• EEVblog #73 – How to screw up your winning product

    Another random drive-time rant.
    Dave rants about how dickhead marketing changes can screw up your winning product.

    The Tropicana Story:

    Be Sociable, Share!

      About EEVblog

      Check Also

      EEVblog #1056 – Digilent Open Scope MZ Review

      Dave looks at the Digilent Open Scope MZ Review, an $89 open source oscilloscope, logic ...

      • Ray Jones

        Ah yes, the 99SE version of a “certain company” became the standard for a long time.

        Certainly my first encounters with the new “better” version, especially the “better” network based licencing became so frustrating that I still stick with 99SE till this day.

        Perhaps if I was doing more PCB work than software these days I’d revisit the latest.
        I have seen the latest and it looks glitzy, but pure unadorned functionality can’t be beat.

        Do the keyboard shortcuts from the earlier Hobart based DOS versions like ‘P’ ‘T’, ‘P’ ‘P’ etc still work on the latest and greatest?

        • Yep, shortcuts still work.

      • edward


        Change is good, when it’s good. It’s bad when it’s bad. I don’t think you’re still using VHS video or analog cell phones, or cassette tapes, are you?

        The key to successful change is to do your homework. Test test test test. And if you can’t test, then release to a small audience and increase the size as you become convinced it works.

      • Brian Hoskins

        Hi Edward,

        I think you’re missing the point a little bit there. I don’t think you can use the change from VHS to DVD as an example, because that’s an actual step forwards in technology and completely *different* products will have been designed to make use of that technology.

        Still, if you design and build a really good DVD player (for example), and it tops the sales chart, then you’re probably not likely to be making good decision if you decide you’re going to completely change it. That’s what Dave is talking about I think.

        That doesn’t mean you can’t build different products that later serve a different audience or perhaps make use of a newer technology.


      • L

        That’s one of your posts that make me keep watching your blog, Dave. You’re not only talk about transistors, but about a bunch of interesting related topics like product design and dickhead MBAs 😉

      • Ray Jones

        Did anybody else notice the “step motion” of the background images whizzing past?

        An artifact of change dare I say!

        Dave, does this happen with your original source video?

        If so, then it proves the old analog video despite the lack of crispness at times handles fast motion in a superior manner!

        • It doesn’t really happen on the original MP4 from the camera. The video gets compressed and converted multiple times before it gets to you.
          You can’t stream uncompressed “analog” video on the internet, the blog wouldn’t exist if it wasn’t for compressed MP4 video!

          • Ray Jones

            Yeah it would be a bit of a stretch to stream analog signals on the net.

            So yeah must be part of the multiple trans-codings that take place to get onto YouTube’s final feed.

            I know if you do a quick DVDShrink you can get the same effect, reference frames only exist every so often, and the incremental differences pile up giving the abrupt step back to reality in the compressed version.

            Mind you things are much better than the first digital feeds I recall when watching OS Grand Prix’s via satellite feeds in the 80’s. Huge stop motion jumps in the background when tracking a car around a corner!

      • ivan

        My case is worst 🙂
        I am working in subsidiary of another “certain company” which is located in a post socialist country. Therefor we don’t have many MBAs, so we bring our managers (and team leaders) “from the street” because (according to the management gurus) this is the “American way [to success]”.

      • Mike Duan

        Protel / Altium (well known Australia company) is one good example same as this story.

        Hi-Tech (grown up in Australia, Now part of Microchip) and its newest PICC and PICC18 compiler (9.6x or newer) is another good example how MBA(s) screws up a good company.

        Now the newest magento 1.4x release is the same as this story. Everything stops working after upgrade from 1.3.2.x. Even though the 1.4x is claimed to be “matured”. What a MBA dominated world!

      • As someone whose entire career has been based on redesigning successful products, I have to agree absolutely with everything that you have said in this post.
        Frequently we’ve suffered top brass completely overriding design decisions with absolutely no technical, economic or ergonomic justification. These decisions were based on marketing plans or worse, motivated by internal politics. And too many times this has resulted in the catastrophic failure of the entire product line.
        While it is true that many of the changes that I have made are embarrassingly stupid in retrospect, I am not ashamed of them as I have made these mistakes in earnest.

      • John M

        My rant actually has nothing to do with Product Design, but the MBA problem is very real. My ex-company ( the details of why I no longer there won’t be discussed here) had these Six Sigma Black Belts come in and make sweeping changes to the way my department was organized and run and the results were appalling. It ended up taking 5 to 6 times longer just to get the paperwork done! Which added 1 to 2 hours of extra labor every day and they were working us overtime to begin with! Yes, yes, it allowed them to micromanage down to a much lower level, but, having that in place absolutely did not improve our efficiency – it made it worse! I remember the Six Sigma person in our staff asking us what we thought of their plan or if we had any questions. Dead silence. Top Down management is inherently ineffective in promoting communication.

      • DarkSider

        As Dave says in EEVBlog #73, a great product “is all about getting a whole bunch of small things right”.

        I agree with a lot of the good points that Dave made in the EEVBlog #73
        video rant about the perils of “Marketing”. There are certainly some dickheads in Marketing. It’s not only dickhead marketers with MBAs that can screw up your product. In the interests of full disclosure, I have to tell you that I have moved over to the Dark Side. I have the utmost respect for the brilliant design engineers that I have had the pleasure to work with. They have created some amazing and innovative products with give great customer experiences.

        In an open society it’s good to hear the view from the other side. This is where the view from the Dark Side may offend some engineers. I present these notes/rants to help good engineers who want to become even better. You thinking through your design a bit further will make for a better design and improve the customer experience for all of us.

        Unfortunately there are some design engineers who, as Dave says, “just don’t get it”. What follows is my list of 16 rants – mainly experiences with engineers designing products for people working in the industrial automation market. I hope that we can all learn from these personal real world issues.

        Are you sitting down and ready to hear this stuff?

        At a tradeshow a marketer (me) was telling a customer about a brand new product’s features and the benefits to his organisation. One of our design engineers was there to get experience and to listen to customers. Unfortunately he interrupted the dialogue and proposed to the customer that we could redesign this Din rail mounted Ethernet switch to have the ports on the side. He promised that it could be done in two to three weeks! Ignoring a few facts like this design is already agreed by all stakeholders, is tested and is in production; the need for a PCB re-spin; the cost and delay of new plastics; the cost and delay of new agency certifications….

        A related point:
        Above all else, we sell the product that we have in hand today. It has cost a lot of money to produce. That helps us pay the salaries to develop the next generation of products. Getting a customer to wait for a potential future product does not help our cash-flow and salaries.

        A remote control designed to have over forty (40) buttons arranged in a grid of same shaped, same textured and same coloured buttons.

        Laser marking on enclosures that is tiny (3mm tall), low contrast (gray on white), 90 degrees to the user (causing user to tilt the head to read). Remember that your well lit bench top is a lot different to the average factory, steel foundry, mine etc.

        Making the user configure the product with subminiature rotary switches and associated pico-sized markings. Once again, no chance for customers over 30; or those working in dimly lit, dusty or dirty places.

        The design that presents error codes to the user as a series of long flash sequences on a LED. Picture a stressed user trying to differentiate between 13 or was it 14 flashes!

        Four levels of nested MS Windows menus in a fault tracking program. A configuration product that has five levels of nested MS Windows scroll bars!

        The Engineer’s response to a crashed module being “just press the reset button” doesn’t cut it when the rack is on top of a crane, in the roof of a sports arena or 10km away down a tunnel or a mine.

        A product design which has a possibility that following the procedure to upgrade firmware may cause the product to become a brick or paperweight.

        An industrial device designed in an enclosure that looks like an old clear 35mm film container, rattles and has the fragility of an eggshell.

        The remote terminal firmware coder who had decided to use the national currency codes in his programming, meaning that users could not use $, £, ¥ and €! Explain that to a bank!

        The firmware engineer who was stuck until he received a ‘marketing decision’ on the order of the bits in a status word.

        Tiny buttons spaced impossible to select singly with gloved hands.

        Connector blocks that require special, non-standard super skinny screwdrivers. Use of regular screwdrivers damages the plastic.

        Six months after agreeing to test all the product’s user features, test engineers wanting features removed as they don’t know how to test them.

        Engineering wanting an easy solution to a PCB population issue. Would we mind asking the customer to install a few components external to the enclosure? Oh yes, for each installation the user has to recalculate the resistor values. Then he has to source the resistors & capacitors and safely wire them up!

        And saving the best until last.
        At the very end of a development project, lead engineers who say that a “Marketing Decision is needed”. At this late minute, they state that they can’t develop a feature by the agreed deadline. The cause is never disclosed. It’s probably for one of these reasons – it’s too difficult to do; it’s boring work; doing this would impact R&D’s performance metrics; and there are no testing resources. The only options presented for Marketing’s choice are to de-scope (remove) the feature or slip the project more than six months!

        PS A final plea – Let’s all really THINK about the ergonomics for the end user!

        • Awesome, thanks for sharing!

      • Art Barnes

        You touched on an important point briefly – at the least, incremental upgrades are necessary. Manufacturers hate to change a beloved product, especially one where reliability is important. When this happens, customers will inevitably become angry and switch over to a competitor. However, the beloved product will eventually fall behind the competition otherwise. One example: how many simple, indestructible Nokia phones do you see these days? How many (yearly updated) Iphones do you see?

      • Roddy

        You engineers are lucky they do “upgrades”! Who needs an EEG if you don’t upgrade things. You should be worshipping the hand that feeds you, even though they are just keeping you around for future projects,

      The EEVblog Store generally ships twice a week, on Tuesdays & Fridays, Sydney time. Dismiss