After the hype: what kinds of intelligence should you invest in?

Computer generated illustration of a brain

With Gen AI already on the slippery slope towards the trough of disillusionment, it is reasonable to expect a wave of anecdotes where A.I. projects fail spectacularly.

I can't bemoan the favorable winds brought by the all-encompassing hype — we have spent many years trying to evangelize leaders on the potentials of those technologies.

But it’s also true that I will be nodding in agreement with those shouting bulls#!t at the most egregious cases of A.I. overpromising.

We already see this happening, as the overuse of “AI” seems to have a negative impact on consumer perception. The now-infamous case of the $1 Chevrolet has become a frequent target of mockery.

Gartner 2024 Hype Cycle for Emerging Technologies

The question then becomes, what do we do, and how can we frame an A.I. that actually benefits organizations. Here are suggestions that closely match a core principle at Agilytic — namely, to align with your strategic priorities.

Here are three examples of areas where we suggest refocusing the discussion instead of tech for tech’s sake.

1. Market intelligence - no more excuses

One of the most interesting changes in the past 18 months has been how modern tools allow organizations to internalize and automate a lot more of their market intelligence.

I used to be a market analyst, so I know how costly this can be for companies, and how seemingly reliant on third party experts this activity can be.

Market intelligence, as the name implies, is all the valuable information about the market that an organization can gather and leverage to their benefits.

I posit that smart marketing departments will gain an edge on the competition by building market intelligence assets about their customers, but also their competitors, social trends, and the macro-economic context.

For example, we can now very easily automate competitive watch across multiple channels, consolidate insights, push alerts, etc…

And it’s not only B2C. B2B companies might have an easier path to building better market intelligence, thanks to more structured information about company financials, commodity prices, etc.

2. Operational intelligence - pushing optimization boundaries

We didn’t wait for the latest A.I. technology to develop optimization models.

In operational intelligence, the obvious next step will be to improve the operations models already running. Document automation can be more precise and faster. Optimizations under constraints can handle more complexity, all the while providing clearer explanations to human agents.

User interfaces are easier than ever to test and deploy. Yes, that includes chatbots, but not exclusively. Don’t underestimate traditional, therefore familiar, interface paradigms — nobody puts search box in the corner!

There is also a great opportunity for push communications, be they machine-to-machine (e.g., an ERP pushing a CRM with all required information to the salesperson), or machine-to-human (e.g., a WhatsApp or Teams notification).

Going from 90 to 99 is less spectacular than going from 0 to 90. But some challenges are worth the extra effort as they will yield returns in speed, quality, and cost.

3. Financial intelligence - lowering the barrier to entry

For CFOs, we are also optimistic, especially for those in mid-size organizations. The technologies become increasingly plug-and-play and pay-per-use. This means solutions previously reserved to large corporations are becoming ROI-positive for smaller stakes.

Consider

  • improving invoice to cash efficiency

  • anticipating fraud

  • automating reporting and alert systems

Those are more within reach than ever.

But what about disruptive ideas?

Here’s the thing with disruptive ideas: they are rare, it’s a first mover game, and when people realize it’s disruptive, it’s already become a fast follower play.

We gladly workshop with our partners on groundbreaking opportunities. The key requirement is having an open mind across the room. Nothing will drag innovation more than a “problem finding” attitude. Realism and pragmatism have their place of course, but not in the first stage of your ideation process.

No matter what, the mistake to avoid is to wait for a big bang before you start building impactful solutions right now.

Wrapping up

As much as we like innovative technology, Agilytic’s north star has always — and will always will be: using technology to serve strategic goals.

Framing AI as an enhancer of current disciplines will ensure AI remains a means and doesn’t become an end.

👉 Curious about what Agilytic can do for you? Get in touch

Previous
Previous

From models to pipelines: the journey of a data engineer

Next
Next

Attribution Modeling in a Cookieless World