Customer Complaints
Man, more and more customers are real unhappy with us these days. Thanks to shortsighted managers rushing new product designs out the door just so we can have "something" in the marketplace ahead of the competition we've sold a ton of unreliable, defective boxes. All we seem to care about is selling test gear and getting the customer's cash up front. Everything else that happens after that doesn't matter. We don't care about their feedback and we're not listening to valid issues they have been bringing up. I am one of the few employees here on the shop floor paying particular attention to what our customers have to say. Nobody else around here seems to be aware of any customer satisfaction issues or cares. It's fucked up. Here's a sampling of what my customers complain about. I've removed instrument specific information. All of these complaints concern RF/Microwave Signal Generators but isn't limited to them only. Currently I am aware of numerous complaints and general dissatisfaction across most of our product platforms.
For those of you who are our direct competitors, take note of the information here. There should be some useful items to take advantage of...
----------
Our customer, Kathrein, evaluated our XXXX and the Anritsu MG3692. He likes our specification, the display and the user interface but doesn’t like the fan noise. The fan is the decision point. The Field Engineer asks “Are there any plans to reduce the fan noise? Do you have any other idea?” Response: XXXX is aware of this issue on the XXXX but we do not have plans to fix this noise level.
*Short answer, no. Nobody cares about the fan noise and nothing will be done about it.
In the call center, we are constantly trying to improve customer satisfaction. We are being hampered by not having access to real-time factory support in this country and in our time zone. We have a lot of legacy products out in the field, especially at military accounts, and when they have a question or a problem, they usually need it resolved quickly. It doesn’t work to wait a few days to get a response back from Malaysia. We need a minimum viable level of service center support in XXXXXX and division support in XXXXXX. At the service center, we should at least be able to work on the products that are still in their support life. For division support, if it isn’t possible to get a live person in XXXXXX for our products that are manufactured in Malaysia, we at the very least need someone in Malaysia who works the same hours we do and speaks good English.
*Well that's what you dumbshits get for sending everything halfway around the world just so you can exploit the Malay as a cheap labor force.
From the SPO (Service Parts Organization) liaison between our customer’s and our suppliers: “Our customers and SPO are faced with many problems due to the company's decision to outsource so much to Contract Manufacturers. These problems have been going on since the end of last year when the XXXXXX to XXXXXX transfer began. It is my job to resolve the source of the customer dissatisfaction and get their repair part to them immediately. Not so easy now days. XXXXXX has provided unrealistic lead times for our downed customers and there is no good expedite process in place with the CMs regarding these customer requirements. Our customers’ only option is to wait the long lead times for their required parts. The TAT in the service centers will suffer. For on-site next day customer contracts, it will have an even greater impact. This company's end user customer will have an even worse experience with our customer support. My concern is that once the transfers have all been completed, XXXXXX will lose any kind of visibility to their customers’ plight. I, however, shall continue to hear their unhappy screams and there will be little I can do to resolve the issues any longer. This will be the hardest aspect of my job to accept."
*Hey. I have two words for you: Shareholder Value. When you outsource this much of the company's operation in a foolish attempt to save a few bucks and boost end of quarter earnings this is what happens. Enjoy, pal.
Nokia Networks and Mobile Phones have had many XXXXXX failures in a short amount of time ... no output signal, unlevel sign is on. Nokia Mobile Phones is introducing the XXXXXX into one of its production lines to cover W-LAN application. They are VERY concerned about the product quality as they have experienced already several failures of Output boards.
*Outsourcing, offshoring, rushing new designs to market without flushing out critical design flaws all add up to inferior products with high cost and low reliability. You've also probably got quite a few counterfeit components in your units. I'll get to that later...
Skyworks Solutions, Inc: "I have 4 XXXXXX ESG's (our company has 50+). We purchased option 405 (802.11b personality) and option 410 (802.11a personality). The 802.11b software produces a waveform that is an ABSOLUTE VIOLATION of the 802.11 standard. To address the problem, XXXXX charges $5000 for a software upgrade (option 417), when the software was broken to begin with. Furthermore, we can no longer even download the broken software for the options that we did pay for (we still use options 405 and 410 for some measurements). We would like to at least be able to continue to download the software for those options that we paid for, even if it is XXXXX policy to sell broken software and then charge for the correction (802.11g is the same modulation as 802.11a, the fact that Option 417 supports 802.11g does not mean it supports a new modulation format additional to Options 405 and 410)."
RenTelco: "I have an XXXXX Synthesized Swept-CW Generator, 10 MHz to 26.5 GHz. What is the specific product that is recommended as a replacement for this unit? We are a rental company and when a customer wants to rent an XXXXX, I need to know exactly what XXXXX recommends as the replacement. I don't want to be sent to a "XXXX" page, I want it to be a specific model number and take me to that product page. I need it to be streamlined, linear, specific. I want to know what XXXXX says is the recommended replacement and that's what we'll put into our database. My sales people need to know specifically what to order for the replacement."
*Our customer-facing external website fucking sucks. It's so messed up that we can't even use it internally to find information on many of our older products. Many of us share your frustration but until the people in charge of our website pull their heads out of their asses this is going to continue to be a real problem.
In April, Telogy tried to order the XXXXX (Modulation Generator) they were told was available ... only to find out Oracle was showing it as not orderable. After telling the customer it was not orderable, we fixed the system and on May 21st customer sent his unit to Malaysia. The unit shipped back to him on June 23rd and was received with no AM modulation. They asked that the unit be repaired within two weeks to meet the needs of their customer. Unit was sent to XXXXXX on July 2nd. A second part needed to be ordered when the first part was received defective. The second board appeared defective as well. On August 2nd, the problem was found to be the motherboard... the option boards were okay. The unit was shipped to the customer on August 9th and is working, however, Telogy is still upset this has taken so long and that they lost a customer in the process.
Ascom Tateco AB placed an order in December for an XXXXX. For several (~7) weeks, there was confusion over which factory builds these upgrades. When the upgrade shipped on February 28th, we shipped 2 units, customer ordered only 1 unit. The credit for the second one is not released yet but hopefully we will do that very soon! So, from a customer perspective this has not been a good case, 9 months from order (dec) to final credit (aug/sept).
*Absolutely pathetic.
"I rely on the XXXXX Phase Noise performance > 100 kHz in the PSA cal software. I learned of the XXXXX discontinuance when one of my customers asked what source my future plans were going to use. After looking into the XXXX performance I realize the XXXX isn't a replacement for the Noise Sideband performance of the XXXXX. I have talked to many self maintainers....Casey (Rockwell Collins), James (USAF), Jeffrey (USAFMETCAL)....that have voiced the same concerns."
USAF: "The XXXXX website sends you over to the XXXX webpage. The XXXX doesn't have the Noise Sidebands performance (160 dBc/Hz typical performance @ 1 MHz to 10 MHz offset) that the XXXXX has. The XXXX isn't an XXXXX replacement. Are you working on a XXXX option that will perform as well as the XXXXX?"
*A recent trend here is we have been discontinuing our high performance legacy products and replacing them with inferior equivalent units that don't have the performance or tight specifications. Oh and our customer facing website is pure ass. Sorry 'bout that.
"I would like to use the XXXXXX to test our 50 GHz Wide Bandwith Option (Opt 122) scheduled for fall '05. The source would be used to verify our Wide Bandwidth IF Phase Linearity specification above 20 GHz. The >$200k price is going to prevent me from specing this source into the PSA cal software ETE. We will use the XXXXX and only test to 20 GHz."
*That's our marketing department having fun with price gouging.
RF Micro Devices requested a quote for Signal Studio for HSDPA. Between the time they were quoted a price (March) and when they placed the order (August) significant enhancements were made to the software and the price was increased during the semi-annual pricing round (May). Customer: "A $970 increase ??? That's a 20% jump. How do you justify that? Why has the quote gone up so much in such a short time? This may be unacceptable. I will have to check on this."
*More price gouging.
Third complaint in 5 months about the loud fan noise in the XXXX.
*Yeah, nobody is going to do a damn thing about it, either.
A need for improved security erase procedures in the XXXX from Northrop-Grumman, Lockheed-Martin, and Raytheon.
*Scrub procedures in our boxes are an overcomplicated hassle and in order to remove the hard drive you have to take the unit apart just to get at it. Same goes for CPU boards with sensitive data stored in memory. I agree this is a problem, but from what I have learned about the situation management has looked into this issue and flat out refused to change designs to include an easily removable disc drive or simplify the scrub procedures.
Concern from several groups in Lockheed Martin to PLEASE stop changing and obsoleting the model #s for the XXXX - this group in Owego, NY has purchased many XXXX and integrated them into systems for specific programs. All the support documentation has to be revised each time we obsolete a product. They are getting frustrated and have informed their FE that each time this happens it also allows them to easily switch vendors. All our competitors are focused on the defense companies like Lockheed - we need to try and reduce obstacles. He was less stressed when I (XXXXXX FE) explained his Cs were being rev'd to Ds - but then we got into the other XXXX. In the last couple of years they have purchased our latest and best Sig Gens. For basically the same XXXX CW generator they will have 3 different model #s in 3.5 years A-C-D. This gets expensive and frustrating for Lockheed to constantly update support packages and reduces our leverage of using common products across multiple programs when the model #s keep changing. Revising a suffix is much better than changing the model # ex A to B or A-B-C - is much less painful for our customers and also builds product loyalty. The XXXXXX has been the gold standard for almost 20 years - that is about the life of many of Lockheed's programs. We have trained our customers to expect stability and quality from this company - we have struggled the last few years.
*Marketing is always trying to push something that's "new" you know, it's that 1950s mentality that if you can tell customers you've got something "new" they'll probably buy it. It's irrelivant if it is actually an improvement over previous models. They're counting on this lame ass strategy, and I'd like to add the models you've already bought from us were hobbled versions of the full-fledged box. Things were running so far behind schedule in R&D that the decision was made to sell what we had at the time just to have something available for purchase. As soon as the design was more complete they obsoleted the previous model.
Anaren Microwave bought several XXXXXX generators in May. Now in September they need two more, and we've announced it will be discontinued on Nov. 1. Problem is support life. Anaren uses their generators for a long time. The C is being replaced with the D, which in their configuration is exactly identical. Trouble is, the C's they bought a few months ago will be out of support in 5 years! Models turn over too quickly these days to leave it at 5 years. In this case, the C should be supported longer than 5 years because the replacement is functionally identical.
Defence Account Manager in Australia: I'm figuring that the XXXXXX is still about 80% identical in hardware to the XXXXX, so why are we treating this like the complete death of a product and the customer has to figure out what he wants to do with it after 5 years! I've got to walk into a meeting and try to convince them that they don't need to go back out to competitive tender again. That the new model is virtually identical and only has a 'D' on the end of it instead of a 'C'. But if I have to tell them that we still don't really have a support plan for it, then we better be prepared to just take back all 23 instruments that they've bought in the last 18 months (19 earlier this year) and hand them new XXXXXX!
Mark, FE: "Rockwell Collins is looking at buying 2 of these software options (Enhanced Multitone and NPR) now and maybe 6 or 8 in the future. However, because of the price of each of the personalities, they are going through an engineering effort to figure out why NOT to buy NPR and that's not the mode we should have our customers in because of our price. We need to respond and respond shortly. It's been over a year since this has been under investigation and frankly, if we're going to be agile, we need to be agile -- and that's making decisions where it really makes a lot of sense. I don't know what the hang up is on offering a 2 for 1 1/2; that is, buy one and get the second one at half price. I think that would make a very good offer and not really cut into our profit line. It's cutting into our profit line now when they're going off and figuring out how not to use our software options. I need an answer and I think the division ought to get off the pot and make a decision on when customers want to buy more than one very expensive software option that we bucket these into a palatable solution that doesn't prevent sales, but actually enhances sales."
*Asleep at the wheel, no one paying attention to customer needs, price gouging, and only give a marginal discount when a customer complains loudly enough. Doesn't make a whole lot of common sense from a business standpoint to keep operating in this manner.
Comments about delivery requirements, product quality, repair turnaround-time and cost received from Flextronics, Southeast University, Nokia, RF Micro Devices, and Huawei Technologies during customer visits.
A request from Northrop Grumman to forward a letter to the management team expressing their outrage at "The company's lack of sensitivity for configuration control regarding the C change to a D. Northrop Grumman ES Test Systems is not a casual consumer of an instrument that sits isolated in some lab. The instruments we utilize, and I believe that was over $1,000,000 on this contract alone, are part of Automatic Test Equipment (ATE) used to sell product to our customers. As such we have drawings that call out all of our hardware. We have a Component Management System which tracks the part numbers, we have a Master Parts List which stores our part numbers, and we have SAP which is used to procure the hardware. All of these systems are affected when a part number changes and requires a revision notice to be generated to initiate this change and engineering time, and is processed by numerous other personnel which costs money. Not only does this affect the C, but it also results in a change to the C. WE HAVE ONLY RECEIVED DELIVERY OF THE C's SOME TIME LATE OCTOBER. THIS MEANS THAT THE ATE IS OBSOLETE BEFORE IT WAS EVEN ASSEMBLED. What I would like is for XXXXXX to take back all of the Cs and all of the XXXXCs and upgrade them to Ds at no cost to Northrop Grumman."
*These kinds of situations should never have happened in the first place. Inexcuseable.
US Navy reported they did not receive their option 420 (pulse building) software with their new XXXXC. "We did receive the signal generator, but the associated software was not in the box. The paperwork that came with the shipment listed only one box in the shipment, but was missing the C-420 Signal Studio for pulse building component." After letting him know he only receives a page letter inside an envelope with instructions for downloading the software from the web, he found the information he needed. "I am a little perturbed that for a price tag of $15,000.00 I had to download the system interface software, burn it to the appropriate media, download and print the manuals, and retrieve the host ID and license keyword from the Signal Generator in a very convoluted process. I assumed for $15,000.00 the company would have included a CD-ROM and Manual in the box."
*This is a totally unnecessary, fouled up process that many customers frequently complain about. For whatever reasons no one is taking a serious look at the way we provide software licensing to make it streamlined and easier for customers to use. Right now when you purchase software options all you get in the box is a stupid form letter telling you to go download the items. But what if the download system isn't working properly or what if you happen to work in a secure environment where outside internet access is prohibited? You're out of luck. Maybe someone in the company will eventually mail you a CD rom with an outdated version of your software options but you will more than likely have to complain for a long time before anyone is going to respond.
Similar to the feedback received last month from the US Navy and other reported OBDs, Northrop Grumman (D-408) and LG Electronics (A-170,160,168) reported they did not receive the new software they ordered. Response from Northrop Grumman: "My initial asssement is that putting the item in an appropriately marked box is more likely to result in delivery than an envelope. Alternatively a CD, even if obsolete seems to get through and having a notice on it to go to your web site for the latest works for me. Additionally, I would note that receiving has become a difficult operation these days when shipped options are buried in the units and are less than tangible to a non-engineer. Perhaps you need to enclose very specific directions to the receiver as to how to evaluate the contents of the shipment."
*See my comment above.
Rockwell Collins and Skyworks are having major problems creating and downloading files (waveform and data files) to the XXXX/XXXX. Our documentation (programming manuals, applications notes and download utilities) are not useful for many customers as they assume a higher level of programming knowledge than many customers have. The Contact Center reports there has been a significant increase over the past year in the number of customers calling with these download problems.
Navy Mid-Atlantic bought the C version (20 and 40GHz) of the XXXX to use as a drop in replacement for the XXXXX signal generator working together with a D scalar. In this combination it isn't possible to make good D scalar measurements in AC mode below 3.2Ghz. The customer is working on the Navy calibration lab and needs to do high quality load measurements with the XXXX/Scalar combination. The customer is very openly communicating his statement right now, "The XXXX is a piece of crap below 3.2GHz." The customer is in contact with the cal labs of the Army and the Airforce and will not hesitate to inform these departments about his experiences with the XXXX.
Lockheed Martin in Denver is experiencing numerous firmware bugs which cause the XXXX to go out of spec. This is a very serious customer satisfaction issue which the customer believes has cost them "tons of time, money and credibility." The four issues submitted to the firmware team were: 1) enhancements to gain cal for ALC MOD GAIN 3 with I/Q on; 2) Unleveled Output Power in Auto power search with LB filters on; 3) Power search span mode has 0.75 dB error with IQ on; 4) Need error reported if power search with RF Off.
Our customer at ITT Industries sent an email to his Field Engineer, "I thought you'd like to know we've found problems with the signal generators (2 XXXXC) we have here and the XXXXA freq counter. When you add these issues to the XXXXXXA issue, then the quality level at your company is not looking too good. It makes it very difficult to take calibrated measurements when you have no faith in the test equipment. Both signal generators exhibit broadband spikes of several db and sometimes the CF spikes by 6 db. This happens when stepping the amplitude in most step sizes and seems to occur in spacings of 10 db for both signal generators. It happens at +2, -8, -18dbm. We didn't try more settings." We suggested the customer upgrade to firmware revision CXXXX6 or greater which eliminated power spikes when switching the mechanical attenuator. He was unable to do that immediately because he had sent the unit in for repair and it had been at our service center for several weeks. "When and if it ever comes back, I'll let you know." The unit was not returned from XXXX when expected. "We just received notice that our SG that was broken has now been rescheduled for 6-13. This will mean that your company has hijacked our equipment for 60 days. This is completely unacceptable! We need to understand why your company needs 60 days to repair and calibrate a piece of your own equipment." An escalation to the Service Center resulted in the return of their unit. XXXX product support is waiting to receive information from the customer to further investigate the issue as "the one (XXXXC) we received back recently still has the original problem. We are putting together a package of screen shots and firmware information. Can you please ensure that the one that is in calibration currently has this problem fixed before it is sent back. You can easily check for the problem by connecting the output of the SG to a SA in max hold and then step the SG power level."
Lucent/Celestica: "There are so many units of XXXX with opt. H99 installed in my production line. Most have the same failure i.e. 'unlevelled O/P' which needs to be sent to your company for repair. I was informed that this failure symptom is covered under service note warranty. That means your company is supposed to know that this special option must have problems and may cause failure to the instruments. Your company should recall back the units that have potential to get affected by this failure and replace with the new units -- or -- there should not be the expiration date for service note warranty as this is your company's fault not the customer's fault."
*You caught us, and you are entirely correct. The output board fails routinely on boxes with option H99. They are of a totally poor design and are known to be defective (bad HFETs, bad circuit design). We sold them anyway. There is a long story behind the H99 failures which I hope to get into shortly. Basically if you're not a big customer like Motorola or Boeing, the attitude is we don't care about you. We got your loot and as far as warranty service goes we don't stand behind our workmanship like we used to. Big customers such as Motorola also caught us on this one... and in that case management created a special "back door" service policy to repair their units and keep them happy. The rest of you poor bastards were left out in the cold while we gave you a big bony middle finger salute. For that I am truly disgusted, and sorry. We know better, and we used to take care of you guys.
In April of 1986 we made the following announcement concerning our warranty/service support:
Announcing Ten-Year Minimum Support
A further definition of lasting value
Lasting value is an integral part of every XXXXX product. Lasting value means enduring quality in both hardware and software as well as in strategies to protect these investments. It also means support and service to help you get the most out of the products and systems throughout their useful lives. Now XXXXX has extended it's support commitment to a minimum of ten years after active manufacture of the product. This commitment inculdes the availability of parts, maintenance services, training, useage assistance, and a whole host of other programs designed to ensure your access to the help you need in using your XXXXX products effectively. What's more, this backup is only a minimum. In many special situations, we will continue to offer support beyond this time period. This continuing effort is an extension of our larger commitment to an ongoing partnership. We stand behind our products with support you can count on- for the long run.
Today, twenty years later all of that has gone clear out the window. When you buy one of our brand new test instruments the standard warranty period is only one year. One. If you want warranty coverage longer than that, you have to pay through the nose for it. Marketing doesn't want you to continue using our legacy products over long periods of time. They want you to buy whatever latest and greatest high cost instruments we've got instead of maintaining your vintage gear.
Tucker Electronics: "There seems to be 2 issues every time we order an option where a key code number can be sent to us via email to open up the programming capabilities of the unit. The first issue is that it usually takes up to 2 or 3 days to get the original order entered and a confirmation number sent to me. The second issue is that it takes a couple of days after that to receive the email with the key code number. We are quoted 2 weeks for the delivery time for the option and we know well that the key code will come sooner via email ... It just seem that when we order 4x4 or 5x5 parts we never have these issues. We place the order, a confirmation number is given at the end of the conversation and the parts ship the same day. Is there a simple explanation?"
*Sure there's a simple explanation! Nobody on our end is paying attention. Weeeeehooooo!
Motorola received the entitlement certificate for his (2) Baseband Studio for Fading order. He was surprised that it arrived in an anti-static bag "why did it come in this type of bag? what a waste of money!" He was confused when he received two copies of the entitlement certificate There were no instructions on the certificate for what to do after retrieving the license file. No indication what to do next or how to obtain the software.
*Yep. We rock.
More customer reaction on the recent expiration (May) of the H99 service note which covered all H99 repairs under extended warranty. NEC Wireless Networks had 6 XXXXX/H99 failures at the same time they were ramping up production after a 1 week factory shut down. The customer still has reliability issues on their 40 units of H99 even after we provided the proactive corrective actions to improve the reliability. They are experiencing 1-2 units out of 40 broken every 3 months. NEC totally disagrees to pay any repair cost for their H99 because they have a strong concern about "high failure rate of H99" and "owner cost" while under cost reduction pressure on their products. Anritsu is pushing the MG3700A to this customer to replace their XXXXXBs and the customer is interested due to cost. According to the FE, "This is our biggest threat in the near future". They are asking XXXXX to provide extended warranty for H99 repair cost and material until the end of support.
*See my above comments about the H99 debacle. Warranty service on these defective output boards was extended, but the redesigns are not much better than the original boards. They won't last long even after repair with an upgraded output board.
Feedback from FEs in Japan about XXXXXC quality issues ... 10 out of 20 failures at Kyocera (unleveled, hard disk failures, download problems). Other customers are experiencing this as well. Two dead units recently for "no modulation" and many H99 units become unleveled.
An exceptionally high XXXXX failure rate (over 100% among their 15 As!) reported by the FE for our customer, Leihua. One unit even failed 7 times!!! One failed 4 times and another 3 times. The biggest problem they are facing is when those XXXXXs reach their EOS (end of support) date next year and fail again, they will have to discard them even though some of these boxes are just ~4 years old! Please give the customer a way out! Those failures are hurting the customers and our company a lot! We already begin to lose SG deals to Anritsu in this account and there's almost nothing we could do.
*Worldwide in the first year of service we've observed a 28%-30% fail rate in these products. Normally when we release a new instrument platform for sale our worldwide failure rate target for the first year of service is 3%-5%, with 5% being totally unacceptable. A major factor in the low reliability of these units is that they have surface mounted tantalum capacitors that are either counterfeit or underrated on voltage for the circuit. In either case they will usually overheat, grenade off the board and crater it rendering the PC board useless. This is a collective result of rushing products through the R&D phase, cutting corners, outsourcing, and offshoring our manufacturing. Can you say Shareholder Value? C'mon now and say it with Supergeek like you mean it... Shareholder Value! Yeah!
For those of you who are our direct competitors, take note of the information here. There should be some useful items to take advantage of...
----------
Our customer, Kathrein, evaluated our XXXX and the Anritsu MG3692. He likes our specification, the display and the user interface but doesn’t like the fan noise. The fan is the decision point. The Field Engineer asks “Are there any plans to reduce the fan noise? Do you have any other idea?” Response: XXXX is aware of this issue on the XXXX but we do not have plans to fix this noise level.
*Short answer, no. Nobody cares about the fan noise and nothing will be done about it.
In the call center, we are constantly trying to improve customer satisfaction. We are being hampered by not having access to real-time factory support in this country and in our time zone. We have a lot of legacy products out in the field, especially at military accounts, and when they have a question or a problem, they usually need it resolved quickly. It doesn’t work to wait a few days to get a response back from Malaysia. We need a minimum viable level of service center support in XXXXXX and division support in XXXXXX. At the service center, we should at least be able to work on the products that are still in their support life. For division support, if it isn’t possible to get a live person in XXXXXX for our products that are manufactured in Malaysia, we at the very least need someone in Malaysia who works the same hours we do and speaks good English.
*Well that's what you dumbshits get for sending everything halfway around the world just so you can exploit the Malay as a cheap labor force.
From the SPO (Service Parts Organization) liaison between our customer’s and our suppliers: “Our customers and SPO are faced with many problems due to the company's decision to outsource so much to Contract Manufacturers. These problems have been going on since the end of last year when the XXXXXX to XXXXXX transfer began. It is my job to resolve the source of the customer dissatisfaction and get their repair part to them immediately. Not so easy now days. XXXXXX has provided unrealistic lead times for our downed customers and there is no good expedite process in place with the CMs regarding these customer requirements. Our customers’ only option is to wait the long lead times for their required parts. The TAT in the service centers will suffer. For on-site next day customer contracts, it will have an even greater impact. This company's end user customer will have an even worse experience with our customer support. My concern is that once the transfers have all been completed, XXXXXX will lose any kind of visibility to their customers’ plight. I, however, shall continue to hear their unhappy screams and there will be little I can do to resolve the issues any longer. This will be the hardest aspect of my job to accept."
*Hey. I have two words for you: Shareholder Value. When you outsource this much of the company's operation in a foolish attempt to save a few bucks and boost end of quarter earnings this is what happens. Enjoy, pal.
Nokia Networks and Mobile Phones have had many XXXXXX failures in a short amount of time ... no output signal, unlevel sign is on. Nokia Mobile Phones is introducing the XXXXXX into one of its production lines to cover W-LAN application. They are VERY concerned about the product quality as they have experienced already several failures of Output boards.
*Outsourcing, offshoring, rushing new designs to market without flushing out critical design flaws all add up to inferior products with high cost and low reliability. You've also probably got quite a few counterfeit components in your units. I'll get to that later...
Skyworks Solutions, Inc: "I have 4 XXXXXX ESG's (our company has 50+). We purchased option 405 (802.11b personality) and option 410 (802.11a personality). The 802.11b software produces a waveform that is an ABSOLUTE VIOLATION of the 802.11 standard. To address the problem, XXXXX charges $5000 for a software upgrade (option 417), when the software was broken to begin with. Furthermore, we can no longer even download the broken software for the options that we did pay for (we still use options 405 and 410 for some measurements). We would like to at least be able to continue to download the software for those options that we paid for, even if it is XXXXX policy to sell broken software and then charge for the correction (802.11g is the same modulation as 802.11a, the fact that Option 417 supports 802.11g does not mean it supports a new modulation format additional to Options 405 and 410)."
RenTelco: "I have an XXXXX Synthesized Swept-CW Generator, 10 MHz to 26.5 GHz. What is the specific product that is recommended as a replacement for this unit? We are a rental company and when a customer wants to rent an XXXXX, I need to know exactly what XXXXX recommends as the replacement. I don't want to be sent to a "XXXX" page, I want it to be a specific model number and take me to that product page. I need it to be streamlined, linear, specific. I want to know what XXXXX says is the recommended replacement and that's what we'll put into our database. My sales people need to know specifically what to order for the replacement."
*Our customer-facing external website fucking sucks. It's so messed up that we can't even use it internally to find information on many of our older products. Many of us share your frustration but until the people in charge of our website pull their heads out of their asses this is going to continue to be a real problem.
In April, Telogy tried to order the XXXXX (Modulation Generator) they were told was available ... only to find out Oracle was showing it as not orderable. After telling the customer it was not orderable, we fixed the system and on May 21st customer sent his unit to Malaysia. The unit shipped back to him on June 23rd and was received with no AM modulation. They asked that the unit be repaired within two weeks to meet the needs of their customer. Unit was sent to XXXXXX on July 2nd. A second part needed to be ordered when the first part was received defective. The second board appeared defective as well. On August 2nd, the problem was found to be the motherboard... the option boards were okay. The unit was shipped to the customer on August 9th and is working, however, Telogy is still upset this has taken so long and that they lost a customer in the process.
Ascom Tateco AB placed an order in December for an XXXXX. For several (~7) weeks, there was confusion over which factory builds these upgrades. When the upgrade shipped on February 28th, we shipped 2 units, customer ordered only 1 unit. The credit for the second one is not released yet but hopefully we will do that very soon! So, from a customer perspective this has not been a good case, 9 months from order (dec) to final credit (aug/sept).
*Absolutely pathetic.
"I rely on the XXXXX Phase Noise performance > 100 kHz in the PSA cal software. I learned of the XXXXX discontinuance when one of my customers asked what source my future plans were going to use. After looking into the XXXX performance I realize the XXXX isn't a replacement for the Noise Sideband performance of the XXXXX. I have talked to many self maintainers....Casey (Rockwell Collins), James (USAF), Jeffrey (USAFMETCAL)....that have voiced the same concerns."
USAF: "The XXXXX website sends you over to the XXXX webpage. The XXXX doesn't have the Noise Sidebands performance (160 dBc/Hz typical performance @ 1 MHz to 10 MHz offset) that the XXXXX has. The XXXX isn't an XXXXX replacement. Are you working on a XXXX option that will perform as well as the XXXXX?"
*A recent trend here is we have been discontinuing our high performance legacy products and replacing them with inferior equivalent units that don't have the performance or tight specifications. Oh and our customer facing website is pure ass. Sorry 'bout that.
"I would like to use the XXXXXX to test our 50 GHz Wide Bandwith Option (Opt 122) scheduled for fall '05. The source would be used to verify our Wide Bandwidth IF Phase Linearity specification above 20 GHz. The >$200k price is going to prevent me from specing this source into the PSA cal software ETE. We will use the XXXXX and only test to 20 GHz."
*That's our marketing department having fun with price gouging.
RF Micro Devices requested a quote for Signal Studio for HSDPA. Between the time they were quoted a price (March) and when they placed the order (August) significant enhancements were made to the software and the price was increased during the semi-annual pricing round (May). Customer: "A $970 increase ??? That's a 20% jump. How do you justify that? Why has the quote gone up so much in such a short time? This may be unacceptable. I will have to check on this."
*More price gouging.
Third complaint in 5 months about the loud fan noise in the XXXX.
*Yeah, nobody is going to do a damn thing about it, either.
A need for improved security erase procedures in the XXXX from Northrop-Grumman, Lockheed-Martin, and Raytheon.
*Scrub procedures in our boxes are an overcomplicated hassle and in order to remove the hard drive you have to take the unit apart just to get at it. Same goes for CPU boards with sensitive data stored in memory. I agree this is a problem, but from what I have learned about the situation management has looked into this issue and flat out refused to change designs to include an easily removable disc drive or simplify the scrub procedures.
Concern from several groups in Lockheed Martin to PLEASE stop changing and obsoleting the model #s for the XXXX - this group in Owego, NY has purchased many XXXX and integrated them into systems for specific programs. All the support documentation has to be revised each time we obsolete a product. They are getting frustrated and have informed their FE that each time this happens it also allows them to easily switch vendors. All our competitors are focused on the defense companies like Lockheed - we need to try and reduce obstacles. He was less stressed when I (XXXXXX FE) explained his Cs were being rev'd to Ds - but then we got into the other XXXX. In the last couple of years they have purchased our latest and best Sig Gens. For basically the same XXXX CW generator they will have 3 different model #s in 3.5 years A-C-D. This gets expensive and frustrating for Lockheed to constantly update support packages and reduces our leverage of using common products across multiple programs when the model #s keep changing. Revising a suffix is much better than changing the model # ex A to B or A-B-C - is much less painful for our customers and also builds product loyalty. The XXXXXX has been the gold standard for almost 20 years - that is about the life of many of Lockheed's programs. We have trained our customers to expect stability and quality from this company - we have struggled the last few years.
*Marketing is always trying to push something that's "new" you know, it's that 1950s mentality that if you can tell customers you've got something "new" they'll probably buy it. It's irrelivant if it is actually an improvement over previous models. They're counting on this lame ass strategy, and I'd like to add the models you've already bought from us were hobbled versions of the full-fledged box. Things were running so far behind schedule in R&D that the decision was made to sell what we had at the time just to have something available for purchase. As soon as the design was more complete they obsoleted the previous model.
Anaren Microwave bought several XXXXXX generators in May. Now in September they need two more, and we've announced it will be discontinued on Nov. 1. Problem is support life. Anaren uses their generators for a long time. The C is being replaced with the D, which in their configuration is exactly identical. Trouble is, the C's they bought a few months ago will be out of support in 5 years! Models turn over too quickly these days to leave it at 5 years. In this case, the C should be supported longer than 5 years because the replacement is functionally identical.
Defence Account Manager in Australia: I'm figuring that the XXXXXX is still about 80% identical in hardware to the XXXXX, so why are we treating this like the complete death of a product and the customer has to figure out what he wants to do with it after 5 years! I've got to walk into a meeting and try to convince them that they don't need to go back out to competitive tender again. That the new model is virtually identical and only has a 'D' on the end of it instead of a 'C'. But if I have to tell them that we still don't really have a support plan for it, then we better be prepared to just take back all 23 instruments that they've bought in the last 18 months (19 earlier this year) and hand them new XXXXXX!
Mark, FE: "Rockwell Collins is looking at buying 2 of these software options (Enhanced Multitone and NPR) now and maybe 6 or 8 in the future. However, because of the price of each of the personalities, they are going through an engineering effort to figure out why NOT to buy NPR and that's not the mode we should have our customers in because of our price. We need to respond and respond shortly. It's been over a year since this has been under investigation and frankly, if we're going to be agile, we need to be agile -- and that's making decisions where it really makes a lot of sense. I don't know what the hang up is on offering a 2 for 1 1/2; that is, buy one and get the second one at half price. I think that would make a very good offer and not really cut into our profit line. It's cutting into our profit line now when they're going off and figuring out how not to use our software options. I need an answer and I think the division ought to get off the pot and make a decision on when customers want to buy more than one very expensive software option that we bucket these into a palatable solution that doesn't prevent sales, but actually enhances sales."
*Asleep at the wheel, no one paying attention to customer needs, price gouging, and only give a marginal discount when a customer complains loudly enough. Doesn't make a whole lot of common sense from a business standpoint to keep operating in this manner.
Comments about delivery requirements, product quality, repair turnaround-time and cost received from Flextronics, Southeast University, Nokia, RF Micro Devices, and Huawei Technologies during customer visits.
A request from Northrop Grumman to forward a letter to the management team expressing their outrage at "The company's lack of sensitivity for configuration control regarding the C change to a D. Northrop Grumman ES Test Systems is not a casual consumer of an instrument that sits isolated in some lab. The instruments we utilize, and I believe that was over $1,000,000 on this contract alone, are part of Automatic Test Equipment (ATE) used to sell product to our customers. As such we have drawings that call out all of our hardware. We have a Component Management System which tracks the part numbers, we have a Master Parts List which stores our part numbers, and we have SAP which is used to procure the hardware. All of these systems are affected when a part number changes and requires a revision notice to be generated to initiate this change and engineering time, and is processed by numerous other personnel which costs money. Not only does this affect the C, but it also results in a change to the C. WE HAVE ONLY RECEIVED DELIVERY OF THE C's SOME TIME LATE OCTOBER. THIS MEANS THAT THE ATE IS OBSOLETE BEFORE IT WAS EVEN ASSEMBLED. What I would like is for XXXXXX to take back all of the Cs and all of the XXXXCs and upgrade them to Ds at no cost to Northrop Grumman."
*These kinds of situations should never have happened in the first place. Inexcuseable.
US Navy reported they did not receive their option 420 (pulse building) software with their new XXXXC. "We did receive the signal generator, but the associated software was not in the box. The paperwork that came with the shipment listed only one box in the shipment, but was missing the C-420 Signal Studio for pulse building component." After letting him know he only receives a page letter inside an envelope with instructions for downloading the software from the web, he found the information he needed. "I am a little perturbed that for a price tag of $15,000.00 I had to download the system interface software, burn it to the appropriate media, download and print the manuals, and retrieve the host ID and license keyword from the Signal Generator in a very convoluted process. I assumed for $15,000.00 the company would have included a CD-ROM and Manual in the box."
*This is a totally unnecessary, fouled up process that many customers frequently complain about. For whatever reasons no one is taking a serious look at the way we provide software licensing to make it streamlined and easier for customers to use. Right now when you purchase software options all you get in the box is a stupid form letter telling you to go download the items. But what if the download system isn't working properly or what if you happen to work in a secure environment where outside internet access is prohibited? You're out of luck. Maybe someone in the company will eventually mail you a CD rom with an outdated version of your software options but you will more than likely have to complain for a long time before anyone is going to respond.
Similar to the feedback received last month from the US Navy and other reported OBDs, Northrop Grumman (D-408) and LG Electronics (A-170,160,168) reported they did not receive the new software they ordered. Response from Northrop Grumman: "My initial asssement is that putting the item in an appropriately marked box is more likely to result in delivery than an envelope. Alternatively a CD, even if obsolete seems to get through and having a notice on it to go to your web site for the latest works for me. Additionally, I would note that receiving has become a difficult operation these days when shipped options are buried in the units and are less than tangible to a non-engineer. Perhaps you need to enclose very specific directions to the receiver as to how to evaluate the contents of the shipment."
*See my comment above.
Rockwell Collins and Skyworks are having major problems creating and downloading files (waveform and data files) to the XXXX/XXXX. Our documentation (programming manuals, applications notes and download utilities) are not useful for many customers as they assume a higher level of programming knowledge than many customers have. The Contact Center reports there has been a significant increase over the past year in the number of customers calling with these download problems.
Navy Mid-Atlantic bought the C version (20 and 40GHz) of the XXXX to use as a drop in replacement for the XXXXX signal generator working together with a D scalar. In this combination it isn't possible to make good D scalar measurements in AC mode below 3.2Ghz. The customer is working on the Navy calibration lab and needs to do high quality load measurements with the XXXX/Scalar combination. The customer is very openly communicating his statement right now, "The XXXX is a piece of crap below 3.2GHz." The customer is in contact with the cal labs of the Army and the Airforce and will not hesitate to inform these departments about his experiences with the XXXX.
Lockheed Martin in Denver is experiencing numerous firmware bugs which cause the XXXX to go out of spec. This is a very serious customer satisfaction issue which the customer believes has cost them "tons of time, money and credibility." The four issues submitted to the firmware team were: 1) enhancements to gain cal for ALC MOD GAIN 3 with I/Q on; 2) Unleveled Output Power in Auto power search with LB filters on; 3) Power search span mode has 0.75 dB error with IQ on; 4) Need error reported if power search with RF Off.
Our customer at ITT Industries sent an email to his Field Engineer, "I thought you'd like to know we've found problems with the signal generators (2 XXXXC) we have here and the XXXXA freq counter. When you add these issues to the XXXXXXA issue, then the quality level at your company is not looking too good. It makes it very difficult to take calibrated measurements when you have no faith in the test equipment. Both signal generators exhibit broadband spikes of several db and sometimes the CF spikes by 6 db. This happens when stepping the amplitude in most step sizes and seems to occur in spacings of 10 db for both signal generators. It happens at +2, -8, -18dbm. We didn't try more settings." We suggested the customer upgrade to firmware revision CXXXX6 or greater which eliminated power spikes when switching the mechanical attenuator. He was unable to do that immediately because he had sent the unit in for repair and it had been at our service center for several weeks. "When and if it ever comes back, I'll let you know." The unit was not returned from XXXX when expected. "We just received notice that our SG that was broken has now been rescheduled for 6-13. This will mean that your company has hijacked our equipment for 60 days. This is completely unacceptable! We need to understand why your company needs 60 days to repair and calibrate a piece of your own equipment." An escalation to the Service Center resulted in the return of their unit. XXXX product support is waiting to receive information from the customer to further investigate the issue as "the one (XXXXC) we received back recently still has the original problem. We are putting together a package of screen shots and firmware information. Can you please ensure that the one that is in calibration currently has this problem fixed before it is sent back. You can easily check for the problem by connecting the output of the SG to a SA in max hold and then step the SG power level."
Lucent/Celestica: "There are so many units of XXXX with opt. H99 installed in my production line. Most have the same failure i.e. 'unlevelled O/P' which needs to be sent to your company for repair. I was informed that this failure symptom is covered under service note warranty. That means your company is supposed to know that this special option must have problems and may cause failure to the instruments. Your company should recall back the units that have potential to get affected by this failure and replace with the new units -- or -- there should not be the expiration date for service note warranty as this is your company's fault not the customer's fault."
*You caught us, and you are entirely correct. The output board fails routinely on boxes with option H99. They are of a totally poor design and are known to be defective (bad HFETs, bad circuit design). We sold them anyway. There is a long story behind the H99 failures which I hope to get into shortly. Basically if you're not a big customer like Motorola or Boeing, the attitude is we don't care about you. We got your loot and as far as warranty service goes we don't stand behind our workmanship like we used to. Big customers such as Motorola also caught us on this one... and in that case management created a special "back door" service policy to repair their units and keep them happy. The rest of you poor bastards were left out in the cold while we gave you a big bony middle finger salute. For that I am truly disgusted, and sorry. We know better, and we used to take care of you guys.
In April of 1986 we made the following announcement concerning our warranty/service support:
Announcing Ten-Year Minimum Support
A further definition of lasting value
Lasting value is an integral part of every XXXXX product. Lasting value means enduring quality in both hardware and software as well as in strategies to protect these investments. It also means support and service to help you get the most out of the products and systems throughout their useful lives. Now XXXXX has extended it's support commitment to a minimum of ten years after active manufacture of the product. This commitment inculdes the availability of parts, maintenance services, training, useage assistance, and a whole host of other programs designed to ensure your access to the help you need in using your XXXXX products effectively. What's more, this backup is only a minimum. In many special situations, we will continue to offer support beyond this time period. This continuing effort is an extension of our larger commitment to an ongoing partnership. We stand behind our products with support you can count on- for the long run.
Today, twenty years later all of that has gone clear out the window. When you buy one of our brand new test instruments the standard warranty period is only one year. One. If you want warranty coverage longer than that, you have to pay through the nose for it. Marketing doesn't want you to continue using our legacy products over long periods of time. They want you to buy whatever latest and greatest high cost instruments we've got instead of maintaining your vintage gear.
Tucker Electronics: "There seems to be 2 issues every time we order an option where a key code number can be sent to us via email to open up the programming capabilities of the unit. The first issue is that it usually takes up to 2 or 3 days to get the original order entered and a confirmation number sent to me. The second issue is that it takes a couple of days after that to receive the email with the key code number. We are quoted 2 weeks for the delivery time for the option and we know well that the key code will come sooner via email ... It just seem that when we order 4x4 or 5x5 parts we never have these issues. We place the order, a confirmation number is given at the end of the conversation and the parts ship the same day. Is there a simple explanation?"
*Sure there's a simple explanation! Nobody on our end is paying attention. Weeeeehooooo!
Motorola received the entitlement certificate for his (2) Baseband Studio for Fading order. He was surprised that it arrived in an anti-static bag "why did it come in this type of bag? what a waste of money!" He was confused when he received two copies of the entitlement certificate There were no instructions on the certificate for what to do after retrieving the license file. No indication what to do next or how to obtain the software.
*Yep. We rock.
More customer reaction on the recent expiration (May) of the H99 service note which covered all H99 repairs under extended warranty. NEC Wireless Networks had 6 XXXXX/H99 failures at the same time they were ramping up production after a 1 week factory shut down. The customer still has reliability issues on their 40 units of H99 even after we provided the proactive corrective actions to improve the reliability. They are experiencing 1-2 units out of 40 broken every 3 months. NEC totally disagrees to pay any repair cost for their H99 because they have a strong concern about "high failure rate of H99" and "owner cost" while under cost reduction pressure on their products. Anritsu is pushing the MG3700A to this customer to replace their XXXXXBs and the customer is interested due to cost. According to the FE, "This is our biggest threat in the near future". They are asking XXXXX to provide extended warranty for H99 repair cost and material until the end of support.
*See my above comments about the H99 debacle. Warranty service on these defective output boards was extended, but the redesigns are not much better than the original boards. They won't last long even after repair with an upgraded output board.
Feedback from FEs in Japan about XXXXXC quality issues ... 10 out of 20 failures at Kyocera (unleveled, hard disk failures, download problems). Other customers are experiencing this as well. Two dead units recently for "no modulation" and many H99 units become unleveled.
An exceptionally high XXXXX failure rate (over 100% among their 15 As!) reported by the FE for our customer, Leihua. One unit even failed 7 times!!! One failed 4 times and another 3 times. The biggest problem they are facing is when those XXXXXs reach their EOS (end of support) date next year and fail again, they will have to discard them even though some of these boxes are just ~4 years old! Please give the customer a way out! Those failures are hurting the customers and our company a lot! We already begin to lose SG deals to Anritsu in this account and there's almost nothing we could do.
*Worldwide in the first year of service we've observed a 28%-30% fail rate in these products. Normally when we release a new instrument platform for sale our worldwide failure rate target for the first year of service is 3%-5%, with 5% being totally unacceptable. A major factor in the low reliability of these units is that they have surface mounted tantalum capacitors that are either counterfeit or underrated on voltage for the circuit. In either case they will usually overheat, grenade off the board and crater it rendering the PC board useless. This is a collective result of rushing products through the R&D phase, cutting corners, outsourcing, and offshoring our manufacturing. Can you say Shareholder Value? C'mon now and say it with Supergeek like you mean it... Shareholder Value! Yeah!
posted by factory_peasant at 11:36 AM
1 Comments:
So many fuck ups, so little time...
DT
Post a Comment
<< Home