Talent Tech Is Hot … But Without the Right Foundation, It Will Likely Disappoint

There is a revolution currently occurring in talent management and recruiting. Many firms are following Google’s lead and shifting to a data-driven model throughout all of the talent management functions. But an equally hot and exciting area is the development of talent technology, or talent tech for short. New vendors in the talent tech area seem to be popping up each month. But don’t get carried away with excitement because I have found that up to 80 percent of all corporate talent management technology purchases don’t produce a measurable ROI for the corporation within two years.

Many of the reasons for that high failure rate can be attributed to bad purchase decisions that are made without an objective checklist covering the most critical vendor and product assessment criteria (note: I will provide a sample of my checklist in my November 7, 2016, ERE.net article entitled Talent-Tech Is Hot … However, Without a Vendor Assessment Checklist, A Bad Purchase Is Likely). But other than weak vendor assessment, the other key failure factor is a weak foundation within HR that is unable to support the implementation and operation of new talent technology.

Don’t Purchase Talent-Tech Until You Have These Seven Internal Foundation Elements in Place

Many talent-tech purchases are doomed from the start because firms that are buying it do not have the needed internal foundation to support almost any of the new tech solution. Through my research and experience, I have found that there are at least seven important “internal foundation elements” that must be in place before a corporation purchases any talent-tech solution. These foundation elements include:

The primary goal must be to improve business results (not HR efficiency) — leaders in talent management are obsessed with efficiency. But that can be a major stumbling block because the ultimate goal and success measure for any talent-tech purchase must be to measurably improve business results. Yes, efficiency is necessary, but it should not ever be confused with business impacts. The business impacts that count are always defined in the corporation’s strategic business goals. Therefore, the first step for talent leaders is to determine, up front, which business goals the talent technology has the highest chance to impact. For example, in recruiting, hiring better performers in key positions (because that increases performance) and reducing the number of position vacancy days in revenue jobs (because that increases revenue) are the top two most important direct business impacts. In the area of training and development, simply offering more training or making the scheduling of training classes more efficient are not strategic business impact. Instead, show that the performance of those who complete training and development increases dramatically within months after completing it. So, before you buy any talent tech, avoid intermediate goals like improving sourcing, job matching, or interviews and instead set business impact goals like improving the on-the-job performance of new hires by 10 percent (when compared to the average performance of last year’s new hires).
You’ll need a great business case in order to pay for the technology — obviously, you will not be able to buy the quality and the volume of the technology that you need without executive support and funding. If your vendor is any good, they will supply you with a pretested outline (pretested with various CFOs) of a proven business case that will enable you to quickly convince your executives to give enough money to purchase their product. If they don’t know the elements of a business case for purchasing their product, it might be safe to assume that this isn’t a good vendor to deal with. However, in the absence of a vendor’s help, work with your own CFO and CIO to identify both the costs and benefits that would make this a compelling business investment. Quantify each of your business case arguments in dollars so that executives from every area can understand its impact.
Focus your talent-tech purchases in the program areas that have the highest impacts — before buying any technology, the very best talent functions identify and then prioritize the individual program areas that have the highest business impacts. Every organization is different, but as a general rule, the highest impact talent areas are recruiting, retention, employee productivity, and employee innovation. Avoid the common mistake of purchasing talent technology in a functional HR area simply because new technology is available in the marketplace. Without evidence that a particular talent area has a high impact and that it needs significant improvement at your firm, the ROI of the talent-tech purchase simply can’t be high (even though vendors are focusing on it). For example, it makes no sense to purchase talent tech in the area of reference checking unless you have data that shows that reference checking has a significant impact on the quality of hire and that your firm is currently not doing reference checks well.
The talent area of the purchase must already be data-driven and the required data must be available — one of the primary reasons for buying technology is that it is digital, which usually means that it uses electronic data to improve its results. However, if HR or the functional area of the purchase isn’t already data driven, the operational success of the technology is already in doubt. Being data driven means that you must clearly define success using either numbers or dollars. And then you must have metrics and a metric process for periodically measuring that success. And unfortunately, that means if that talent area doesn’t already have extensive performance metrics, you can never know if the technology has the desired impact or a positive ROI. Fortunately, in some cases, technology packages already have the appropriate metrics embedded in them. But any set of metrics can only work if your organization is already gathering the necessary data in each of the important metric areas. And to further complicate matters, when you’re dealing with technology that involves algorithms and artificial intelligence, the volume of data that you need increases dramatically. So until you have defined success, created success metrics, and ensure that you have the data necessary for those metrics, you shouldn’t move forward with most technology purchases.
Weak talent processes must be fixed first — when you insert technology into poorly designed talent processes, you are reducing your chances of success to near zero. So, that means that currently bad systems, weak processes, and haphazard data collection must be fixed first. I know of no exception to this rule. In some other cases, your internal systems may be so complex, full of workarounds, and so outdated that applying technology to them is equivalent to putting a new engine in a dilapidated car. Obviously, talent functions must be willing to invest the time and resources required in order to re-engineer its own internal processes before you apply talent tech to them. Unfortunately, most talent leaders have long ago forgotten how to re-engineer processes to increase their speed, quality, and capabilities. In most cases, the first step is to work with your potential vendors to identify the minimal operational requirements, so that you know in advance what processes you need to re-engineer.
Make sure that your managers will actually use the outputs of the technology — even though technology will dramatically increase your capabilities, you need to consider the possibility that once these new capabilities and outputs are delivered, and no one will actually use them. Historically, one of the major problems with talent-technology solutions has been the failure for users to fully take advantage of the results that they produce. For example, predictive analytics that warns you about upcoming employee turnover will produce absolutely no business impact if the managers ignore the turnover alerts that you sent to them. Even if a manager is willing to take action, if you don’t provide them with “prescriptive solutions” (i.e. the best tools for keeping these employees), they may in frustration actually take the wrong actions, which may actually in this example result in increased turnover. The lesson to be learned is that if you want to ensure the full usage of new technology, work with a sample of managers who will use the outputs to overcome any resistance factors and to convince them that it’s in their and their team’s best interest to fully use the new technology. You might also reveal the percentage of productivity improvement they can expect after utilizing the technology.
HR must have the level and type of talent necessary to manage the technology — you don’t need to be a rocket scientist to realize that all new technology needs a support team in order to be effective. Unfortunately, I found that many in HR fear technology and metrics, so the only option might be to hire technologists and data scientists into HR. But even that can be problematic because they may be forced to work in isolation because of the clash between the lovers of technology and the larger number of the defenders of historical HR practices. Also be aware that managing talent-tech vendors themselves also requires advanced vendor management skills within HR.

Don’t Purchase Talent-Tech Until You Have These Seven Internal Foundation Elements in Place

The primary goal must be to improve business results (not HR efficiency) — leaders in talent management are obsessed with efficiency. But that can be a major stumbling block because the ultimate goal and success measure for any talent-tech purchase must be to measurably improve business results. Yes, efficiency is necessary, but it should not ever be confused with business impacts. The business impacts that count are always defined in the corporation’s strategic business goals. Therefore, the first step for talent leaders is to determine, up front, which business goals the talent technology has the highest chance to impact. For example, in recruiting, hiring better performers in key positions (because that increases performance) and reducing the number of position vacancy days in revenue jobs (because that increases revenue) are the top two most important direct business impacts. In the area of training and development, simply offering more training or making the scheduling of training classes more efficient are not strategic business impact. Instead, show that the performance of those who complete training and development increases dramatically within months after completing it. So, before you buy any talent tech, avoid intermediate goals like improving sourcing, job matching, or interviews and instead set business impact goals like improving the on-the-job performance of new hires by 10 percent (when compared to the average performance of last year’s new hires).

You’ll need a great business case in order to pay for the technology — obviously, you will not be able to buy the quality and the volume of the technology that you need without executive support and funding. If your vendor is any good, they will supply you with a pretested outline (pretested with various CFOs) of a proven business case that will enable you to quickly convince your executives to give enough money to purchase their product. If they don’t know the elements of a business case for purchasing their product, it might be safe to assume that this isn’t a good vendor to deal with. However, in the absence of a vendor’s help, work with your own CFO and CIO to identify both the costs and benefits that would make this a compelling business investment. Quantify each of your business case arguments in dollars so that executives from every area can understand its impact.

Focus your talent-tech purchases in the program areas that have the highest impacts — before buying any technology, the very best talent functions identify and then prioritize the individual program areas that have the highest business impacts. Every organization is different, but as a general rule, the highest impact talent areas are recruiting, retention, employee productivity, and employee innovation. Avoid the common mistake of purchasing talent technology in a functional HR area simply because new technology is available in the marketplace. Without evidence that a particular talent area has a high impact and that it needs significant improvement at your firm, the ROI of the talent-tech purchase simply can’t be high (even though vendors are focusing on it). For example, it makes no sense to purchase talent tech in the area of reference checking unless you have data that shows that reference checking has a significant impact on the quality of hire and that your firm is currently not doing reference checks well.

The talent area of the purchase must already be data-driven and the required data must be available — one of the primary reasons for buying technology is that it is digital, which usually means that it uses electronic data to improve its results. However, if HR or the functional area of the purchase isn’t already data driven, the operational success of the technology is already in doubt. Being data driven means that you must clearly define success using either numbers or dollars. And then you must have metrics and a metric process for periodically measuring that success. And unfortunately, that means if that talent area doesn’t already have extensive performance metrics, you can never know if the technology has the desired impact or a positive ROI. Fortunately, in some cases, technology packages already have the appropriate metrics embedded in them. But any set of metrics can only work if your organization is already gathering the necessary data in each of the important metric areas. And to further complicate matters, when you’re dealing with technology that involves algorithms and artificial intelligence, the volume of data that you need increases dramatically. So until you have defined success, created success metrics, and ensure that you have the data necessary for those metrics, you shouldn’t move forward with most technology purchases.

Weak talent processes must be fixed first — when you insert technology into poorly designed talent processes, you are reducing your chances of success to near zero. So, that means that currently bad systems, weak processes, and haphazard data collection must be fixed first. I know of no exception to this rule. In some other cases, your internal systems may be so complex, full of workarounds, and so outdated that applying technology to them is equivalent to putting a new engine in a dilapidated car. Obviously, talent functions must be willing to invest the time and resources required in order to re-engineer its own internal processes before you apply talent tech to them. Unfortunately, most talent leaders have long ago forgotten how to re-engineer processes to increase their speed, quality, and capabilities. In most cases, the first step is to work with your potential vendors to identify the minimal operational requirements, so that you know in advance what processes you need to re-engineer.

Make sure that your managers will actually use the outputs of the technology — even though technology will dramatically increase your capabilities, you need to consider the possibility that once these new capabilities and outputs are delivered, and no one will actually use them. Historically, one of the major problems with talent-technology solutions has been the failure for users to fully take advantage of the results that they produce. For example, predictive analytics that warns you about upcoming employee turnover will produce absolutely no business impact if the managers ignore the turnover alerts that you sent to them. Even if a manager is willing to take action, if you don’t provide them with “prescriptive solutions” (i.e. the best tools for keeping these employees), they may in frustration actually take the wrong actions, which may actually in this example result in increased turnover. The lesson to be learned is that if you want to ensure the full usage of new technology, work with a sample of managers who will use the outputs to overcome any resistance factors and to convince them that it’s in their and their team’s best interest to fully use the new technology. You might also reveal the percentage of productivity improvement they can expect after utilizing the technology.

HR must have the level and type of talent necessary to manage the technology — you don’t need to be a rocket scientist to realize that all new technology needs a support team in order to be effective. Unfortunately, I found that many in HR fear technology and metrics, so the only option might be to hire technologists and data scientists into HR. But even that can be problematic because they may be forced to work in isolation because of the clash between the lovers of technology and the larger number of the defenders of historical HR practices. Also be aware that managing talent-tech vendors themselves also requires advanced vendor management skills within HR.