How to write a specification for a product

how to write a specification for a product

Estimate the Production volume

We have a section on the revenue and profit model that you would use for your project, and another on economies of scale. In the authors experience, most electronics designed and produced for Niche products and startups (in Australia) has production volumes of several hundred to several thousand units. Again, in the authors experience, a few companies have been successful in making highly specialist devices (usually medical or heavy industrial) with sales of just 25 to 100 units. Why is this number so important? - With the higher quantity of units sold the startup costs get divided up more. e.g. If it costs $15k to design something and $100 to build it, if you can sell 200 of them they will have cost you $175 each.

You need to estimate your production volume as part of your specification as it makes a substantial difference to the way the project will be financed, designed and marketed. Producing this is often quite hard to do. For our electronics design work we normally only need the estimate accurate to perhaps -50%, +100%; ie predicting you will make 1000 a year, then actually making 500; or 2000; should not overly affect the design decisions.

With reference to the design, a cross-over occurs somewhere between 75 and 400 units (depending not just on the design, but the timescales, financing and nature of the circuitry) between two quite different construction techniques - surface mount and thru-hole. Surface mount costs more to design and set up, and less to make. Further crossovers appear at higher volumes for different types of custom parts - such as custom LC displays - and different types of tooling. In larger quantities some functions that would be normally be incorporated as modules - for instance RF circuitry - may instead by implemented on the PCB as seperate parts. The design cost would go up, but the unit cost would go down. As manufactured quantites go up, the time to test the units becomes an important consideration, a designer will devote more effort to implementing power on testing, and will perhaps offer to design dedicated test equipment to complement the manufacture. All these decisions are dependent on the production volume - but really only to an order of magnitude.

Some clients believe that it is dangerous to tell a design company - such as AirBorn - exactly what their product is, and how many they expect to sell. But a designer needs to know this in order to their job well. The designer is supposed to be working for you - not just for themselves. It is all a case of trust - know who you are dealing with, know their position in the company, ask the right questions and make a judgement call. Provisional patents and NDA's may well assist you to safe guard your intellectual property but really your first line of defence is dealing with someone you can trust.

. back to that new product brief.

Avoid detail: Use one line per idea

  • The best way to write a product brief is as a list of bullet points, or clauses.
  • Bullet points have the advantage of being clear and concise.
  • Somehow, using bullet points seems to encourage writing and then re-reading
  • Go back and edit the bullet points later

Avoid distraction: Get the ideas down

  • There is no time like the present, describe your idea
  • Again, To be useful, keep the main detail short - 1 page is ideal
  • Again, Concentrate on what is to be done by the product, (vs How. so much)
  • Again, Distinguish between requirements and WIBNI (Wouldn't It Be Nice If)
  • Start with the 3, 4 or 5 most important ideas that describe your new project

If the ideas are flowing, don't read - just write or type - you can tidy up the details later

If you are having trouble getting the ideas out, that is what this page is for. Putting it personally, as people we sometimes need other factors to help us work on something creative. Have you had success in doing this before - if so, what helped you then? A small reward may help - even a cup of coffee and a biscuit. Perhaps

An example of our design work listen to some music. Perhaps walking, or a change of environment. Perhaps carry a pen and notebook and write down ideas as you think of them during the day. Perhaps you can role play - imagine yourself describing this project to a co-worker, friend or customer - what would you say? What is good about your project? Write that down as a start. Get your ideas out on paper even if they are not perfect. Afterwards you can use these ideas as a very good starting point - you can fix up any mistakes later.

If despite your best efforts you have been unable to get your ideas on paper, realise that it happens to the best of us. It is called Writers block. The best remedy is time, or perhaps persistence - who knows - it will probably sort itself out. However, do consider if there is a reason why you might not want to complete the task. Could completing your project description end the fun of imagining the new project? You could always start writing a description for a new project straight afterwards. Maybe the new project will be even better. Could completing the project description mean you have to start the hard work of building it? Why? You have written out a project description, there is no rule which says you must therefore complete the project. If it is too hard, no one would want to do it. At the end of the day, as people, we sometimes have to look deeper to find our motivations, and why we do things, and why we sometimes do not.

Review your New Product Brief so far

    What do you have for your product brief so far:
  • A Project name (And perhaps a nice short name as well)
  • Project description (An opening statement)
  • A Killer feature the project has (if that is the way it sells)
  • The 3, 4 or 5 most important ideas describing the project
  • How many units you expect to sell
  • Who you are going to sell to, or how the product compares with what you already sell
  • Lots of assorted points about the project, and probably one or two items in great detail

If you have made it this far, you are most of the way there. The item you want to work on right now is the last one - the big list of points. You may find it better to do this after a break, or even the next day - the author does not know the size of your project.

When the author does this sort of work, there always seem to be one or two items that get covered in great detail. My impression is that there is nothing wrong with this, and it can only help going forward - but really it is not necessary to cover all items in great detail. In fact, if one or

two elements are covered in great detail, the best place for those points is on a seperate sheet with the main title at the top - it is good work, but it is not the basis for a short, succint specification. It should be an addendum. Take it out of the main spec for now.

    Take that list of assorted points:
  • Prioritize - what most helps describe the project?
  • Compartmentalise - group like items together
  • Rationalize - join items together if it seems to work, or take lines out if they are not really useful - cull - make every line count.
  • Refine - after a break you will see things differently, reword stuff

Everything should be made as simple as possible, but not simpler. --Albert Einstein (Paraphrase from "On the method of Theoretical Physics", The Herbert Spencer Lecture, Oxford, 10 June 1933)

Are we done? YES!

You should now have that list of assorted points slimmed down to a set of bullet points that concisely describe your project. A final review should happen to make sure you have not left out something. Often the thing that gets left out is glaringly obvious - so perahps try and role-play someone being told of the project for the very first time, to test for the possible omission. You may also like to read through the paragraphs in the next section summarizing items we would recommend reviewing.

The author has been involved in many hundreds of electronics projects at one level or another over the years. This page represents opinion based on experience. I hope it is of benefit to people in developing new products or ideas.

    Your new product brief:
  • Is a concise specification for a new product that can be built
  • Even if you don't build it, it represents a possible venture - you should keep it, you may look back in years to come and smile about it
  • (Having written one document like this you will find it easier next time)
  • The brief probably encapsulates all the information required by a subcontractor to finish your design work.
  • And now the hard work is just beginning - you need to build it!

Genius is one percent inspiration, ninety-nine percent perspiration. --Thomas Edison (Spoken statement (c. 1903); published in Harper's Monthly (September 1932)

Reviewing the new Product Brief

A specification should describe all the required attributes of the project - make a separate, if attached, wish list. When writing a new product brief it is important to distinguish between required and desired. Think carefully before you state a desire as a requirement - if you specify that the project is to be no larger than a mobile phone, be sure that the market really warrants that feature, because it may cost you a substantial amount to achieve.

Your opening statement should have embodied the function of your device succinctly. Reread it, and working from there make sure you flesh out the function of the project. Add points, rewrite, revise, but remember that often "less is more", the brief needs to be short and simple as well.

Your brief needs to describe how your product works. It does not always have to list every detail, because some are implied by its function. For instance, if your product brief stated that the green indicator LED was to flash at power on, and the red indicator LED was to flash if there was a fault, you would not really need to seperately state that a Green indicator LED was required, and also a Red indicator LED was required - but looking through the list below may help you to remember an item which has not been covered at all.

  • Inputs
Inputs often flow into functions which control outputs. For example, a reed switch might detect when a gate is open, triggering an piezo sounder for 30 seconds, and activating a display indicator. Or on a bagging machine, a microswitch may detect a bag landing at the base of a ramp, and start a cycle to raise the bag up the conveyor to the stacker. If you work through the inputs required for your project, you may find it helps define many of your projects functions. If we had to name this class of inputs, we would call them activating inputs
  • Outputs

    Outputs are usually the easiest items to list - the Motor than runs the pump, the sound alert, the relays that are the main control outputs, that sort of thing. Often they are controlled usually, in fairly simple ways. If you work through the outputs for your project, again you may find that it helps define much of the way your project works.
  • Modes

    We carefully worded the advice on outputs - because in nearly every project project the way outputs and so forth are controlled is usually fairly simple, but there are usually exceptions. Most normally the exceptions mean different modes of operation. For instance the gate alarm may have an off-season mode where the Piezo does not sound, but the display indicator still operates. Most devices also have a different way of operating when they first turn on, or are being tested, or if they detect a fault.
  • Programmability

    When you are first working through your project, leave programmability a little to the end. It can be quite engaging to consider, for instance, whether it is useful to make it programmable for the piezo sounder to operate for 10 seconds, 30, 60, or 120 seconds, however it is not as relevant as the conditions that make the piezo sound in the first place. Programmability also suffers from fashion. Years ago it was common to have two (+) different models of product, each with the different features. Programmability has been variously achieved with jumpers, dipswitches, and also with a single button used in many different ways depending on the state of LEDs. Programmability can be achieved with a display and menu buttons, and this is perhaps the most normal method at this time. I predict future devices will be more likely networked, with all programmability through the network using a browser interface or other software.
  • Tying it together

    It may sound too simple, but with Inputs, Outputs, Modes, the functions joining them all together, and the Programmability mostly considered last, you have most of the ingredients for most projects. By considering these and trying to join them you are probably well on the way to a concise point-by-point specification. If it is not coming together, you could try talking to a design engineer (such as Airborn Electonics ) You may benefit by going into more detail on the points, see Specifying a new design: Technical ingredients We also need to consider:
  • Power Supply

    Choices are usually between DC power, Battery (esp Lithium Ion Rechargeable) and Plugpacks.
  • Protection (such as fuses, although fuses are less used these days), and replaceable parts
  • Connector types (such as unpluggable screw terminals, RJ45 CAT6 connectors,etc)
  • Physical mounting or enclosure for the circuit board
  • We wish you the very best with your new project, and hope that you choose AirBorn Electronics to help you with the Research & Development work.


    Category: Insurance

    Similar articles: