or Business Intelligence (BI) solution is that it makes assumptions about or imposes limitations on the structure of source data, how that data needs to be processed to capture value, and the types of questions and uses that deliver value. Biotech firms have highly customized infrastructure related to the types of outputs they research and manufacture, and that customization guides how analysis of R&D and manufacturing processes will accrue to the bottom line. The questions that biotech companies need to answer in designing a data pipeline require both an understanding of the scientific phenomena occurring and the measurements represented by the data produced when those scientific phenomena occur. 

With over 50% of the onshore delivery team coming from a science background, many with advanced studies in physics, chemistry, and biology, the CorrDyn team excels at connecting the dots between inputs and outputs of this analytics process in a biotech context.  Biotech firms need a solution that makes their data work hard out of the gate, not out of the box, and having specialist expertise is part of delivering this.

What Hard-Working Data Looks Like

The equipment used in biotechs; DNA sequencers, synthesizers, gas or liquid phase systems, volume measurement systems, electrospray ionization or optical density measurement systems, reagent tanks and scales, cleavers,  homogenizers, ERP systems, sample storage systems, IR spectrometers, mass spectrometers, Raman spectrometers… the list goes on. This is complex and often bespoke technology, so it makes sense that the data strategy should be too.  For most biotechnology companies, the best approach for squeezing value from their data is to work with an external data specialist, and crucially, one who understands the biotech industry.

Data is only lazy when no systems are in place to derive value from it. An external data team can bring the expertise to make the data work hard and offers several critical advantages for biotechs about to embark on a data analytics journey:

1) Tried and tested – A team that brings data infrastructure and business intelligence expertise and understands how to apply it to scientific industries is invaluable. They understand the type of data that biotech manufacturing equipment generates. They know how that data sits within an organization just beginning its data competency journey. Most importantly, they understand where to start managing it and how to transform it from an unwieldy mass of lazy data to a hard-working and valuable tool capable of answering the questions they want to address.

2) Efficient transformation – When looking to make the most of the data you have for the first time, the infrastructure in place is likely not capable of delivering everything needed from a data analytics perspective. The infrastructure is likely designed to collect and store data but is not geared for scaled computation. An external team has the expertise to come in and deliver what is needed efficiently and effectively. For example, at CorrDyn, we provide critical guidance and support to develop the essential on-premise infrastructure and then deploy a cloud-based solution enabling elastic scaling capabilities. This allows us to scale analytics for our clients as they need them and within budgetary constraints. without significant disruption to ongoing processes.

3) Navigating tradeoffs: A good partner tells you what you need; a great partner tells you what you don’t need. By understanding the elements of a data strategy that will drive the most significant outcomes, an external data team can guide you through which investments are unnecessary or won’t generate sufficient ROI. For example, there is an assumption that real-time analysis is always better. In some cases, this is necessary, but rarely at the cost required to achieve that level of latency. Real-time analytics are more complicated and costly. We will often recommend a near real-time strategy as this solves the vast majority of use cases but is substantially more cost-effective than a true real-time option.  

4) Capacity on tap: One of the downsides of an in-house team is that a data analytics program's needs are inconsistent. The start of a project needs a larger investment, but as it becomes a well-oiled process, less capacity is needed. There will be other points where demand for data analytics will surge, for example, when a new product launches, and managing the peaks and troughs in-house is challenging. By working with an external team, biotech firms have the capacity they need on tap. This is how CorrDyn’s flexible analytics team model works. With ongoing projects, we help clients deal with surges in demand but with the ability to scale back when the need is low, for a more efficient overall approach. 

Return that Justifies the Investment

You may think that all sounds great… but it also sounds expensive. Can I justify the cost? From our point of view, the answer is yes, but let us substantiate that answer.

Let’s look at all the options we have considered for tackling lazy data:

• Adding to the workload of the existing team risks not having enough expertise to drive results or always deprioritizing this important work for more urgent internal efforts
  
  
• Hiring an in-house team is an involved, expensive, and potentially risky process. Who hires the first hire, and how do you make sure that they have the expertise to deliver? Plus, once the strategy is up and running, it may only be needed at a partial capacity
  
  
• Out-of-the-box is typically not flexible enough to meet the specific needs of most biotech firms

Working with an external team provides a middle ground on the short-term investment level, but crucially, it drives return in a way that justifies the spend. A data team with dedicated biotech experience has the skill set to navigate implementing a solid strategy and deliver the ‘quick wins’ that offer immediate value from previously lazy data. 

Let’s see what that looks like for biotech manufacturers: 

• Making the right decision on approach: As we have discussed, real-time is not always necessarily the right business decision, a data consultant helps to guide the right approach for your business goals
  
  
• Seeing what’s happening on machines in (almost) real-time. This insight allows processes to be fine-tuned quickly and at an acceptable cost to drive maximum efficiency
  
  
• Using data to inform conversations about machine or product failures within minutes. Without actionable insights, troubleshooting problems with machines can be a long process that results in downtime, incurring high costs. Understanding what the problem is through data analytics can speed up the process and adds immediate value
  
  
• Being able to detect and alert on known faults in near-real-time. Similar to machine diagnostics, simply identifying known defects quickly helps to ensure minimum wastage as machines do not produce defective products due to a fault
  
  
• Being able to provide process managers with near-real-time insights on processes. As data analytics can be automated to take some of the heavy lifting from the shoulders of process managers, their time can be better spent managing processes, using the insights to make improvements. 

Traction with these goals is typically achievable with an initial investment in CorrDyn of under $150K and less than 9 months. This is lower than the cost of a single full-time employee’s annual salary, with several times the efficiency, and with a full team of experts brought to the table. 

Similar principles can be applied to all areas of biotechnology firms to make data work hard, from R&D to product delivery. With time, a data partner can work to build more effective models that deliver deeper business insights, but at least some value needs to be delivered fast. Working with data experts that understand the industry allows these quick wins to be identified immediately, and organizations can start to reap the benefits of the results. 

Conclusion

We are being tongue-in-cheek when we talk about the lazy data within biotech firms, but it underlines a genuine point. Underused data is a massive opportunity for them to drive organizational efficiency. It is a valuable asset, and nothing gives greater insight into a company’s strengths and weaknesses than its own data. By implementing a strategy that immediately delivers quick wins, and drives deeper organizational benefits over the long term, biotech firms unlock a gold mine of insights that can capture actual value for the business as a whole. 

Get in touch here if you want to hear more about how CorrDyn can help transform lazy data into a hard-working insights machine.