AI – What are the next steps, MVNOs?

AI is all the range in technology reporting today but according to McKinsey, outside of companies offering AI front end services, AI has not translated into P & L for many companies that have embraced AI in principle. And amazingly, for such a new(ish) technology, there is an extremely high level of awareness and intent at CEO / C-Suit levels among the companies that were surveyed by Mckinsey in June 2025. The same report has delineated the main opportunity areas in the AI revolution (By far the most succinct that I have come across so far) as per the diagram below, reproduced from that report.

Figure 1: Source Mckinsey in June 2025

This article tries to analyse possibilities in the verticals suggested above and see how ready the industry is in taking advantage of the opportunities. Specific to the communications sector is missing the fulfillment vertical (Provisioning, activation, business workflow etc).

Let us look at some of the the theme areas separately and try to panel beat them into the MVNO industry … .shall we?

R and D

Have you ever spent sleepless nights trying to design the next product that will knock off your competitors? For traditional MNVOs, the tools at your disposal are limited to talk minutes, Mb of data and SMSs. That is it. However complex your product is, it is simply a reconfiguration of these parameters in addition to optimising other factors such as time of day, which are set to be counter cyclical.

Generative AI and more easily Agentic AI is capable of designing your product and taking a survey to benchmark your products against competitors in your chosen market. Of course, not everyone can be able to just type the prompts on ChatGPT and get a product out. For starters, ChatGPT has not been trained with the requisite data. Most of that training data remains in private (and often highly regulated) hands. Moreover, you would want the AI model to be integrated with our digital platform in order to make the process seamless.

There are some providers who are ready to offer these services in the market, including the organisation that I work for.

Procurement

I am of the opinion that this is probably the lowest hanging fruit and can be done without advanced tools. One possible  use case is shortlisting of service providers.  Most of this information is publicly available making it likely that generic ai models have been adequately trained on them

Supply Chain

I would like to translate this to a wider logistics management domain in order to create a use case that resonates with most MVNOs. The most important use case would be demand forecasting so as to make orders for inventory just in time. In the MVNO industry, the operators tend to have very high value items like handsets, laptops and routers. It is thus imperative that the order decision is made on time. Order too early and you tie up a lot of capital unnecessarily. Order too late and you have some interesting calls coming through to your customer care, and an irate CMO who is not meeting the targets.

This is where a well trained model will be able to predict when you need to re-order your stock items while optimizing your cash flows. Again, the trick is to ensure that you model trains within the inventory management system, preferably as part of your integrated digital platform.

Sales and Marketing

In sales and marketing, customer insights stands out as the area that renders itself easily to being integrated into an AI tool to analyse the vast quantities of customer data that the MVNOs and MNOs are capable of generating. Again, it requires that your reporting engine is integrated with your digital platform which in turn should be AI enabled.

This is probably one area that can take advantage of AI with the lowest amount of risk, and possibly cost. It  is also one of the areas that can have a significant impact on the top line if well implemented.

Customer Care

This is probably the area that is most mature for an integrated approach. MNVO adjacent industries have been using AI enabled chatbots for their customer care for a while now. And for margin strapped MVNOs, this makes perfect sense in order to reduce costs. Just like other areas, the trick lies in integration between the AI chatbot and the customer management system.  Secondly, the training data is in almost all cases held privately, but as an MVNO, you probably have this somewhere.

Conclusions

I hope that these few possible use cases have demonstrated the possibly endless possibilities that exist for use of AI technology for the industry. Sadly, if the industry players do not keep up, disruptors will come up and eat their lunch with more compelling offers at much lower delivery price, hence margins.

Get cracking!