Omniadigital's Podcast
The Omniadigital Podcast explores how organisations optimise their current digital landscape to maximise its return, transform to a new digital technology or maturity with expert advice, and solve one-off problems utilising people, process and technology to enable the data across the full digital landscape.
Hosted by Lucy Lynch, the podcast features practical, experience-led conversations with leaders and practitioners working across data, analytics, AI, technology, and business transformation. Each episode focuses on how organisations move from ambition to execution, and from activity to measurable value.
We examine what it really takes to optimise existing systems and capabilities, transform operating models and ways of working, and solve complex business problems using connected intelligence. The discussion goes beyond tools and trends to cover governance, culture, leadership, risk, and delivery, the factors that determine whether digital initiatives succeed or fail.
This podcast is for senior leaders, transformation teams, and professionals who want clarity, not hype, and outcomes that stand up to scrutiny.
Omniadigital's Podcast
Rewriting the Integration Playbook
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
This episode is a deep dive into the world of enterprise integration with Bejoy Thomas, founder of NJC Labs, a Salesforce Summit Partner and global Innovation Award winner. You explore Bejoy’s 20‑year journey from writing “handwritten code to connect systems” to leading a 100‑person integration consultancy operating across the UK, US, Australia and India.
There are people in the integration world who follow the playbook, and then there are people who rewrite it. And our guest today, B Joy Tom Thomas, is firmly in the second category. So welcome to the Omni Digital Podcast, and I'm Lucy Lynch, Chief of Staff. So welcome, B Joy.
SPEAKER_01Thank you, Lucy. Thank you so much for uh for the invite and uh great to be with you in this podcast.
SPEAKER_00Yeah, no problem. So I was looking back at your sort of history. You've got more than 20 years under your belt as an integration architect, and you've kind of like spent time in your career solving the kinds of enterprise challenges that most teams sort of quietly avoid. You've worked across really complex, high-stakes environments, built solutions that kept organisations moving, and developed a reputation for delivering integration that's not just functional, but elegant, efficient, and genuinely transformative. So this experience leads me to how you founded NJC Labs, which you're now a Salesforce Summit partner. And I was just wanted to sort of pick up, you know, your sort of accolade as your Salesforce Partner Innovation Awards in the MuleSauce category, which is a global recognition that puts you in the sort of top tier of integration partners worldwide. So today we're sort of talking about that journey, how you got from architecture to founder and from founder to award-winning partner, and what it really takes to build integration capability that accelerates delivery rather than slowing it down. So BJ, welcome. I've been really looking forward to this conversation. It was so awesome to meet you at the partner dinner at Data Decoded. And you know, I just kind of was love wondering, you know, one of those sort of questions that you that you think about when you're, you know, just kind of like having a walk. What were the sort of pivotal sort of decisions or beliefs that shaped NJC Labs into what it is today? And what advice would you give to other founders?
SPEAKER_01Great, yeah. Thank you, Lucy. That that was a very good uh intro than usually what I say to customers in my pre pre-sales activities. But yeah, that was impressive, and thank you so much. And thank you for the dinner as well, right? It was uh it was great meeting you uh and all the founders, and it was a it was a great event. So yeah, coming back to your question, yes, uh I was always in integration, knowingly or unknowingly. When I started my career at that time, I didn't know I was doing integration, but uh it was writing handwritten code uh to connect systems and applications. I was doing that for four or five years uh in the telecom industry, connecting various applications by writing code and then started using middleware platforms to connect enterprise applications and been doing that until 2015 using various platforms, Oracle and a few other uh ETL platforms and so on so on. Then 2015-16 onwards, I started working with uh with MuleSoft. Uh, it is one of the best integration platforms out there in the market. And I I started I became a trainer, and then that's what my I started the journey towards a founder. So I started going to customers and partners of Salesforce and MuleSoft to train them, and then in various um uh sessions, training sessions, customers started asking me how do we do this? We have a challenge with connecting these two applications, how do we do this in MuleSoft? And uh at that time it was uh it was a quite niche market, especially in the MuleSoft ecosystem. So I found okay, so there are a lot of customers who need support in uh in enterprise integration. So I founded the company in 2018-19 time frame, and we started in the UK, so uh headquartered in the UK and JC is headquartered in the UK with offices in US, Australia, and India. Now we are a team of around 100 people, uh amazing uh team, and with the the help and support of the team we and the work which we have done for customers globally, we got the innovation award, and uh that award was for one of the complex integration projects, not very common type of integration scenario the customer had. And uh we did it less than a month, that and their project, and that's the reason MuleSoft uh gave us this global innovation award. It was very innovative in the way we delivered that project here.
SPEAKER_00So wow, that's amazing! And what advice would you give to other founders? Because I know there's always that thing, you know, the thing that kind of keeps is it the same thing that keeps you up at night, that gets you up in the morning? You know, you have different kinds of headaches, right? And you're kind of juggling a whole sort of myriad of uh of different problems that you're trying to contain, get on with your normal life, be a regular guy, you know, and all of that stuff alongside. So I just wonder if you've got any advice for like I don't know, not you know, things that you might have happened upon or like mistakes that you might have found or other mistakes you've seen, and other founders have just thought you might want to impart some of that and like um knowledge.
SPEAKER_01Yeah, um definitely, right? Uh yeah, I think I missed that uh question earlier. So I have done a few other things as well in my past past career before starting NJC labs, and uh that was nothing related to what I knew. So integration is something we I always know, and that was something which I was living on, right? Bret and better for for me for years. Instead of doing something in integration or enterprise integration space, I did a lot of other stuff, and then I found that that wasn't really successful, even though that gave me a lot of uh learnings. But I found that if you are starting something new and if you want to found and then scale to scale, then probably you should start something which you'd absolutely clear of what you know you are doing and how you have got a vision on how to scale that up, right? So maybe that was something which I uh the lesson which I learned. I was dependent on many people when I started other ventures, and when there is a lot of dependency, and if you have haven't got that investment to meet that dependency, then you probably cannot scale. Whereas in this one, I knew integration, I knew MuleSoft. I used to be a trainer, so I can train and scale the team. So that I thought it was quite useful in starting this.
SPEAKER_00Brilliant, yeah. So you found your your niche, right? You had that real deep knowledge that you could you could you wouldn't have to rely on on other people, right? Is that you had that internally and you know, in a small team, which you could then scale up, you could deliver the training yourself, you could you could kind of answer all those like multiple questions that you might get, right? Because you just know it inside out. And I think that's that's also really good information because often people have an idea, oh great, you know, I'll start this company, and um there's a real demand for it, but then you have to think like, well, who's who's gonna deliver it when hey, I'm not here, or if I'm sick or if something goes wrong, depending on one customer as well. Like you've got to look at all those risks, right, as a founder as well. Like, where's the cash flow coming in? Look at the pipeline for the future, you know. Um, and and also people, they're our biggest asset, right? You know, forget tools and technology for a second, like if we think about people, but they're also quite tricky as well to manage and and put together in teams as well. You've also got to make sure that you know there's a good culture and there's a good sort of like skills match as well, that people can continue developing as well, because that whole thing about um attrition and and and uh staff retention as well, really important when you build it when you're building something and you want to scale it, you don't want all that knowledge and experience to just walk out the door when the next kind of better offer in inverter commerce comes along, right? Absolutely. So I think they're also things that maybe founders don't always think about in the in the first, you know, kind of like year, you know, what if the risks? And I always love looking at risk. I don't I don't know why, but I'm quite a risk-adverse person, maybe. So I always like to have like a hundred different scenarios going on in my head, like, oh what about if this happens or or that? What how can we mitigate that risk? So I think that's also quite important. And and maybe you don't think of these things 20 years ago, right? As well. I don't know, but I I definitely know I didn't then.
SPEAKER_01Yeah, yeah, absolutely. I think that's uh that's an important point, right? The team uh which you have got, and uh, and uh after all, we are in a services business, we're helping customers, and our team is the most important aspect of this business. And we as a as a brand is all about our team. Um and I'm very proud to say that we have got an amazing team, and many people or the entire team has been with us for years now, and it's mainly because of um of the flexibility which we are giving to the team, the knowledge we are giving to the team. And I think everyone appreciates the amount of training or effort which we are putting to train and develop our team and uh and also the flexibility. And uh, we have a lot of women uh at some point. I don't know if it maybe it's still more than 50 percentage of our global team uh were women, and uh and and then the certified uh engineers, integration engineers maybe at some point, even still it's more than 50 percentage is women. So we uh give a lot of uh importance to that, and uh we also give a lot of flexibility as well, maybe working from home or remote work, and then we have uh a big team in India as well, so we we meet once or twice in a in a year. Uh, maybe we bring everyone from all across India in all across India, and then we meet and stay for one or two nights and then collaborate. And so, yeah, so the team is important, and we do a lot of uh activities to make sure that collaboration is happening, and uh in all other countries we we do that, even though it's a remote uh only team.
SPEAKER_00That's amazing. I'm I'm really, really pleased to hear that. I mean, you spent sort of like you know, over sort of two decades inside enterprise and integration. What are the biggest challenges that you still see organizations struggling with, VJ?
SPEAKER_01Yeah, I think um in the integration world, we always say um data and application challenges, right? Um so organizations typically, especially enterprise organizations, have got um hundreds of applications by knowing or not knowing, right? So they eventually we'll have a lot of the core systems, systems of record we we call the main systems like ERPs, the SAPs, the CRMs, and payroll and uh all these systems. And on top of that, there could be every department will have a lot of applications as well. I have seen customers and the marketing team uses maybe three or four different types of applications for email marketing. And with the um we we had this SaaS as a kind of the main thing for everybody, everybody could just subscribe a service, uh, SaaS service, and through that, a lot of uh enterprises, the different departments they started subscribing for services. Maybe one department will have multiple applications for the same purpose itself, and uh for large enterprise, they grew through acquisitions. So, as part of acquisitions, they will have different types of ERP systems alone, right? So, multiple ERP, two, three, four ERP systems, different CRM systems, and all of that. So, the complexity is how do we manage all of this? And uh, that's where enterprise integrations become uh very key, and there are different ways to do the integration as well. But the the challenges is the number of applications, the data in in silos, maybe these applications might not be connected very well. Sometimes it could be a legacy system, there is no proper interface to connect to. So there are different challenges which organization sees, and um, especially the recent time with the AI, everybody wants to implement AI and agentic use cases, right? And that's the the new challenge which every organization is facing, where um uh customers do not have accurate data available most of the time. So a lot of majority of our customers who we are speaking to, the major challenge is data. So we need to have a clean data uh in order to do any sort of agentic use case and access to clean data as well. So recently I've been uh working with a customer, uh, a retail segment milk producer, one of the biggest in the Middle East region. And in order for them to identify the the forecasting for next month, like how how many milk cans I can uh sell to a particular region, there are a lot of parameters which are important to make that decision. Not just the the sales uh history of last few years, but then there are weather, what's the weather forecast for maybe next few weeks, what's the the size of the shelf uh in a particular retailer at a particular location, and how many such retail shops are available, the route which the van is going, for example, right? So there are a lot of parameters, and many of this maybe customers do not have access to. And the integration scenarios are numerous, especially in this agent tech enterprise. So that is one side of from a business outcome perspective, there are quite a lot of things to do, and uh the productivity improvement is another area from an AI use cases perspective, right? So so looking at challenges, yes, there are too many challenges to overcome, and that that also creates opportunity for us as an organization, and especially since we are in the data and integration space, we all are, that creates a massive opportunity to help solve these problems for enterprises. Hope I answered your question. I was saying too many.
SPEAKER_00Yeah, no, no, no, that's good. Because my next my next one was about you know, what do you do differently that clients feel that immediately sort of feel? And I can already tell from the way you've described that, I would say, you know, it's your ability to ask better questions, right? To really think outside of, you know, what is what is this business problem, but then you know, what are all the multi-layered, like multifaceted other questions that you could ask that will expedite, you know, an even more effective solution. And it was interesting what you were saying um about cattle, actually, and I wasn't expecting you to say that. And then I thought, oh, that's great. We kind of got to at least 15 minutes before we use the words AI, which was great. So that's great. But I do want to ask you about agentic AI, and so like it's a question really for also for you know customers listening out there who are experimenting in this, what would be good things for them to watch out for in their implementation? Because I know obviously you have to have, you know, like you were saying, you have to have, you know, your good clean data, you need to have your guardrails up, and you know, a good sort of implementation flow. But tell me, tell me some things that they could potentially be watching out for so that they don't make some unholy mistakes that would be terrible.
SPEAKER_01See, I think uh the challenge, right? Nobody's uh sure at this moment what's the right thing to do, right? And uh so when nobody's we as experts also are not sure to be very transparent, right? Because we have too many options and even customers uh will have a lot of options from different product vendors they are dealing with. So when they have too many options, I have seen customers who go for a big band kind of an implementation. So they identify all the automation and agent deck use cases within the organization, they select a tool and then go and implement uh that tool at the enterprise-wide use cases, and that's a it's a massive program, like multi-million, like hundreds of millions they need to spend in order to achieve that. And then often that probably won't give them the result back. So that I think it's not the right thing to do. So maybe start small, right? So identify a few use cases instead of hundreds of your hundreds of use cases, start with maybe two, three, or five use cases, and then identify how agent implementation can help to their business, mainly from a profitability or a bit perspective, right? So, how is it impacting the business? Is that a use case actually needed rather than just for the sake of implementing agent deck use cases? And then what's the what the what are the right uh products to be used? Because every product now has got an agentic capability most of the time. So are we using the right product? And um so usually the first type of uh first few use cases would be a FAQ and knowledge base with these are the agentic use cases, which and it's quite easy to implement, right? So provide a FAQ and knowledge base to all your employees or all your customers, because you just need to feed in your data, your information, and then people can you ask questions to agents and get information out of it. It's a straightforward use case. So that probably is the first thing to do, and probably there might not be any failure in those two use cases, right? Then as soon as it comes into the transactional use case, so say I want to create this in it in a system, I want to make an order for something, or I need to have a purchase order request needs to be placed within the uh finance uh finance department, some something of those things that involves multiple systems to be connected, so that's where you need to have all these right strategies in place. What's the agent tech platform you're building? How it can be scaled, how it can be democratized, right? If you build an agentic platform and if your business users or internal consumers are not using the platform, then it's not use. So, how do you democratize? How do you make sure these agents are working correctly? How do you avoid hallucination when especially when you have got hundreds of agents at some point, may not be now, maybe in next six months, one year time. How do you make sure all these agents interact with the agent? You have the right governance in place, you have the right security in place. So we see a lot of news in the market saying that there is a security challenges with the agent tech interfaces, so data and other compliance violations on going on everywhere, right? So you need to have all these in place. Governance and security, I would say, is quite important, and the platform we choose is right uh is important instead of going for a big bang implementation where maybe start small, pick a low hanging fruit, and then try to implement that. So those are the some things which I would uh I would uh suggest to business association.
SPEAKER_00And and I think that's sort of spoken like someone that's kind of like walk the walk, right? Because you know, often I'll I'll speak a lot with you know chief data officers and you know people at the C-suite level, and it's always in hindsight they go, Oh, I really wish in my first 90 days that I've done like a smaller project as a smaller case study rather than just try and scoop everything up and go, okay, we'll do this big ta-da moment, and it'll all be amazing, and you know, we'll get that instant sort of gratification and that that ROI will come back, you know, in buckets. But actually, organisations don't work like that, right? They're very slow and clunky and like you said, like siloed and nothing really sort of fits together or talks to each other, communicates, uh, teams of disparate, remote, hybrid, all that kind of stuff. So a lot of work has to go into place first, right, as well, and a lot of good listening skills to to understand what is going on under underhood. And I think you know, I'm just thinking, I'm gonna be talking about C-suite there, and I'm just thinking that integration work often sits at that intersection of pressure, ambiguity, and deep complexity. What mindset or sort of leadership approach has helped you thrive in this space? Because I know that I've met you, and not obviously everyone that's listening to the podcast has met you in person, and you're a very chilled out, lovely, charming guy, and I'm sure you have pressure held song, and you must deal with a lot of pressure and ambiguity because I know that I certainly do. So, what what sort of leadership approach has helped you sort of thrive in this space? Because you are very relaxed and just very easy to talk to. So you you really give off this uh impression, you know, and I'm sure it's real, of of of being very relaxed in this environment. But how do you get to to that state, Joy? Because I'm always like bullying the headlights, going, Oh my god, Matt says I don't have a poker face, like I just shows.
SPEAKER_01Yeah, I think uh that's some feedback which I get from everyone from my team. I'm so relaxed and chilled. Then maybe I don't know, see, with with respect to integration and customers, right? So the one approach which I always uh follow is I'm a good listener, I think that's what um it's quite important for any any sort of business, but especially integration. So a good integrator needs to have a good listening skill, try to understand what a stakeholder is saying, and then able to translate that information back to another stakeholder in the right way, right? So that translation, so listening and translation, and not just from a technology aspect or the product which we are using or platform which we are using, it's all also about the stakeholder management as well. So multiple stakeholders are also in always involved in integration. So we are we are at the middle of an organization, and if integration works, nobody's even aware there is a team exist in the organization. If it doesn't work, then everybody knows okay, integration is failing. And this is so it's a big escalation to the to the organizations, right? So it's important that we understand what each stakeholders are looking for. Multiple parties are always uh involved in integration. Um, and since we are we are connecting all these different systems, we mostly will have sales sources, CRM, most of organizations or ERP, SAP, or Oracle based ERP. So all of these experts from these different areas we need to talk to and understand what they are looking for. And say, for example, a conduct in sales source might be treated completely different as a same uh entity in an ERP system. So the ERP person might be saying some. Something different to a sales force case. So, how do we relate all of this? How do we translate? That's important. So, maybe that listening skill probably is helping helping me a lot in creating that atmosphere and asking right questions to right stakeholders, setting the expectations very clearly. One thing I always do is um maybe that's also helping me with it to be not under tremendous pressure. I set an expectation that this is what you are going to get. And if it is not possible to deliver, I'll say a friend that it's it's too difficult for us to deliver this program. So some sometimes we get requests like a you need to implement uh 50 or 100 uh integrations in three months, and I will say, Okay, it's not possible for six months, right? So it's it's impossible um to we as a team cannot do, right? So maybe you need to look for other size. So I'll set the expectation very clearly. And sometimes we might lose the project, but we have we have done a lot of projects where we help to solve the challenges customers had. Uh, a lot of rescue missions, I would say, right? So uh yeah, and it's mostly setting that expectation always helped me in my experience.
SPEAKER_00Yeah, I think that's great, and I think it's that whole thing about transparency, right? And and and and I know that Matt also, you know, uh shares that. So talking about Matt uh small. So I know that obviously we're fully fledged, you know, business partners. And in your world, what does the why does this partnership make sense for for you and where do you see the biggest opportunities that we could create value together for for customers in the future?
SPEAKER_01So um we being coming from an integration space, right? So we we are originally still uh integration experts, and then in order and slowly moving in towards an agentic enterprise vision to meet to help customers to achieve agentic enterprise. And one thing which is uh lacking in this vision to be achieved is data. So we need to have we are an engineering team, we are not naturally can adapt the skills of uh data, data engineering, but there's quite a lot of elements to data when we say data, like there is data strategy, there is data lineage, there is data governance. So it's important to get that piece correct. And that's where I think we as an organization can collaborate with you and your partners in that side. So we integration and data is absolutely the foundation for implementing an agentic enterprise for the for the future. So having this collaboration is really, really important for our customers, especially, right? And uh, I think that's where this can this can scale quickly, especially in the next few months and years.
SPEAKER_00Yeah, yeah, absolutely. And if you could give one piece of advice to any organization trying to modernize its digital estate, what would it be, B Joy? Maybe hire us. I would love that. That's it, okay. That's it. Well, folks, you've hit it, you've heard it here first. That's a wrap. You need to hire B Joy right now. That's your call to action. Thank you, B Joy. I think that was really great insights, a lot of laughter, which I always like, and a masterclass in sort of what good integration actually looks like. And you know, winning your you know, Salesforce Partner Innovation Award is like absolutely fantastic. And it says everything about the standard that you and your team operate at, which is you know why we we love working with you. And I think that's exactly why this partnership with Omni Digital and and NJC Labs like works really well. So if today's conversation sparks something for you, for our listeners, or any organisation, then please reach out. This is the work that we love to do, and we look forward to seeing you on the next episode. So thank you so much, BJoy, for being our guest today. Thank you.