Your tech vendors’ apps are a great place to start with AI—just be sure you can connect them.
Nearly every leading enterprise application company now offers AI agents, AI-enhanced applications, or other AI-powered features as part of their product offerings. Many are very good and offer practical, quick ways to use AI to drive incremental value. However, the value and focus are often limited to a specific business function or product set. Apps built for HR, supply chain management, finance, or sales will connect only to that function’s core data set by default. As a result, if you use them in silos, you’ll be leaving value on the table.
These siloed apps remind us of the initial migration to the cloud, when most solutions promised to eliminate silos of information and deliver seamless integrations. But while many of the early cloud-era solutions caused silos, these new AI-powered tools don’t have to. By sharing data between solutions, you can gain the quick value these apps offer while preparing your company for long-term success. The key to enterprise profitability from AI is to build a larger data foundation—and with your tools working together, they can be a big part of your overall AI strategy across functions, vendors, and departments.
Avoid AI silos in your organization by integrating apps.
If you’re starting your AI journey with a vendor-provided app, how can you avoid silos? Ultimately, knowing is half the battle. Plan to treat each app you use as part of a bigger system, and you’ll already be better off. As you implement these apps, consider these steps:
Adopt a “plus 1” policy for AI pilots.
No app should be standalone. For instance, if your finance organization is considering an Anaplan AI pilot, how can that pilot pull in data from beyond finance? Consider what’s available within your existing Snowflake or Databricks infrastructure, and bring that data into the app. And remember, the greater the data set, the greater the AI results. Ensure you’re not relying on a single data stack, and feed your AI tools with as much data as possible. You’ll both be able to leverage additional organizational data that will help provide necessary context for the models and set a solid data foundation for quicker ROI across your technology investments.
Widen your data aperture.
On a related note, when working with AI, remember all the various types of data you can use. Most companies arm their customer-facing agents with structured data such as sales history, account firmographics, contact info, opportunities, and marketing engagement. However, AI is often excellent at making sense of unstructured data. Sources like call notes, contract data, social media engagement data, annual financial reports, and company commentary can all provide a broader lens to inform agents and drive the best outcomes.
Treat every AI project like a finance project.
Consider recent Gartner survey data, highlighting how CFOs increasingly oversee their companies’ data and AI strategies. Finance is probably already touching most data flowing through your organization, so they’re the natural point of contact for a connected project like an AI pilot. Bring in the office of the CFO early, and connect your tool to key finance applications, and you’ll avoid running AI pilots in a vacuum.
Avoid agent sprawl.
Agents have found homes in numerous business functions because of their ability to work effectively with minimal supervision. But just because they’re independent doesn’t mean they don’t need to be managed. When agents are disconnected and not centrally governed, they often perform poorly. In extreme cases, this can introduce potential security risks, such as data security, compliance, and even safety, depending on the agents’ focus. Fortunately, there’s a middle ground between micromanaging your agents and letting them go unmanaged. Companies like Boomi have introduced agent governance tools, giving companies a central clearinghouse for agent design, distribution, and safety within their teams. By starting early with agent governance, companies can avoid running into much bigger problems down the line.
Deploy agents with more comprehensive workflows.
While an individual agent may excel at a single task, combining agents into a complete workflow will create greater value. For example, in Spaulding Ridge’s AR Collections applications, numerous agents are embedded in the full workflow of collecting payments. One agent recognizes customer payment patterns and alerts of potential risks. Another agent notifies and reminds customers of payments due, providing context-aware insights to reduce collection risk. Yet another agent advises AR collections representatives on prioritized actions for effective collection. Having these multiple agents working across a business process increases value and creates an environment where humans and agents can engage at scale.
Make your AI apps work harder for you.
Vendor-specific AI apps are a great place to start with AI and they should also play a major role in your broader AI strategy once you break down silos and connect your data across the enterprise. By broadening your data sets, aligning with finance, avoiding agent sprawl, and deploying AI as part of comprehensive workflows, you can transform AI experiments into company-wide growth drivers.
At Spaulding Ridge, we can help you with every step of the process, from getting started with individual apps, to establishing an AI strategy, to building the data infrastructure that will drive the enterprise of tomorrow. If you’re ready to get real value out of your AI investments and build a foundation for long-term success, we’d love to partner with you. Come meet us at Dreamforce or schedule time for a conversation!