Hey guys, it's sasha, there is a huge amount of information out there about how you can go and value a company. The problem for me is that a lot of this information a lot of the ways people do. These valuations, in my personal opinion, has massive flaws, and that's why i do something really quite different. Let me tell you exactly what these flaws are now getting the valuation of the company is key.

It is critical to understanding whether or not you want to go and invest in a company. You need to understand exactly what your view is and what the right price for a share is right now versus what the market valuation is, and if you see a massive upside, you may then go and choose to invest on that basis. The industry standard way of assessing company valuation is through something called a discounted. Cash flow model called a dcf.

If you haven't seen one of these before it might seem somewhat complicated, and it really isn't all that bad if you begin actually trying to dig underneath of what all the numbers are and what it's actually trying to do now, i am going to simplify things. A lot in here in this very short youtube video. It is not a detailed finance lecture; it doesn't cover every single little tiny bit. I'm gon na simplify things a lot just before people go and write comments about how wrong i am and how i miss something out.

That is the reason why also, i am not a financial accountant. I am not a financial advisor. I am not here to go and provide you with any specific advice, because i cannot do so. If you do need financial advice, you will need to go and find the help of a suitably qualified, professional somewhere else.

Now that's out of the way. Let's talk about these dcf models, the way these dcf models basically work is they go and project what the cash flow p, l for the company and the balance sheet for the company will be over the next period of time. Usually they talk about periods of five years and at the end of the five years they calculate something called the terminal value which is sort of their expectation. What the value of the company at that point will be.

They then go and add up all of the money at the bottom of these models. So the actual money that you're essentially going to be earning as a shareholder in the business together tone tournament value, and then they discount those individual values and discounting basically achieves two different things. One is a sort of way of assessing what the value of the money at some point in the future is versus today against the value of that money. If you had it invested somewhere else, it is sort of a way of saying if you weren't putting your money in this company, what would sort of be the equivalent return that you would normally expect on this kind of money, which would maybe prevent you from putting That particular amount of cash into this investment - that's kind of one side of things in a way again, i'm simplifying horribly here, but just so that people can try to understand it, and the second kind of element of why people often use discounting is to sort of Account for the risk potentially of the company uh in out years, so the further out you go from today, the more risk there is and the more fluctuation there is probably around the projected numbers.

So if you want to sort of minimize the risk of overestimating what the revenue is going to be in five years, you probably don't want to give it the same weighting as the revenue you would expect next year, for example, so people also use discounting in order To affect for that, and essentially not count the revenue and the numbers in five years, quite as much as you do now now some people do do these two things in two separate sets of assumptions in the model, but i'm just trying to keep things as simple As possible, so then people talk about the total sum, which is the sum of all of these different cash flows and the terminal, value etc, and they add them up and essentially what you get at the end is the net present value of a company. All you have to do, then, is just go and take that number and essentially divided by the number of shares in circulation, and that's how you get your average price per share and that's how people typically do these calculations in a very simple way. But here are some issues that i have that i think are really important that when people talk about this, i don't really hear people mention - and i think can be massive massive issues with how people then make assessments as to what the value of a company is. The first problem for me is that almost a hundred percent of these models are done top down rather than bottom up.

As someone has done a lot of big finance deals, i've done lots of big mergers and acquisitions and acquisitions of portfolios, and things like that. This really irks me, although this is pretty much the industry standard way of doing it now. What do i mean by top down and why? Why do i have an issue with it? To put it simply, a top-down model: um just takes the overall company, p l to date and essentially looks at a few years of performance, maybe like up to three and then it projects what the numbers will be going forward, but it just takes the top level. So, rather than working with any of the things that make up those numbers, they take the total, p, l basically say: okay, the revenue was doing this three years ago, this two years ago, this last year.

Therefore, i think the next number will be that and they sort of project based on those total numbers now usually they'll project like just something as simple as just the ebitda or some kind of other net profit type figure. In some cases, they'll go and project individual top level items, so maybe they'll go and take the revenue projected and then separately project the cost line as well, but that's pretty much as far as a lot of these models typically go. So if you're doing this, if last year say your profit was 1 billion, what are they going to be doing? Is it adjusting it based on some kind of flat growth factor in most cases, so people say i think that the company's going to be growing at a rate of 20 per year because of a bunch of different things that i thought and it sort of averages About 20, so they'll say that next year it's gon na be 1.2 billion dollars and then the year after it's gon na be 1.44 billion dollars. So you can understand how this works already.

The issue is that this approach makes really little effort in trying to understand what actually are the drivers for these numbers? Why is it that they achieved say 1 billion last year? Why do you think, will it be 20 next year, rather than 40 or 10, and why won't it be a different number the year after what? Why is that a constant number, and - and how do you make those determinations and the difficulty with top-down models is almost impossible because you don't really work with the underlying drivers that create those performance figures in the first place. So let's say i have a company with three recent years of growth in their p l, i'm going to put it up over here. How do i value their growth? This is how most people will do it, but you could also do this or this, but what? If the performance dips, because of the way that their products and propositions are going or the way that their cost base is rising? What if this is the early part of an exponential growth cycle which of these is the accurate one? Well, it's very, very difficult to actually make that assertion. Make that assumption, based on a top-down model, because again you're not really understanding, what's actually driving those numbers and you're not making projections on those underlying drivers you're only extrapolating the top level figures.

So this is the reason why i think this is really important. A minor tweak and a small number of assumptions within the model can make a huge difference to what the final valuation will be. The second issue for me is that these models are usually really really bad of converting company investments and company r d and company current expenditure on building infrastructure into future revenue. I haven't noticed when i'm doing these models, that when people go and do them, people will separately extrapolate say what the revenue number is or what the p you know, the net profit number is or whatever it is, they that they use and then they go and Separately, maybe if in most cases they won't even get that far but they'll go and extrapolate say the cost number, but what they don't really very effectively take into account is when a company is very heavily investing in itself when the company is injecting huge amounts of Money in massive one-off architecture builds massive, one-off injections into r d, massive things that will fundamentally change the performance of that company in the future.

And i don't correctly take into account what that might do to the performance going forward because of that same kind of top-down projection issue. So if a company is investing a huge amount into those things right now, does that get reflected in these models within the performance? Over the next few years, and also, if they're, injecting that much right now, you probably wouldn't expect the same uplift and performance that you're going to be seeing over the next two to three years to still be there in five years time, because likelihood is then i'm Gon na be continuing to put the same amount of money into those things forever. A good example of this is tesla right now, they're building these huge gigafactories all over the world in different continents - and this is an incredible incredible amount of investment. They're pulling in are people genuinely fairly, attributing the potential value of what these things are going to bring over time.

I don't know you have to make your own decision on that. Are the dcf models really accounting for what the impact of these is going to be. On the revenues next year in the revenues the year after that, i i don't know, my opinion is that i think there is a better way. The last one that i find somewhat amusing is that literally every single book on investing every single lecture, every single disclosure - that is a legal disclosure on an investing website, tells you something like past performance is not indicative of future results.

Have you seen that one before? Well, here's the funny here's the funny bit for me. Everyone then values. These companies then goes and does exactly this, because this is how a dcf model works. You basically go and look at the last three years worth of performance and say well.

The past performance was this therefore feature performance is just an extrapolation of where those particular numbers are going, and, although i do understand, i am aware that the only data available on any companies - the data - that's already happened, but in my mind, there's a much better way Of doing it through doing a proper bottom-up product or service like model that i'm going to be talking about in a great level of detail in upcoming videos. So if you haven't subscribed already and you're interested in that kind of stuff make sure you go and subscribe to this channel, because that is going to be a lot of what i'm going to be talking about over the next few weeks and months. I'm going to be talking about how inviting specific companies using that particular approach. Thank you very much for watching.

I really really appreciate it and i'll see you guys later.