While vendors may be hyping up generative AI as the next big thing, enterprise buyers are moving cautiously. This year has seen a flurry of announcements from companies eager to leverage this technology, but behind the scenes, CIOs have been taking their time to evaluate and implement these new solutions.
The ROI Dilemma
Not all enterprises are looking to spend more money on new technologies. Many are actually trying to cut back on spending or at least maintain their current budgets. However, when technology does enable companies to operate more efficiently and do more with less, it’s hard for them to ignore the potential benefits.
Generative AI certainly has the potential to improve efficiency and productivity, but it also comes with its own costs. Whether it’s a higher cost for features in a SaaS product or the price of hitting a large language model API when building software internally, there are expenses associated with implementing this technology.
The Importance of ROI
It’s essential for companies to understand if they’re getting a return on their investment (ROI) from generative AI. A July Morgan Stanley survey of large company CIOs found that many were proceeding cautiously, with 56% of respondents reporting that they had not yet invested in this technology.
Only 14% of respondents said they planned to invest in generative AI within the next six months, while 30% said they would do so within a year. This suggests that companies are taking their time to evaluate the potential benefits and costs before making a decision.
Case Studies
Several companies have already begun exploring the potential of generative AI. Box, for example, has unveiled a unique AI pricing plan designed to account for the high cost of running large language models (LLMs).
Sharon Mandell, CIO at Juniper Networks, says that her company is participating in an initial pilot with Microsoft around Copilot for Office 365. While she’s heard anecdotal feedback from users who love it and those who are less impressed, measuring increased productivity remains a challenge.
The Challenges of Measuring ROI
Sharon Mandell highlights one of the main challenges facing companies trying to measure the effectiveness of generative AI: "You don’t have data on people’s level of productivity. So no matter what, you’re using somewhat anecdotal information until you get really good at understanding these dashboards from Microsoft showing you how people are using it."
This makes it difficult for companies to determine if they’re getting a genuine ROI from their investment in generative AI.
Conclusion
While vendors may be touting the benefits of generative AI, enterprise buyers are taking a cautious approach. Companies need to carefully evaluate the potential costs and benefits before making an investment, and they should be aware that measuring ROI can be challenging.
As companies hear about the potential power of generative AI, it’s only natural that they would want to learn more about it and put it to work to help their organizations run more efficiently. However, executives are right to be somewhat cautious, recognizing that these are still early days and they have to learn through experimentation if this is truly transformative technology.
Recommendations
- Conduct a thorough evaluation: Companies should take the time to evaluate the potential benefits and costs of generative AI before making an investment.
- Set clear goals and expectations: Companies should set clear goals for what they hope to achieve with generative AI, and establish metrics for measuring success.
- Monitor and measure ROI: Companies should regularly monitor and measure their ROI from generative AI to ensure that it’s delivering the expected benefits.
By taking a cautious and informed approach, companies can minimize the risks associated with implementing generative AI and maximize its potential benefits.
