SIS INTERNATIONAL RESEARCH: NYC FOCUS GROUP FACILITY
PureSpectrum - Schedule A Demo
Our new GreenBook Directory site is live!
COVID-19 guidance, tips, analysis - access full coverage here

From Insights to Action: Walk Before You Run With AI

With new technology, it can be easy to get carried away. Don't let AI get the best of you, avoid common mistakes with implementing AI with these introductory tips.

Insights and marketing research teams in every industry are faced with the issue of holding a bounty of information without an effective way to search or share the unstructured data or knowledge assets owned by the company. 

In a world of expansive research and data, there is simply too much to learn. Nobody can find the information. Or worse, nobody knows that it exists. 

This is the reason why so many data-driven brands are planning, budgeting, piloting, or fully deploying AI-powered knowledge management systems right now. 

Just as every business has been affected by digital and mobile technology, every business will be impacted by artificial intelligence (AI). That being said, there needs to be a rigorous test to determine if the AI is ready for your particular needs when it comes to data, research, and insights.

Businesses that take the time to evaluate different AI-powered knowledge management systems with realistic expectations and make strategic plans for the implementation and adoption of a system, will have a competitive advantage.

Walk Before You Run With AI

The first step to evaluate solutions is to find any low-hanging fruit in terms of the problems you’d like to solve, the employees you’d like to support, and the data you’d like to draw insights and research from.

Some common challenges that enterprises can solve for include:

  • Redundant Research: Millions of dollars are spent recommissioning research that has already been completed.
  • Dark Data: After 90 days, documents and data that are not actively in use are likely never to be opened or used again.
  • Employee Turnover: When someone starts, leaves, or changes teams, previous knowledge is lost. Finding answers means deep exploration or duplication of the predecessors’ work.
  • Acquisition Chaos: Times of rapid change during mergers or acquisitions require teams to get up to speed on a new influx of data.
  • The Hunt: Storage systems require users to recall document titles, dates, or tags to search. This limits how people search for content and doesn’t point to answers or specific data within documents. Furthermore, the categorization of documents is rarely consistent between people and teams, making it even harder to find information.

From our perspective, we tend to look at the biggest information challenge. Specifically, what is the data that you wish was being used better? Or, what is the data that you struggle to find? 

Imagine being able to leverage video assets such as focus groups, presentations created by teams in different locations, audio recordings of meetings, and consumer research tucked away in PDFs simply by asking a natural language question and then you are delivered your answer, wherever it lives, in seconds. 

These are just a few examples from our lens; however, there are all sorts of opportunities in information technology and human resources where AI-powered knowledge management can also have a profound impact across the organization.

When you understand the capabilities of the vendor, that’s great. But it’s fundamentally better to focus on the problems you and your team are experiencing, determine if they can be solved for, proved with a big gain, and to start there with a pilot program.

Example Pilot Programs for AI Knowledge Management Systems

Rather than jumping in headfirst at the enterprise level, it is best to identify specific use cases for a pilot program. This involves identifying certain data and groups to solve for, executing on a solution, then rolling out that success to other users and use cases. One way to identify a starting use case is to ask your teams about their pain points or challenges while gathering information. The following are examples of places to start.

Organizational Efficiency

Smart companies are looking for efficiencies where they can, in every corner of the company.

It is a fact of everyday life that insights departments have to continuously try to do more with the same or less. One strategy is to help other employees extract all the possible value they can from the insights and research that has already been created. Knowledge created for one reason can be repurposed for another.

While processes have definitely improved over the years, frustrations remain when employees are shown shared drives with endless amounts of assets they will never find, read, or obtain knowledge from.

Imagine instead a system where everyone is empowered with the company’s legacy knowledge. They can ask specific questions regarding the task at hand, instantly find answers within internal assets and resources, and move forward with their work without wasting time searching or repeating work already done. Data-driven decisions can now be made at the drop of a hat.

Reduce Demand on Subject Matter Experts

When a company relies heavily on subject matter experts, those individuals are often overloaded with one-off requests beyond their regular responsibilities. 

Regardless of processes put into place to document and train other employees, individuals still reach out to the experts for information on a regular basis. This inhibits their productivity and value to the company. And if they come to a breaking point, some SMEs choose to leave, resulting in turnover issues and lost knowledge. In this use case, the goal is to empower teams to rely less on subject matter experts for one-off requests.

Does this type of scenario circumvent the importance of subject matter experts? No. But, it is critical to communicate the benefit of gaining back time to focus on their high-priority responsibilities. 

Bring More Research In-House

Large companies tend to rely on outside vendors to conduct research. These projects tend to be expensive and are often repeated as different brand teams target the same market and are unaware that outside research was already shopped out and delivered.

It’s easy to see how redundant research happens. A brand gets a new manager who identifies the need for a new marketing plan to reach the brand’s target consumer. The brand manager has no efficient way of investigating whether the previous team had valuable insights that could be reused, or whether other divisions of the company have run analysis on the same sort of consumer.

Without an effective way to quickly find answers, they need to start from scratch. The process takes a month and has a sizable budget. A few years later, they move on to another position.

Their replacement starts in the same position – with questions that need to be answered, and with no easy way of searching existing company assets for relevant data. So, the cycle of redundant research begins anew.

To choose a team or brand for a pilot implementation, look for early adopters of other technologies who innovate and lead naturally.

Getting Started

Defining a starting point for a successful implementation is a critical step in the process. 

Remember to start small with a targeted pilot program and grow adoption through positive use cases around small user groups, individuals with the biggest need, or highly sought after content.

For more information on evaluating, implementing, and adopting AI-powered knowledge management systems, check out our full guide here.

Please share...

Join the conversation