Can we outperform machine?

We have designed platform that tells us what should be the rational valuation zone of any particular company for the current as well as for the next year. This platform (machine) keeps track of all the available information in the market about that company and keeps updating its valuation zone as required. Also it advises on how much you should buy at certain price.

Sounds good, isn’t it?

But if we follow this machine’s advice, there is one surety that you will not buy expensive stocks after using this tool. Returns will still depend on stock selection and your analysis and understanding of the business.

We select growth stocks from the market which are still trading within their forecast valuation zone. Believe me, it is hard to find a good quality growth stock for cheap. But once we find that we look for our conviction in that business story.

We find our conviction through our research and analysis. Our analysis result then decides on what percentage of allocation should we give to that story in our portfolio.

Analysis Framework

Analysing or conducting a research is not any less than going on the battle field. Where do you start from? What should you look for? And many such questions keep flowing in the mind.

If you are the regular reader on our blog then you must have also read our analysis on Tech Mahindra last week. We will be using our analysis on this company to understand our analysis framework.

Our analysis is broken into 3 parts. As mentioned earlier that conducting analysis is like going on battlefield, the reason we believe in that is because of this favourite quote by Donald Rumsfeld:

As we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns – the ones we don’t know we don’t know. And if one looks throughout the history of our country and other free countries, it is the latter category that tend to be the difficult ones.

Known Knowns

Difficulty level: Beginners

Best used for: historical analysis, first glance

This is the easiest bit of analysis. You get all the information in the Annual reports and financial statements like what are the debt levels, past ROE, what were its margins in the past and what are today and where does it stand in the market.

All these information is known, historic and factual. It helps to understand the status quo of the business.

Known unknowns

Difficulty level: Advanced

Best used for: Forecasting, base case, sensitivity, understanding a range of outcomes

This is the toughest part of analysis and no one can predict this 100% with surety. Once we know the factual condition of company, we need to know beyond this to forecast its future.

We cannot know what the next year margins will be. But we can look at its revenues and its cost drivers to reasonable range of outcomes. For example, we found this information from Tech Mahindra management commentary that they expect their revenues to be in range of 10% – 12% for the 2018 financial year. We also know that its margins dented by 300 basis points in fourth quarter. We also found the drivers of this dent was loss of business from networking (business they bought) of around 160 basis point. 100 basis point was due to the change they are enforcing in their business and remaining 40 basis points was due to the forex fluctuation.

Loss of business from networking might not dent similar basis point in 2018 is our expectations. We can still expect it to affect its profitability in range of 50 basis – 100 basis point in worst case scenario. We expect there should be some positive affect due to the change they are doing in their portfolio and products in coming quarters. And finally we expect Indian currency from here on to strengthen in the coming year. So there should be no negative impact to its margin due to forex.

Hence we expect some positives in range of 50 basis to 100 basis in its profits margin for 2018.

As an analyst I always face many known unknowns – every assumption in DCF calculation is ultimately a forecast. Some of these unknowns are easier to predict and understand. On the difficult side it is very hard what margins will be affected by buying any other business and consolidating with its parent.

As mentioned earlier we cannot predict known unknowns to 100% surety. But we can gain an understanding of range of possible outcome. This helps us to quantify their effects to future earnings. For example, if Tech Mahindra in its 1st quarter report loss in margins by another 200 basis point, and mostly because of networking business and change they are enforcing then it is a time to re-look at both of its drivers and understand the sensitivity of it to the business.

Regardless, predicting unknowns is the difficult part and requires deep understanding of competitive dynamics, industry growth and intricacies of the business.

Unknown unknowns

Difficulty level: Black Swan event

Best used for: Keeping yourself grounded

The final category unknown unknowns is the toughest area and is impossible to predict. It creates the biggest “surprises”. Like we had recently seen in Kitex Garment stock volatility. It is hard to predict – like walking by and finding Rs 2000 note on the ground. Cannot predict take-over announcement, loss of key contract and development of new technology.

So how should you deal with these unknown unknowns?

We have two approaches to deal with this situation:

  1. Concede and acknowledge that you cannot predict all the aspects of the future. There is an element of unknown ahead and being aware of it sets you up well to react rationally when it occurs.
  2. Try your best to minimise the size of this category by focusing on wide range of known unknowns. Think outside the box and just don’t limit your known risk of those in annual report. For example in Tech Mahindra situation, recently they purchased IT business in relation to health care services (IT services). They have never done business in this segment. What could go wrong? Did they bought profitable business? Or is it HCI is a loss making business.


If you are the investor in the stock market then you are dealing with uncertain future that lie ahead of us. It is still possible to plan ahead and be more prepared for the range of future outcome. Although we cannot eliminate them completely but we can endeavour to minimise the size of unknown unknowns. The drivers that are involved in Tech Mahindra example are wide and growing with the number of M&A activities they are conducting to buy growth. This is giving a threat of bigger unknown unknowns list in our analysis and this is the reason our conviction level in the story of Tech Mahindra is 50%.

Machine is saying to invest 60% most but with half conviction level in the story, we prefer to go by most 30% (60% * 50%) of our allocation to this business.

Hope this blog helps you to conduct your research in stock market and understand what we mean when we talk about conviction.

Let me know what do you think about this post? And what is your conviction level for Tech Mahindra is?