FY26 closed 3.9% / 17.2% EBIT vs raised mid-year 4-4.5%.
- Isolation two sap two — answer hedged.
- Software deal spillover return — answer hedged.
- Whether ai deflation materially — answer hedged.
These two SAP programs and the two Telecom accounts, are they overlapping or separate events? What gives confidence these might be isolated events with no underlying causality at a vertical or geography level? And if you add back 50 basis points impact, you are essentially talking 2% to 5% growth in services, which might be softer than anticipated - anything to explain that delta?
Two Telecom clients and the two other clients are completely different. The two other clients where there is a half-a-percent decrease in FY'27 guidance, one is a large Manufacturing client, other one is a Retail client. So there is no overlap in these clients. What gives us confidence? This is restricted. In the lower end of our guidance, we are baking in that the softness continues. Given it is a well-known fact that 2% to 3% deflation happens, barring this, getting to 2% to 5% is a reasonable growth in the given environment and in so much of uncertainty. Outside of these two data points that we shared, the business continues to perform at the same pace. This of course does not include any acquisitions and they will get closed sometime during the year and we will call it out separately.
On the product business - you mentioned a spillover of last 14 days in this quarter. For the full year, our product business is down 4% YoY. Is the revenue missed this quarter a spillover we can expect back in Q1? And could the next year performance for product business look better?
Yes, I think whether these deals will come back depends on many variables. So it's too early to say and the timing of closure is unpredictable. You would have noticed this would be the first time in recent years that the total revenue growth is half a percent below the Services revenue growth. So, at the lower end, you can compute what our assumptions for the Software business have been.
You talked about 3% to 5% deflation, but that's kind of what we see anyway, even before AI. In renewal deals, we were seeing 10% to 15% cut in pricing over the life of the deal, which would translate to 3% to 5%. So with AI there is nothing significantly different - would that be right?
No, I think we were always careful that this is the incremental impact. I mean, considering the traditional productivity what we normally commit - we should be now looking at an incremental kind of impact or reduction in the overall solution.
On the 3% to 5% deflation estimate, what is the risk that this number keeps expanding over the next 2 to 3 years as model capabilities improve? Across the five GenAI offerings, which are seeing maximum uptake and where will you over-index? And on guidance, how much softness comes from geopolitical pressures vs events in the last 2-3 months including Opus 4.6 release?
Starting with the deflation number, 3% to 5% that I shared is mostly based on the industry mix of services. Most of the enhancement in models are really driving more and more velocity and efficiency in the SDLC lifecycle. That piece could go through a little higher deflation based on the model outcomes. In rest of the areas, it is Agentic, it is human-in-the-loop, and even the latest model on Anthropic's Mythos, ability to run production environment fixes without human-in-the-loop is very limited. For us, we called out 2% to 3% and that holds true even now. We have called out 5 key areas - Physical AI, AI Factory, Custom Silicon Engineering for Inferencing, AI-led Marketing Services, and IP and Platforms. AI Factory is where we are seeing tremendous traction - one large deal we called out is a $100 million+ AI factory deal for design, implementation and support of a next-gen AI data center for a large Technology company. We are already into two major clients for this, and hope to get to another three or four more in this coming year. Semiconductor Engineering also announced a new deal on Physical AI for ASIC development on advanced nodes. Coming to guidance, there has been some impact in March, which is what is reflected in our significantly lower revenue in Q4. Two large Telecom clients have cut down on discretionary spend for this calendar year. A couple of SAP programs got discontinued. We are seeing some softness in Europe, and the US seems to be quite robust except for this specific client situation. For FY'27, we see a half a percent reduction due to two clients and their own business challenges.
The impact in Telecom this quarter, sequentially some $12-odd million began in March. Should we assume a full quarter impact in next quarter? Over and above the 50 basis points from two other clients, this itself is quite a drag - is that a fair understanding, almost a 1% drag?
Yes, I wouldn't say that it was only in March, though the decision got communicated to in March, but it does have a little more impact than just one month because these were the SOWs which were expected as a part of CY'26 spend, and it was dragging and finally it was called off. I would say it's a little more than a month. We are assuming at least that it is there till end of the year and that is all accounted in our guidance.
On the cancellation of SAP programs - is this purely a budget decision or a technology or strategic decision?
These were related to the client budgets. They kind of de-prioritized this modernization. There is a general understanding that the timelines for some of this is also going to get extended from SAP. Some of that is probably playing into these decisions.
On the products business, considering many were end of life that were tried to modernize, in context of AI developments, how are you thinking about long-term trajectory? Likely to see higher cadence on new product launches?
Firstly, this Q4 revenue decline is not related to anything on AI or any of the latest developments. The last fortnight of the quarter is crucial for all the deal closures in the Software segment. The situation in West Asia led to deferral in client decision-making and some of the delays in the US government also caused this. We expected it to get done before March, but they didn't get done. There are three broad portfolios - One is data; second is operations, which is all our IntelliOps and AEX; and the third is experience, which has got some of the declining products. The two categories which I called out, data and operations, are growth-oriented. They will get offset by the declines in the experience portfolio. So, our expectation is low-single digit, flattish or marginally declining, in the coming year. As the recurring part of the portfolio improves, that will also help. Especially government buying is all perpetual licenses, so we are not able to clearly predict it at this point.
In first quarter, we usually see seasonality with productivity pass-through. Given a couple of account-specific issues will have full quarter impact, should we see higher than usual seasonality going into Q1? And on deal TCV with AI deflation impact - should we expect deal TCV to remain muted around $2 billion?
Rakesh, you can assume the usual Q1 seasonality. Despite the headwinds that we had in Q4, which will continue into Q1, the mega deal ramp-up is on track. And that growth will offset the headwinds from the two client challenges that we talked about. So otherwise, you can assume a usual Q1 seasonality.
And on deal TCV, the AI impact?
Yes, of course. I mean, $100 million deal would be much lesser today - maybe 80 million, just on a rough ballpark. So, deal TCV is flat. But technically, it does require at least 25%, 30% more effort to convert and get to the same number. But I also want to call out - we have lost some deals which are voluntary losses. We have walked away from some deals which will not make sense and that would have easily contributed at least $1 billion more to this number. It's only prudent to be a little bit more careful about this and spend the energy and organizational bandwidth on more reinventing for the future and enhancing our AI positioning and delivering more value to our clients using AI instead of really fighting some of the traditional deals where it's hyper-competitive. If it doesn't make sense, we walked out on quite a few in the last six months.
The second question is on capability side. The infrastructure managed services work for enterprise customers - do we have scope of work with hyperscalers as well? If yes, how does that differ from enterprise customer work?
Yes, of course. It is different because you're managing the hyperscaler networks, the AI Factory, a lot of the operational work; all of that is different. The tools used are different. The underlying network technology is significantly different. So, it's really a similar kind of capability, but it's different technologies on which we have to work. We have invested in training and retraining a lot of our infrastructure and data center teams to drive or to really participate in the new AIDC programs. And we've also hired a lot of lateral talent because a lot of this is also very geography specific. We have been hiring a lot more lateral talent on this front in the last maybe a year or so.
On the client-specific issues - two in Telecom, one in Manufacturing, and one in Retail - the three would probably be in ER&D part, and Retail in pure IT services part?
No, no, I mentioned the Telecom is in the discretionary spend. It's a digital business which is part of our ITBS portfolio. Manufacturing client is both engineering services and BPO; both large. Retail client is also mostly in our digital business. We were building a new platform and some of that is related to this.
Should we think about this low to mid-single digits being equivalent to high single digits given the demand or lack of growth in the last three years and now this being the fourth year of poor demand? Can we say some of this is also due to AI, the deflation impact?
No, I would say that very little has really played out in already reported numbers. We expect this to happen in FY'27 and onwards. That's also why you'll see this guidance lower than what we had given last year.