Customers like to have a clear understanding of what’s going on. MSPs hold more than enough data to let customers know what is going on with their services. However, they struggle to transform this data into a useful form that really helps the customer. Streams of data are too arcane for most customers; they need something that paints a clear picture.
Invoices tend to be devoid of anything useful – just a list of services against what the customer owes. Portals can be useful but tend to be technical and only beneficial to administrators in the customer’s world. Intelligence has been built into such portals. Red markings for problems, green for things being “OK,” yellow for potential issues, and so on. But this still doesn’t say much about how well the services support the business or how the customer can change what they are doing to make the service better fit their requirements.
Here comes artificial intelligence to the rescue (again). AI can also examine a customer’s data. It provides insights into what has happened, what is happening, and what will likely to happen in the future.
So far, so good. Except all that this did is narrow down the masses of data into not so much a mass of data. What does this really mean to a customer?
Data must be actionable to be useful
It is easy to scrape data and send something like a table that says that so many users used so many resources, resulting in this much cost. Value may be added by pointing out which users were the heaviest users during the period, allowing the customer to investigate this to figure out why this is the case.
How about sending a report that states how the customer’s usage of the service is growing month over month and that it looks like they will need to renegotiate their contract in so many months? Analyze the data and say that this may not be necessary if they cut out some users who haven’t used the service for a certain period, that there have been big resource hogs that don’t fit in with general usage, and so on.
How about analyzing workflows and advising the customer on how to optimize them? How about looking at the security of the content involved and providing insights into how changing things could enable a company to be better aligned with, say, ISO27001/2, the EU’s GDPR, HIPAA, or other areas that have a direct impact on the customer’s actual business?
Such reports could examine how data could be better formatted to allow greater straight-through processing both internally within the organization across its multiple applications and services and outside the organization in its dealings with suppliers and customers.
An MSP could even provide advice on how to optimize service usage to bring down costs. Counterintuitive, but helps the customer acknowledge how you are collaborating with them rather than just trying to fleece them.
Maybe comparing the customer’s usage against an anonymized analysis of other customers’ usage can tease out greater actionable insights.
Generative AI is transforming reports
Generative AI can do a lot of this. Based on the output from data analysis carried out by ordinary AI, it can create readable reports that customers can really understand and act on.
It can also provide real-time reports. A quick report stating that the customer is likely to require extra resources for a specific workload within the next hour, day, or week can inform the customer as to why this is likely to happen. The report can then contain links that enable the customer to take the right action for them. Perhaps they maintain resources as they are and take a hit on performance, offload lower-priority jobs to free up resources for the specific workload, apply more resources to enable existing jobs to continue working alongside the higher-priority workload, etc.
Generative AI is becoming a slightly dangerous game-changer. This is because it struggles to escape its bay years, moving from being a toddler to walking and then trying to be a Usain Bolt world champion within its field. Careful usage of both straightforward AI and generative AI can, however, create a different business model that works for MSPs and their customers alike.
Organizations may want to use generative AI to send equivalently useful reports to their own customers. Along with this, links that enable them to leverage sales through providing meaningful and persuasive messaging.
By being the guinea pig, the MSP can identify the best ways to implement generative AI to achieve such aims. Once proven to generate marginal incremental revenues through optimizing the MSP/customer relationship, leveraging the capabilities as a service on its own should be easy.
Just please do not try it the other way around. Using customers as guinea pigs for such a service is highly likely to blow up in an MSP’s face. As the saying goes, please eat your own dog food first.
Photo: Sky Motion / Shutterstock
