When it comes to AI driving productivity, the standard concern (at least outside the contact center) seems to be: What if people don’t efficiently redeploy all the time AI saved them to do even more work? What if they just end up goofing off?
Aside from the unflattering picture that paints of people whose employers tend to describe them, in other contexts, as their most important corporate resource, I’m not sure it accurately depicts how organizations tend to operate. I have to question whether any organization is built—or can be built—to maximize employee efficiency.
A recent survey from software vendor Quickbase details how much time employees spend slogging through unproductive tasks, and according to the report, more than half the workers surveyed (58%) spend more than half their time on efforts that they don’t consider “meaningful work that drives results.” What are these unsatisfying, time-wasting efforts? One of the main culprits is the state of enterprise data today. According to the report: “45% of respondents say they are spending 11+ hours per week chasing information from different people and systems.”
The current assumption among many in the AI world seems to be that all we have to do to unlock maximum employee productivity is clean our enterprise data and get it into the LLM (or LLMs) that drive the AI use cases. Then employees will have a single source of truth that can answer their questions, write their reports, and take care of other routine, boring tasks – freeing up their time for still more productive pursuits.
That data cleansing and integration effort is arduous enough. We also have to ask why most enterprises’ data is so varied, scattered, and generally messy in the first place. It’s not like anybody planned it that way or is doing it to torment users. Enterprises are dynamic environments, where different organizations and teams have different needs and practices, so in large enterprises it’s neither quick nor easy to effect system-wide standardization of any sort. And even when you can do that, it’s likely just a matter of time before your enterprise makes a big acquisition, or gets acquired, and the problem starts all over again.
People are messy, and organizations made up of people are exponentially messier. We can use AI to improve our processes and tools, and probably make employees incrementally more productive. But the idea that AI can overcome barriers and silos that people raise up for human reasons (usually unintentionally) seems unrealistic.
We’ll be exploring these questions around productivity, data readiness, and many other core issues at our first-ever Enterprise Connect AI event Oct. 1 – 2 at the Santa Clara, CA, convention center, for which registration is now open. We’ll be announcing the first elements of the conference program next week, so stay tuned – this event will be tailored to the needs of IT decision-makers who know they have to start planning now for Generative AI, and are grappling with the strategy to move their enterprise forward. I hope you can join us!
