12 min read
Mar 05, 2020
Human resources

A 5-Step Framework for Implementing Artificial Intelligence (AI) in HR

The AI revolution is around the corner. Is your HR department ready?

Harshit SrivastavaSpecialist Analyst

Did you know? According to a Gartner report, about 85% of businesses could start using artificial intelligence (AI) in their human resources practice by 2022. (Full report available to Gartner clients.)

AI is one of the most transformative technologies, with the ability to disrupt all areas of a business. Human resource (HR) departments are not immune to its disruptive impact. In fact, the fast-paced growth of AI will force every business to rethink its entire HR strategy within the next two years.

But first, let’s understand the factors behind this fast-paced AI growth:

  • Wide range of use cases for HR. AI models can help HR teams analyze resumes to identify high-performing recruits, predict attrition, and boost employee engagement via chatbots. 

  • Business profitability. AI not only makes HR processes more efficient, but it also automates HR processes such as resume screening and candidate scoring.

  • Increased availability of AI-based HR services. The service-based approach enables business professionals without technical expertise to implement AI in their practice.

When it comes to AI in HR, however, there’re a lot of mixed opinions. While some people feel that AI will lead to job cuts, others feel that it will create more jobs. The resulting confusion paralyzes the decision-making capacity of HR leaders. 

To help you declutter your thoughts on AI, we’ve analyzed the two biggest trends driving the AI disruption in HR departments. Based on these, we’ve also formulated a five-step framework that can put your business at the forefront of the AI revolution.

Group 3@1x Created with Sketch.

Trend #1: The AI revolution is much closer than you think. Brace yourself!

If you haven’t yet started using AI in your HR practice, the time to act is now. According to a Gartner report, about 55% of businesses have already started using or are piloting AI-based projects in their HR departments. (Full report available to Gartner clients)

The recent developments in this field have far more transformational effects on HR departments than ever before. 

For instance, IBM’s AI is able to predict which employees will quit with an accuracy of 95%. With such a highly accurate prediction rate, the results will help you make better decisions related to talent management. Managers can reach out to high-risk employees so they can better understand their challenges and come up with ways to motivate them to stay. HR managers can also use these inputs to revise existing employee policies to create a more positive work environment.

The fast-paced development of high-impact AI solutions tells us one thing: We are on the brink of an AI revolution in HR, and all businesses must act now.

Group 3@1x Created with Sketch.

Trend #2: If lack of tech expertise is holding you back, don’t wait. AI is not just for techies!

One of the biggest myths around AI is that it requires extensive technical knowledge and therefore, businesses need to hire developers and data scientists. This tendency is revealed by a recent study that found that about 43% of HR leaders feel that they will not be able to keep up with AI developments over the next decade.

But, contrary to popular notion, AI does not require much technical knowledge. What has made this possible is the service orientation of AI solutions:

  • Software vendors developing AI-powered features: As AI has become widespread, software vendors have started embedding AI technologies into their products. For instance, Zoho Recruit uses AI to assign a "percentage match" score to help recruiters prioritize the candidates that closely align with a particular job requirement.

  • AI-based solutions offered as third-party integrations: Some AI tools offer features such as candidate screening, process automation, and chatbots that integrate with your current HR and applicant tracking systems (ATS). One example is Ideal, an AI-based recruiting solution.

  • Independent AI-based services: The third factor simplifying AI for non-technical users is the proliferation of businesses offering AI-based services. For instance, Clarity, a background-check tool built on AI, uses thousands of data points to perform checks with a high accuracy rate of 90%.

Learning to use AI technology is akin to learning any software solution: You’ll need to spend some time learning its interface and features, but you can quickly become proficient.

Group 3@1x Created with Sketch.

A 5-step plan to stay on top of the AI revolution

It’s clear that businesses need to act now. With careful planning and preparation, you can reap the benefits from early adoption of technology and also gain an edge over the competition.

However, to successfully deploy an AI solution at your business, you need a calculated and piecemeal approach: assess the impact of using AI and scale it per your business requirements.

To help your business implement AI in HR, we’ve created a five-step AI framework. This framework will help you achieve your AI-implementation goals while avoiding immediate disruption to your HR practice.

Step 1: Identify AI opportunities in your HR department

Your approach to AI adoption should be strategic and well-planned. To achieve that, we recommend using an “urgent vs. important matrix” to identify and prioritize AI applications:

List all the AI-based applications relevant to your HR practice, such as candidate scoring, attrition prediction, interactive chatbots, performance analytics, and resume parsing.

Classify these applications as important and not important. The important applications are the ones that will help you solve high-priority challenges. We recommend creating a table similar to the one below:

The next step is to classify these on the basis of urgency. Urgent applications are the ones that will help you solve imminent threats to your business. For instance, if your business’s current performance practices are perceived as unfair and are causing attrition, then "employee performance measurement" will be an urgent application. Sometimes, if a lot of your competitors are using an application, thereby giving them a considerable human resource advantage, these apps should also be classified as urgent. We recommend creating a similar table as below:

Use the tables above to create an ‘Urgent Important’ matrix similar to the one below:

Step 2: Prioritize AI applications according to the matrix

After you’ve created the urgent versus important matrix, the next step is to establish a deployment timeline according to the matrix findings. 

The top left, important and urgent application should take top priority on your list, followed by the urgent, but not important items, as failing to implement those could have negative consequences. 

The next priority should be important, but not urgent applications, and finally, the bottom right box. 

After you’ve completed these steps, the priority table would look like this:

Step 3: Estimate cost of deployment of AI applications

Budget, especially for small and midsize businesses, is a key factor when purchasing a technology solution. For this reason, the next step is to select the set of applications that will fall in your budget. We recommend the following approach:

  • Check if your software vendor already offers these applications. Estimate the cost of an upgrade to include AI features.

  • Perform an online search to identify products and services specific to your AI requirements.

  • Set a yearly budget for AI implementation. For instance, if your budget is $2,000, you can select the top two applications from the table to prioritize for now: 

Step 4: Set a timeline for implementing these applications

The next step is to set a realistic deployment timeframe for each of the applications, targeting high-priority applications first.

The ideal way to set a timeline is to contact the vendor or service provider and inquire about the time required for complete deployment. It’s important to note that timeframe should not just include software implementation, but also the time it will take your staff to start using the software.

Step 5: Train your employees for using these applications

Employees are at the heart of a successful AI implementation strategy. For this reason, it is important that you start training your HR staff on how to use AI-based systems. 

  • Identify the right stakeholders for AI projects: Look for team members who show an interest in AI and have the necessary background knowledge—employees with a background in analytics and that are currently using HR software are ideal targets.

  • Brief stakeholders on expectations: Create a set of objectives and share it with the selected staff members. This knowledge will help them create AI strategies in line with business expectations.

  • Sponsor AI-related courses: You can also select courses that align with your business objectives. For instance, you can train an employee on chatbots, and designate them with the responsibility of creating and deploying a chatbot solution.

Group 3@1x Created with Sketch.

Future-proofing your AI strategy

After you’ve created a framework based on your business requirements, the next step is to future-proof it for long-term use. To ensure the success of your AI-strategy, there are two additional considerations you need to be mindful of while implementing AI strategy. These are:

  • Data security: AI-based solutions need extensive access to sensitive employee and company data, so you need to be extra cautious while evaluating these solutions for data security measures. To help you better secure your HR data while implementing

    AI, we recommend using the 5-step framework to avoid a data breach.

  • Revise AI strategy quarterly: AI is evolving fast, bringing new innovations every day. In order to be competitive, you need to constantly assess the technological developments in the market, and revise the urgent-important matrix regularly.

Back to top