In my previous posts, I tried to talk about how the freelancer and the customer interact. Most of the thoughts presented in these articles were taken for granted, which was pointed out to me more than once. Today I would like to start a conversation about how we, having safely gone through all the vicissitudes of the initial discussion, begin the project. I will again be banal to disgrace, again and again I will grind that which is clear to everyone. But, as practice shows, we all with the stubbornness of the masochists are attacking the same rake.
Getting started
So everything is fine. You have successfully convinced the customer that you are the most wonderful / coolest / kindest performer who performs the main Wishlist “so that herak has one button, it’s ready and everything is ready.” I congratulate you on this, we begin negotiations on specifics.
A little distracted.
95% of those who read this post have currently experience / lead such a project / lead a project and do not know what it is. Which one Which, even if you ruin everything for yourself, will not end on the date set by you, and will never be worth as much as it was announced. Having passed such a project, having quarreled with the customer (well, and I am so beautiful, I overfulfilled the work plan, went beyond the scope of the TK, and he was beating his fingers), we lean back in the armchair, sweat from the forehead in the morning, rub off blushing eyes, critically examine corns from keys, and safely promise ourselves that this will not happen again. And run into the same r on the next project. I thought for a long time why this is happening, and I would like to share with you my thoughts and techniques that allow you to avoid such problems.
During the start of work, we are all curtsy with the customer, we read letters, we brag about the portfolio. At the same time, when a real opportunity arises to show off your skills, we spit on it, trample into the dirt, and lay a time bomb under us. What am I talking about? About the cost and timing.
')
I have come across more than once, well, and what’s really there, I myself am guilty of the fact that in small groups of developers / individual developers the percentage of correct estimate of cost and time is inversely proportional to the size of the project. the face is individual, but always visible. As a rule, the face is equal to the double sum of the largest project. Those. if the contractor has previously worked with the project for 150 TR, then the estimate of the cost and terms for the project, the cost of which is 300 TR, will be incorrect. Not “most likely so much,” and not even “may be so much,” but precisely wrong. Record a fountain pen right on the cerebral cortex.
If you evaluate the project with double excess of the cost of the most expensive of your project, then you brazenly, shamelessly lyingYou're lying to yourself, the customer, and the laws of this world, if you list everything.
Well enough to get scorched, I think everyone understood everything, let's move on to what to do about it.
Medicine
Medicines are known to be of two kinds. Cold pills and a complex of drugs for severe illness. To what kind include the excess of time and budget? Are you still in doubt?
Let's start with the preparation of the patient.
The problem, if you understand the essence, in our brains. Just do not have to start quoting Zadornov now and tell us that our nation is the essence of a set of
debunks . No, of course we have a sense of responsibility for the result is much less than the same Japanese or Germans. But 95% is an indicator of going beyond the budget, not only for the programming industry. And not only in our country. Statistics took from a pile of books on the management of projects that had to shovel.
But we digress. The problem is in the brain. Admit it, and it will be easier and easier to find a way out. After all, there is always the temptation to refer to an inadequate customer, changing TK, the idiocy of those who implemented the API for the libraries that were used in the project, and the phase of the moon, in the end. But I will reveal the secret - most of the customers are exactly like that. Of course there are deviations for the better and for the worse, but the basis is the same. And the libraries were written that way, because there is a reasonably adequate reason for which you may not know. Well, the phase of the moon affects the okiana, to which you can hardly refer yourself.
To begin with, how to set deadlines and budget. If the project is small, thousands by 20, and the term of work is 5 days, then we have no problems, right? And why? Because on the one hand, all our mistakes are displayed as a percentage of the project parameters. And if for a reduced deviation of 30-40% is 1.5-2 days and 6-8 tr. With which to sacrifice, then there are other orders in large order. The reality is that the deviations I indicated are a very good indicator. Why? We go further. The second reason that the deadlines are called less reliably is that the project is physically easier to evaluate and survey. On large projects, the vision of the complexity of work is very vague.
Separately, I would like to highlight projects that are similar to what has already been done. How many wonderful discoveries are to us ... Everyone loves stories when the author finds himself in a stupid situation, therefore I will give an example from personal experience. As it had to face a project that was like two drops similar to that. that already had to do. Do I have to say that the project is twice as long as possible, not to mention the bumps with the customer and the loss of profit? Of course, it was not without interference from the customer, whose representative at the beginning promised to clarify all points, and after signing the contract, he began to mess up. But we have already agreed that we ourselves are to blame first of all for the Fails.
Notice how the problem description essentially gives answers to them? Consider in more detail.
First, TK is devastatingly little time. Yes, there are situations when you can work without it. For example, I have a customer with whom we work without TK at all, and it doesn’t cause any issues to change terms and costs after changing the original agreements. But for a hundred projects, I have only one such positive example.
The more TZ, the less time is spent analyzing it, darling. From here - problems with cost and terms. Large TK usually means a lot of work. And sometimes it’s just dumb to call the customer not 100 tr. for the project and 1.5 months (which of course we took stupidly from the head), and three hundred and 4 months. Do not be afraid!!! If the customer is not ready to pay for your work, and this is work, and not picking his nose, then there is nothing to begin.
The problem is complicated by the fact that the customer, as a rule, is not interested in accurate valuation. Yes, yes, you heard right. The customer is interested in a quick assessment. This follows from the fact that turning to freelancing, he gets dozens of offers. And since the project often had to be done “just yesterday,” the main violin is played by the time it takes to get estimates of the cost and terms on the basis of which the choice is made. Therefore, the analysis of TK is your sole concern.
How to deal with it? Simple enough - break the project into blocks. The circuit is called a PERT chart. How to carry it out - there are plenty of examples on the Internet. The breakdown does not take much time, and it is much easier to evaluate the individual parts by laboriousness. Evaluate as if it were a separate development. At the same time, to determine the exact dates, it is not bad to multiply the amount received by 1.2 in order to ensure time for the project to settle down. How much to multiply the final figures in order to secure a reserve in the form of a buffer is a personal question for everyone :)
Here, most likely, there could be some kind of integrated program for the implementation of assessment work, but unfortunately I did not find it. Next, I will talk about other stages and techniques, and you will understand why I say that. Even in the project management books there are 3-4 software products, at the same time at all stages the analysis is performed in one program or another. During the work I have used several products, and I can say with confidence that if in one program something is done easily and simply, then in the other it is not provided at all, or done in one place.
I was thinking of writing a program for managing projects, but it was quite difficult to do work alone, especially in conditions of high employment, and I couldn’t help my guys to do it :) So as soon as I’ll tell you about the results at the end of the project.
Something I signed for ... I think that's enough for today, in the next part we will look at the problems of starting work directly on the project.
part two .