The history of a startup is sorting out other people's mistakes. Part 1
Good day, ladies and gentlemen!
In continuation of the exciting topic of startups ( one and two ) and possible errors at their launch, this article appeared.
Initially it was supposed to fit everything in one post, but then something all began to somehow swell and not intervene in the framework of decency. Therefore, 2 parts.
')
An example in this case would be one startup from two stubborn enthusiasts and the mistakes they made. Just want to say that the guys are great, and I would like to wish them continued success. I hope that further it will be easier for them, because they have already stuffed the cones :)
In the second part of the article, I will tell you about the mistake that was already made on my part, to be honest, I screwed up in one place.
At the time of our acquaintance, the guys had a successful crowdfunding campaign, a website, a promotional video and a prototype.
They managed to live and learn in different countries- USA, Venezuela, and eventually came to Hong Kong. One of the founders was a mechanical engineer, and the second was an electronics engineer, plus writing software. They had knowledge, and probably, in some areas, more than mine, but there is no practical experience, which did not let them down.
On the day when we met, it was May 15, 2014 (note the time for further discussion). After tasting the traditional cuisine of Guangdong province and reaching the office, we (on our side was me and the sales person) sat down in a cool conversation and got down to business. The first thing asked was potential customers about their plans.
- The first batch will be 1000 items. We are currently developing a management fee, but it is not yet ready. We will finish it in a week.
- And when do you plan to send?
- June 16th! We have already announced on our website, sent confirmations to customers, they are already up to date and are awaiting shipment!
Here we with our sales looked at each other.
- And how did you count the time?
Potential customers continued:
- After the end of the development of the board 3 days to order and delivery of components. Week on the production of 1000 boards. Then a week for installation and another week for the final assembly. We plan to get the equipment in 3 weeks.
Our salesperson and I looked at each other again. I asked a simple (for me, because at that time already 2.5 years in the EDMS business (Electronics Development and Manufacturing Service) question:
- What if some components have half a year delivery time?
Here startups looked at each other:
- It's like half a year? And so does it happen?
- Well, depending on which components you choose, the delivery time can vary greatly. Something can be immediately in stock, then it can come in a week, and something will have to wait six months.
The guys are very sad.
“There is a spot-buy option,” I continued, “Then the time may be reduced to 4 weeks.” Again, if the components are there. Just pay will be ready. Standard delivery time is 4 weeks.
The guys cheered up a bit.
“Well, okay,” I say. “And how did you decide that the time to make the equipment was 3 weeks?” Have you already confirmed this time?
- Not. This we ourselves have rated so.
“Look,” I say. Standard production time 5-6 weeks. Do not forget about the fact that it is necessary to make an advance payment, we are laying a week on it, because money through the bank can go up to 5 days. And do not forget about the DFM-loop (Design for Manufacturing). If there are many injection molds, it will easily take another week to prepare DFM reports. And then you still need to look at them, agree; if necessary, make changes to the details. Feel free to throw another week. In total, something turns out about 9 weeks.
The guys opened the calendar in the laptop and began to estimate the time.
- 9 weeks - well, about 2 months ... Fees 1 month ... Electronic components, we will assume the same ... So, with ... We can tell customers that we have decided to make some changes that will greatly improve everything and we very, very much ask them to wait another 2 of the month!
The mood among potential customers has greatly improved. 2 months delay is not small, but still somehow bearable, they decided.
Sales and I looked at each other again. Here he got into our dialogue:
- That's not all. If you need to sign an NDA (Non-Disclosure Agreement), it will still take some time, because we need to send it to the head office for a signature. Next, we first produce several samples to debug our production, then send it to you for approval. And only after that we begin production. In addition, we will need to put you a quotation for the final product and equipment. Given that the board and BOM (Bill of Materials) are not ready for you, it’s impossible to say for sure when we can put up our offer. Let's take the middle case and calculate when you can actually send.
After that, he took a marker and sketched the main stages on the board. It turned out that the first shipment may be around mid-September (that is, 3 months after the planned date).
Sadness-sadness piled on potential customers ...
And so, I think it has already become clear where a mistake was made when planning - the guys had no idea how the procurement process works — how much the procurement of components takes and how long the production of tooling takes. Plus overhead temporary expenses were not considered by them in principle, because they simply did not know that expenses exist.
In general, the guys probably knew how to design circuit boards. I am talking only about printed circuit boards, because, as it turned out later, there were also problems with the knowledge of the principles of the development of the final product. For example, CAE (Computer Aided Engineering) for modeling plastic casting was not used correctly and the results they received were not correct in principle, but due to lack of experience, they simply could not interpret the most modeling results correctly. But they had beautiful pictures.
Next, consider what is happening, so to speak, “under the hood” and that the startups did not know.
When the fee is ready (by the client), it is transferred to the commercial offer. The shortest chain (at that time): customer-sales-engineers, purchasers-sales-client. It is clear that this whole journey takes some time.
For example, a client is satisfied with the proposal, and he gives the go-ahead to the production. Then a project manager is appointed (who leads him before being put on mass production). Next, a BOM is created and then the purchase of components and accessories begins.
So, startupers (as I understood later) at that moment thought to buy everything on Digikey, and to order boards from companies that are engaged in the production of prototypes. You can, of course, do so, it all depends on the goal. If the goal is to invest as much money as possible and make the products as expensive as possible (at least ten times more expensive than it could be), then this approach, beyond all doubt, works. We will consider the usual process when the price should be competitive (I do not say as low as possible, since this is not always a priority. By the way, priorities will be a separate article).
We look what happens next. Once the BOM is ready, purchasers start purchasing components from suppliers. Suppliers are taken not from the ceiling, and not from the Internet, and not even from a store around the corner, but from AVL (Approved Vendors List). Anyone who does not hang out in that sheet. There are suppliers who are encouraged by customers, approved by the component manufacturer (such as Microchip, NXP, etc.) or that have passed our own audit.
Let's take a break for one subtlety - besides AVL there can be AML (Approved Manufacturers List), which can be part of AVL. In AML, usually those who produce something directly. Specifically, for us, for mechanical parts, I insisted on the removal of AML from AVL, because AML for mechanical parts is a separate long story.
Why use AVL? In order to avoid problems with counterfeit components, plus the stability of supply. Pay fines for 800 thousand dollars for one fake resistor? Yes, easily! (the story of one company, not ours, no :)).
As mentioned above, the delivery time of components depends on the availability of a vendor in stock. If the warehouse is empty, then you have to wait until it is produced. By and large, the scheme is not complicated, but there are some nuances. Our potential customers had no idea how the production organization works and how long different processes take.
Here for the first part of the article on this and all.
What conclusion from the described can be made?
1. Knowledge is power!
2. Learn the details about the area you are working in. There may be rather unobvious moments that strongly affect ... everything. All of them can influence :) As they say, the devil is in the details. By the way, the Germans, as I found out, have exactly the same saying.
3. Find someone who is already working in this area and has experience, consult for advice. It is better to spend money on a box of beer than to spend money on modifying a snap.
4. In the end, get a job if you have just finished your studies. For example, we sometimes have internship programs. An intern wanders from department to department and studies all processes. The path can be very different - IT, production, sales. And you can please me in the department, yes :)
PS Shipment of products for some time already. In one of the photos under the translucent details are visible points of the included LEDs. And if we worked with us, I would help them to solve the problem with the visibility of the LEDs :) It would be a beautiful uniform lighting without visible LEDs.
In continuation of the exciting topic of startups ( one and two ) and possible errors at their launch, this article appeared.
Initially it was supposed to fit everything in one post, but then something all began to somehow swell and not intervene in the framework of decency. Therefore, 2 parts.
')
An example in this case would be one startup from two stubborn enthusiasts and the mistakes they made. Just want to say that the guys are great, and I would like to wish them continued success. I hope that further it will be easier for them, because they have already stuffed the cones :)
In the second part of the article, I will tell you about the mistake that was already made on my part, to be honest, I screwed up in one place.
At the time of our acquaintance, the guys had a successful crowdfunding campaign, a website, a promotional video and a prototype.
They managed to live and learn in different countries- USA, Venezuela, and eventually came to Hong Kong. One of the founders was a mechanical engineer, and the second was an electronics engineer, plus writing software. They had knowledge, and probably, in some areas, more than mine, but there is no practical experience, which did not let them down.
On the day when we met, it was May 15, 2014 (note the time for further discussion). After tasting the traditional cuisine of Guangdong province and reaching the office, we (on our side was me and the sales person) sat down in a cool conversation and got down to business. The first thing asked was potential customers about their plans.
- The first batch will be 1000 items. We are currently developing a management fee, but it is not yet ready. We will finish it in a week.
- And when do you plan to send?
- June 16th! We have already announced on our website, sent confirmations to customers, they are already up to date and are awaiting shipment!
Here we with our sales looked at each other.
- And how did you count the time?
Potential customers continued:
- After the end of the development of the board 3 days to order and delivery of components. Week on the production of 1000 boards. Then a week for installation and another week for the final assembly. We plan to get the equipment in 3 weeks.
Our salesperson and I looked at each other again. I asked a simple (for me, because at that time already 2.5 years in the EDMS business (Electronics Development and Manufacturing Service) question:
- What if some components have half a year delivery time?
Here startups looked at each other:
- It's like half a year? And so does it happen?
- Well, depending on which components you choose, the delivery time can vary greatly. Something can be immediately in stock, then it can come in a week, and something will have to wait six months.
The guys are very sad.
“There is a spot-buy option,” I continued, “Then the time may be reduced to 4 weeks.” Again, if the components are there. Just pay will be ready. Standard delivery time is 4 weeks.
The guys cheered up a bit.
“Well, okay,” I say. “And how did you decide that the time to make the equipment was 3 weeks?” Have you already confirmed this time?
- Not. This we ourselves have rated so.
“Look,” I say. Standard production time 5-6 weeks. Do not forget about the fact that it is necessary to make an advance payment, we are laying a week on it, because money through the bank can go up to 5 days. And do not forget about the DFM-loop (Design for Manufacturing). If there are many injection molds, it will easily take another week to prepare DFM reports. And then you still need to look at them, agree; if necessary, make changes to the details. Feel free to throw another week. In total, something turns out about 9 weeks.
The guys opened the calendar in the laptop and began to estimate the time.
- 9 weeks - well, about 2 months ... Fees 1 month ... Electronic components, we will assume the same ... So, with ... We can tell customers that we have decided to make some changes that will greatly improve everything and we very, very much ask them to wait another 2 of the month!
The mood among potential customers has greatly improved. 2 months delay is not small, but still somehow bearable, they decided.
Sales and I looked at each other again. Here he got into our dialogue:
- That's not all. If you need to sign an NDA (Non-Disclosure Agreement), it will still take some time, because we need to send it to the head office for a signature. Next, we first produce several samples to debug our production, then send it to you for approval. And only after that we begin production. In addition, we will need to put you a quotation for the final product and equipment. Given that the board and BOM (Bill of Materials) are not ready for you, it’s impossible to say for sure when we can put up our offer. Let's take the middle case and calculate when you can actually send.
After that, he took a marker and sketched the main stages on the board. It turned out that the first shipment may be around mid-September (that is, 3 months after the planned date).
Sadness-sadness piled on potential customers ...
And so, I think it has already become clear where a mistake was made when planning - the guys had no idea how the procurement process works — how much the procurement of components takes and how long the production of tooling takes. Plus overhead temporary expenses were not considered by them in principle, because they simply did not know that expenses exist.
In general, the guys probably knew how to design circuit boards. I am talking only about printed circuit boards, because, as it turned out later, there were also problems with the knowledge of the principles of the development of the final product. For example, CAE (Computer Aided Engineering) for modeling plastic casting was not used correctly and the results they received were not correct in principle, but due to lack of experience, they simply could not interpret the most modeling results correctly. But they had beautiful pictures.
Next, consider what is happening, so to speak, “under the hood” and that the startups did not know.
When the fee is ready (by the client), it is transferred to the commercial offer. The shortest chain (at that time): customer-sales-engineers, purchasers-sales-client. It is clear that this whole journey takes some time.
For example, a client is satisfied with the proposal, and he gives the go-ahead to the production. Then a project manager is appointed (who leads him before being put on mass production). Next, a BOM is created and then the purchase of components and accessories begins.
So, startupers (as I understood later) at that moment thought to buy everything on Digikey, and to order boards from companies that are engaged in the production of prototypes. You can, of course, do so, it all depends on the goal. If the goal is to invest as much money as possible and make the products as expensive as possible (at least ten times more expensive than it could be), then this approach, beyond all doubt, works. We will consider the usual process when the price should be competitive (I do not say as low as possible, since this is not always a priority. By the way, priorities will be a separate article).
We look what happens next. Once the BOM is ready, purchasers start purchasing components from suppliers. Suppliers are taken not from the ceiling, and not from the Internet, and not even from a store around the corner, but from AVL (Approved Vendors List). Anyone who does not hang out in that sheet. There are suppliers who are encouraged by customers, approved by the component manufacturer (such as Microchip, NXP, etc.) or that have passed our own audit.
Let's take a break for one subtlety - besides AVL there can be AML (Approved Manufacturers List), which can be part of AVL. In AML, usually those who produce something directly. Specifically, for us, for mechanical parts, I insisted on the removal of AML from AVL, because AML for mechanical parts is a separate long story.
Why use AVL? In order to avoid problems with counterfeit components, plus the stability of supply. Pay fines for 800 thousand dollars for one fake resistor? Yes, easily! (the story of one company, not ours, no :)).
As mentioned above, the delivery time of components depends on the availability of a vendor in stock. If the warehouse is empty, then you have to wait until it is produced. By and large, the scheme is not complicated, but there are some nuances. Our potential customers had no idea how the production organization works and how long different processes take.
Here for the first part of the article on this and all.
What conclusion from the described can be made?
1. Knowledge is power!
2. Learn the details about the area you are working in. There may be rather unobvious moments that strongly affect ... everything. All of them can influence :) As they say, the devil is in the details. By the way, the Germans, as I found out, have exactly the same saying.
3. Find someone who is already working in this area and has experience, consult for advice. It is better to spend money on a box of beer than to spend money on modifying a snap.
4. In the end, get a job if you have just finished your studies. For example, we sometimes have internship programs. An intern wanders from department to department and studies all processes. The path can be very different - IT, production, sales. And you can please me in the department, yes :)
PS Shipment of products for some time already. In one of the photos under the translucent details are visible points of the included LEDs. And if we worked with us, I would help them to solve the problem with the visibility of the LEDs :) It would be a beautiful uniform lighting without visible LEDs.
Source: https://habr.com/ru/post/292236/
All Articles