There’s an easy way to describe what I mean by embedded intelligence. Conjure up a memory of the best, most knowledgeable, most effectively consultative and strategic thinking HR leader you’ve ever known. Now conjure up a memory of that master of regulatory options and obligations, that HR administrator who really knows the innermost details of all your organization’s HRM policies, practices and plans and how to make them work for you. Now imagine these two wondrous beings living inside your HRM software and bringing all of their KSAOCs to bear on whatever you’re trying to do or should be trying to do in the people management business. That’s the spirit of embedded intelligence. We could call her Naomi, your friendly and very knowledgeable HRM/HRMDS consultant “in a box.”
If you’re an employee making what appears to be a simple change to your home address, Naomi would tell you, from inside the person indicative data processes of your HRMDS software platform, that:
- The new home address you’ve just input is simply incorrect (e.g. you’ve got a typo in your zip code or there’s no such street address) — and here’s how to fix it;
- There are income tax withholding changes as a result of your new home address — and here’s what they are;
- The HMO in which you’re enrolled doesn’t have jurisdiction at your location — and here’s what you need to do to select and enroll in a new HRO or other health care plan for which you’re eligible;
- You’re no longer going to receive the mass transit subsidy to which you’ve been entitled because it doesn’t apply to your new location — and here’s the amount by which your gross and net pay will be affected.
Naomi doesn’t just tell you what the issue/problem/error is, she takes you by the hand (really routing you through workflow) to the appropriate next steps in completing your business event, to include addressing all of the related ripple effects.
If you’re a manager getting ready to interview a high profile candidate for a tough to fill position on your next generation software architecture team, then Naomi would tell you, from inside the staffing processes of your HRMDS software platform:
- All that’s known about this candidate — and what more you may want to know as a part of your evaluation of her;
- Why the position in question is tough to fill, the efforts already made and the filling experience-to-date — and how you might adjust either the position attributes or the related total compensation entitlements of compete more effectively for these scarce KSAOCs (if they are indeed scarce);
- What the KSAOC profile is of those working successfully in this type of position and what the desired KSAOC profile is for this particular position — and how you might adjust those profiles in light of available developmental resources within the organization;
- What questions are best for getting at the specific KSAOCs which you’re seeking when used in an interview format — and what other interviewers have already asked and their notes on those responses;
- What your own track record has been in making hire decisions in terms of the downstream performance and retention of those hires — and what you might consider changing in order to improve your hire decisions; and
- What you may and may not ask subject to the regulations of the relevant geographic jurisdiction — and the list goes on.
Embedded intelligence, AKA Naomi, replaces, in highly automated HRM delivery systems, the human intelligence that was inserted in much more manual HRM processes when you were lucky enough to have only the very best HR professionals at your disposal, when and where you needed them.
There are many different types of intelligence that can be embedded, to include static and dynamic content, personalized and context-sensitive content, business rules and process knowledge, alerts and triggers, actionable analytics and benchmarks. When considering how best to organize, prioritize, make investment decisions and then implement embedded intelligence, it helps to have some type of a taxonomy. The one I use starts with the easy to do and gets progressively more difficult to accomplish as their impact on achieving business outcomes becomes more direct. My embedded intelligence taxonomy is:
- User inquiries to standard text;
- User-initiated standalone data changes (with attribute, event and context edits) with generated inquiries to standard text;
- User inquiries to personalized text;
- User-initiated standalone data changes (with attribute, event and context edits) with generated inquiries to personalized text;
- System-initiated distribution of standard text;
- System-initiated distribution of personalized text;
- Personal life, work life, or organizational life event initiated chain of event data forecasts and/or changes (with attribute, event and context edits) with generated inquiries to standard text;
- Personal life, work life, or organizational life event initiated chain of event data forecasts and/or changes (with attribute, event and context edits) with generated inquiries to personalized text; and
- External events with generated inquiries to standard or personalized text.
Then repeat all of the above, adding a strong advisory component, along with the relevant analytics, to each.
What’s desired is intelligent (as to content, context, business rules, analytics and advisory), pro-active, organizational role (e.g. employee or position seeker or beneficiary) and work role-based self service (e.g. manager of a specific team regardless of the team manager’s home position or generic call center representative position). It’s via embedded intelligence that we take self service from data entry to self sufficiency.
Embedded intelligence replaces what we lost when we removed the best HR professionals from most interactions with individuals and managers/leaders (or removed them altogether) — their knowledge of business rules, good practice, context, agreed upon corporate culture, regulations, labor contracts, management preferences, business unit-specific considerations, etc. It improves upon human embedded intelligence by also removing human inconsistencies, errors, misinterpretations, incomplete answers, dated information, biases, lack of availability, institutional knowledge lost through terminations or retirements, etc. Automated embedded intelligence reduces greatly the organization’s exposure in all situations where evidence of having trained/told/confirmed the relevant rules/policies/regulations at “point of sale” is an important element in preventing and/or defending against unacceptable behavior by a member of the workforce. Today, embedded intelligence is expected in any self service environment with those expectations set by commercial Web sites like Landsend.com, Amazon.com, and social Web sites like Wikipedia rather than by enterprise applications software.
The business case for embedded intelligence includes:
- Further administrative cost savings, reduced compliance exposure and reduced cycle times — and this can make or break a shared services of BPO provider;
- Improved “customer satisfaction”;
- Increased business literacy for better decision-making;
- Improved business outcomes from total compensation expenditures;
- More effective deployment of KSAOC-rich non-employees;
- Improved career-management decisions and outcomes;
- Improved productivity of individuals and groups;
- Improved selection and retention decisions and outcomes;
- Improved HRM performance metrics and actions;
- Better definition and organization of work;
- Better definition and organization of work roles;
- Improved forecasting, modeling and development of needed KSAOCs;
- Improved generation, selection, deployment and retention of KSAOC-rich persons;
- Improved generation, collection and deployment of organizational knowledge and social networks;
- Greater motivation of the workforce toward desired behaviors and KSAOC growth via targeted total compensation plans and work environment programs;
- Creation of a work environment that removes barriers to and encourages effective performance;
- More effective labor organization relations; and
- Better day-to-day coaching, mentoring, assignment, development planning and performance appraisal.
If this sounds like embedded intelligence goes to the heart not only of improving the total cost and quality of service delivery but also of improving strategic HRM, which is critical to achieving organizational business outcomes, you’ve got it in one!
So which HRM software/HRM BPO vendors are doing this well? Great question. Several of the benefits administration providers have embedded quite a lot of content within their benefits enrollment processes which can be passed through the HRM delivery system of their clients. Several of the most established HRM BPO providers, in an effort to reduce the volume, time to resolve, exposure to errors, and cost to support call center traffic have also made a dent at the basics of transactional edits and some embedded content. A few talent management software vendors have worked hard in this area, but most have neither the resources nor the expertise to get as far as they might like. We are starting to see partnerships between major HCM consultancies and talent management software vendors whose purpose it is to embed the consultant’s intelligence within the TM software, at least for clients who sign up and pay for the enhanced capabilities. And there’s a lot more coming if what I see cooking in the backrooms is a predictor. With more and more similarities among the features/functions of the TM software vendors (although their actual capabilities, in terms of their applications architectures and foundational object models remain VERY different, and there are also big differences in their strategies/progress toward SOR/TM/social tech convergence as well as their process and global footprints), I believe that the push for embedded intelligence will be a source of considerable differentiation over the next many years. Without Naomi “in the box,” HRM software has a long way to go to catch up with Amazon.