Business operations- How long to spend on a free diagnostic?

[Tony_Scarpelli]

In another thread it becomes obvious that many people may take different levels of diagnostic to offer as a free diagnosis. In attempts to become more productive I thought it important that we each share a part of our procedures and operations in that we all might see ways to improve our own productivity.

It is too long to try and put down every scenario but I will start with a few that I see that are hardware as IMO repairing software is more standard.

A computer comes in, won't boot up is the complaint of the customer. You check it in, plug it in and there are no lights, fans and you do not hear the hard drives spin up...what could it be? Is is plugged in? Yes. Is the power supply turned on? yes. Then it is a power supply. This takes me about 30 seconds. I then verify it with a PS tester and plug in a bench supply to test whether there is further damage to the other components. Of all hardware failures around here this is about 90% of it. I keep 2-3 dozen power supplies on my shelves for this. We see this most after a storm, in the dry winters with STD or after the local power company has had a power outage.

In about 30 seconds or less Yes, it is a power supply.

Same scenario but this time you get some indication of power working, could still be power supply so you pull out the $12 power supply tester to see if you get plus and negative 12 and 5v channels if one is failing, replace the power supply to see if other components work.

Depending on the brand of computer (emachine) you have a very high likelihood that there is damage to the motherboard and you continue to check it so you take a test bench power supply and plug into the mobo and boot again. This time it comes up fine so you know it was the power supply this all took 3 minutes or less.

Same scenario again but this 3rd one the video does not come on but you know it is not the power supply you just checked or installed so now you plug in a test bench video card and walla its working so you know it was the power supply which also took out the on board video so you advise client they can replace motherboard or just add a video card to see if you can get some more life out of that motherboard. Still we are under 3 minutes to give the client an accurate estimate.

Fourth scenario power works, monitor works but still no boot, you get past post so you suspect its not the bus so what do you do now? You boot to Ubuntu (a fully self loading and functioning PC OS). Now I can see if it is hardware is working or narrow it to the hard drive/OS. I can even take the hard drive out and boot it in safe mode to a test bench machine to see if the hard drive works or I can install it as a slave and run diags on that drive...Ok now we might run over 3 minutes But this scenario happens in less than 10% of repairs.

Fifth scenario the system boots but is very unstable, you know its not power supply, you also ruled out the bus with Ubuntu and it is still flakey so you have narrowed it to memory, cpu or motherboard so you swap memory with bench memory and see if it removes the system instability....if it does then you likely fixed it but you can decide at this point to run a full hardware diag to see if anything else was damaged by a ESD.

I myself have learned over 20 years that if you find 1 thing broke you can fix it/replace it but if you find multiple failures then I begin to recommend not fixing it and getting a new system as there was likely some level of ESD or power outage/fluctuation. If the client repairs it against my recommendations then I do not warrant any of the repair but i am happy to do it.

When I replace memory or cpu or motherboard I always check the installation of the OS to see if there was corruption causing additional instability.

In my experience and from my parts purchases list the hardware most likely failed is P/S after that it is a toss up between Nic's, Video and HD's, after that Motherboards I see the least repairs on memory or CPU's (other than cpu fans).

Also in the above scenarios I always check for properly working fans but I left that out of my simplified diagnostic as it takes next to no time to do.

I realize it sometimes is a good idea to run systems checks on memory or hard drives I just do not do it on every system. If you go to the doctor for a soar throat they usually do not give you a foot x-ray. That would drive up the time and expense too much.

 

[NYJimbo]

Different scenarios require different amounts of time. A simple power issue can be diagnosed very quickly by just about any tech. But what about issues where the machine is booting but causing problems later on, like BSOD's for example ?

Seems like someone would have to spend a lot more time than 3 minutes to pin down a BSOD. I know that some techs might think that booting with Ubuntu or a similar *nix will isolate the hardware from the installed software but as we all know, the various *nixes will not talk to all hardware. Some will not even display on the LCD of laptop without fiddling with configuration or drivers. Another example of a failed diag using a *nix is when a machine BSODS or freezes under windows but not a *nix because the resolution of a video card is not "tested" under the *nix but it is under windows. A tech might assume that the issues is software if it DOES work under the *nix but not Windows, but in reality the Windows environment pushed the card or chipset further than the *nix. So in this case using a *nix is a bad method to test a computer if it is software or hardware.

Another possible issues is something very simple like an overheating issue. Customer complains about anything from slowness, rebooting, bsods, etc but you cannot pin it down in the first 3 minutes because you actually have to drive up the temperature of the computer to a point where it fails. Sometimes you cannot do that without running a program like prime95 or any virus scanner for at least 3 minutes AFTER a full bootup.

There are quite a lot of different scenarios where you would have to put in a lot more time to properly diagnose to even get an idea where to go next, but I think most techs would agree with that.

So to get back to the question of what you will do for free, well I think a proper diagnosis depends on the job, so you either have a fee or don't have a fee because if the job requires more time for a DIAG you cant blame the customer for you needing more time to diag the machine properly.

Me, I don't charge a fee for desktops but I do for laptops.

 

[SilverLeaf]

I was going to post this in the other thread, but it's still locked.

Yeah, I can usually determine the likely cause of a problem pretty quick, and be right a good part of the time. But, if someone comes in with an obvious problem (say a virus), I still do a full hardware diagnostic. Some customers may decide that proceeding with a $99 virus removal on a $400 computer that also has a hard drive throwing read errors and a PSU with a fan that doesn't spin anymore is just not worth it. Without a proper examination, how would they know? I use this as a selling point, and most customers are happy to pay the $39 diagnostic fee.

 

[brandonkick]

Wouldn't give an estimate, at least not one I'd stand behind.

Too big of a risk for far too little of a reward.



I'd rather know whats wrong and have a solid understanding of what it will cost to fix it. No surprises this way. No customers expecting other free services.

So do the diagnostic and charge them for it if they don't want the machine fixed.

 

[NYJimbo]

But, if someone comes in with an obvious problem (say a virus), I still do a full hardware diagnostic.

Too true. Many a computer coming in with viruses ends up having a hard drive or other issue that they didn't even know about. If you are in the middle of a clean up and the drive starts to die then what ? So you have to diag more than just the obvious.

 

[AlexT]

I don't charge anything for diagnostics and i don't have a limit on the time i spend troubleshooting a machine. However i'm confident enough that i can figure out the problem almost every time fairly quickly and the others i take on the chin. Part of how i sell my services in the current economic climate is that i guarantee that i will not do any work that costs money until i call the customer with a quote and get the go ahead from them. This is paying off for me as it puts the customer at ease and to be honest, there has been a very small % who don't go ahead with repairs and 75% of those is because i recommend that they don't because of the associated cost.

 

[Tony_Scarpelli]

Different scenarios require different amounts of time.

There are always exceptions. Maybe 1 in 20 computers requires more than 3 minutes to get to the bottom of the problem.

Look up the Pareto Principal. 80% of the problems are solved by 20% of the possible solutions. Using statistics of all past repairs you can easily narrow down what you look at by past experience.

http://en.wikipedia.org/wiki/Pareto_principle

Furthermore just like homicide detectives solve crimes starting with the most likely suspects: Husbands statistically 80% are a females killer and visa versa so too we can gain productivity by starting by what it is most probably. Some times it is not the husband it is the boss on those cases it is the last suspect not the first. That is why we have to be very good at reading clues.

Seems like someone would have to spend a lot more time than 3 minutes to pin down a BSOD.

The fact that it might take more than 3 minutes to diagnose a minute amount doesn't mean that it doesn't take 3 minutes to diagnose most computers.

But even with the harder to diag issues experience should direct us in the most likely cause of the problem. Of all BSOD what % are caused by Video drivers and other software issues? Most right...so thus again using experience and statistics of the most likely problem we keep checking off systematically down our list of likely suspects.

It is like a cop going after a diamond hoist. That is way different than the average grab and run crime but it also narrows down the list of possible suspects you have to screen.

I know that some techs might think that booting with Ubuntu or a similar *nix will isolate the hardware from the installed software but as we all know, the various *nixes will not talk to all hardware.

Again statistics 80% of the time this gives a true reading and we have to depend on our experience and wisdom to get us the last 20% of the way. BSOD are not that difficult or mysterious to an experienced tech. I spend more time grabbing the stuff on my bench or stuff that should be on my bench than wondering about what is causing this BSOD.

ON the MOST difficult to trace down BSOD after you ruled out software all you need do is unplug everything from the motherboard except power and video and if it still BSOD you have your answer. If it does not, then one at a time you plug in the rest of the peripherals until the BSOD comes back and then walla! you got it don't you? This is very rudimentary although it might be baffling to some... I do not see how an experienced tech can have difficulty tracing down a fault? OK maybe if I am always answering incoming phone calls but if I am focused on the task this just doesn't take that much time.

And 1 in a 100 times you might get a false negative thinking something has proven not to be at fault so you go through everything else and have to start again from the beginning. These do happen but they just do not happen that often.

Another possible issues is something very simple like an overheating issue. Customer complains about anything from slowness, rebooting, bsods, etc but you cannot pin it down in the first 3 minutes because you actually have to drive up the temperature of the computer to a point where it fails.

This is the exception but it takes not time what so ever to notice if the case and cpy fans are dirty, slow and sometimes there is even a burnt smell to an over heat issue. You must use all your senses and experience. Again heat issues are obvious by the clients description and conditions you see. You get even more clues from the brand and how much dust is in the cpu fan, case fan, age and condition, presence of cat hair, roaches or mice droppings.....inside of the workstation. The minute I suspect there could be a heat issue we boot to bios and watch the temps you can see them going up too fast if that is an issue. We have multiple workstations so we do not need to sit and wait for it to heat up. By the time we get back to it, is already running at its system temps in most cases but even if it was a cold boot we use observation of condition with clients feedback to guide us.

There are quite a lot of different scenarios where you would have to put in a lot more time to properly diagnose to even get an idea where to go next, but I think most techs would agree with that.

Again you are looking at exceptions to disprove the rule when the rule is most of the time you can and do know exactly what is wrong the first few minutes you put hands on a computer. It is true less experienced techs fumble around and do every manner of needless diagnoses until he stumbles on the cause but that is not what we are talking about here.

Perhaps you need to work out a diagnostic model of repair and print it out and put it up on your workbench wall. I do this for my new techs. It helps greatly when they get overwhelmed.

We have a rule if you do not know what is wrong within 10 minutes call me or the manager and we will use it as a training moment to walk you through the steps of the diagnostic model.

So to get back to the question of what you will do for free, well I think a proper diagnosis depends on the job, so you either have a fee or don't have a fee because if the job requires more time for a DIAG you cant blame the customer for you needing more time to diag the machine properly.

This is the first thing you have said that I agree with. If I have a new tech that takes too long, our incompetence has no baring on the client. The client does not pay for our stupidity. This is why it is so important to stick with a good working diagnostic model.

Me, I don't charge a fee for desktops but I do for laptops.

That makes sense since it takes more time to work on them. We handle them the same pretty much but we find with the exception of heat issues and jack issues they have less hardware issues than workstations overall.

 

[Rosco]

I do not do any daig for free.

I charge $25.00 flat rate for pickup.

If they do with the repair it goes right toward it.

onsite i charge a regular hourly rate.

 

[Tony_Scarpelli]

Too true. Many a computer coming in with viruses ends up having a hard drive or other issue that they didn't even know about. If you are in the middle of a clean up and the drive starts to die then what ? So you have to diag more than just the obvious.

I should have mentioned this earlier. I recommend that you try a IR thermometer. They are at harbor freight for $25 on sale. It is amazing how you can use it to quickly check things. Failing hard drives and other things too often get much hotter than normal operating temps before a failure.

Most drive mfgrs have a normal operating temp and you quickly learn this as you Ir every drive, every Video card and onboard video chip. It takes 3 seconds and in the beginning you will just be learning what the temps and variances are but sooner or later you begin to get a feel for failing components. I also use it on laptops before working on them to get a heat signature.

Heat is the enemy of electronics and you can get some good insight with it. The thing I like most is that it doesn't take 5-15 minutes to diag every item. I am against doing a diag of every memory stick even with no symptoms of it failing, or every hard drive, or every DVD burner, or every video card. This can easily turn into an unprofitable venture.

There is a trade off of doing a thorough job and just wasting time. And if those procedures slow the overall return of clients computers in the name of thoroughness you might well be giving worse service by doing it.

however I don't fault anyone for being thorough, its your time, if you really feel it adds value and the customers see it and reward you then that is your economic model. I won't criticize that.

 

[NYJimbo]

I should have mentioned this earlier. I recommend that you try a IR thermometer. They are at harbor freight for $25 on sale. It is amazing how you can use it to quickly check things. Failing hard drives and other things too often get much hotter than normal operating temps before a failure.

Yes, an IR thermometer is useful when it make sense to use it. But they are only good for when a thermal event takes place. If a fault with a drive is not due to a spindle issue or something that generates heat then a thermometer is a waste of time.

 

[NYJimbo]

Again you are looking at exceptions to disprove the rule when the rule...

I was going to go into this deeper but I just realized that 99% of the members of this board seem to have no opinion on any thread so what's the point of trying to drag out this discussion.

However, I would like to note that if a tech is using analogies, theorems, rules or principles to decide his troubleshooting process then maybe he is not really trying to correct a problem but is instead fixated on his way of doing things and so if the problem doesn't fit the diagnostic then the problem is unfixable.

Example: "I have spent x minutes and done x things and it is not fixed, therefore it is not fixable because a principle tells me so".

Whatever.

 

[Tony_Scarpelli]

http://en.wikipedia.org/wiki/Dunning–Kruger_effect

 

[AlexT]

My business partner is very methodical in how he troubleshoots, personally i kinda bounce around a bit. I have to say that i believe there is a strong case for both methods. I think whatever works for yourself is the best way for you to troubleshoot a problem. We work on a no fix no few basis so it is in our interest to fix the customers machine and between our very different methods of troubleshooting we have a huge success rate. It's for that reason i wouldn't knock anybody's method as long as they get the job done.

 

[GoHighVoltage]

Rule #1 - My time and experience is worth money.

Rule #2 - I never give free estimates. I charge for diagnosis. Time is worth something.

 

[Scales]

I was going to go into this deeper but I just realized that 99% of the members of this board seem to have no opinion on any thread so what's the point of trying to drag out this discussion.

However, I would like to note that if a tech is using analogies, theorems, rules or principles to decide his troubleshooting process then maybe he is not really trying to correct a problem but is instead fixated on his way of doing things and so if the problem doesn't fit the diagnostic then the problem is unfixable.

Example: "I have spent x minutes and done x things and it is not fixed, therefore it is not fixable because a principle tells me so".

Whatever.

I am one of those 99%. I barely have time to read a few threads let alone get into a discussion. Carry on

 

[mkeathley]

I have my techs run a HDD sector test and MMRY test on every machine. Yes it takes time. Matter of fact I got one part timer workign 2 days a weke who right now just comes in and starts these tests.

But we replace probably 60% of hard drives. 1 bad sector = 1 bad drive.
It is frustrating havgion a computer sit on the shelf a week, only to open in up and know in 5min its got a bad powersupply. But even then while on my spare power supply I test the HDD and MMRY as well.

 

[compnet]

Some virus infections take 5 minutes tech time, some take hours. Some hard drives take 5 seconds to take out, some take 15-20 minutes. Some dc jacks take 5 minutes to get to and are on a cable, some take 20 minutes and require soldering after that. Screens, keyboards, even other software fixes have a wildly varying time to complete.

In shop you have to just average all out and make sure the slow ones are worth it for your price. Same with free diags. Sure there are some that take a lot longer than others, but as long as you feel the customers coming in is worth the average time, than why worry about the outliers?

If we were going to charge individually based on the exact time it took to diagnose and fix every issue, most of our work - easy viruses, and dead hard drives and power supplies - would be close to free and we wouldnt make any money. Some things - nasty, elusive problems that we waste hours chasing down - we would have to charge hundreds of dollars for and the customer wouldn't pay..

 

[brandonkick]

I think the "3" min per diagnostic or estimate or whatever you want to call it, is too small of a term to be a blanket statement like this.


The power supply problem could PROBABLY be diagnosed in three min. I wouldn't waste my time with a PSU tester. They don't test the supply under load and I'm willing to bet on many of them you can supply say 11.5V and the 12V LED will light. That half a volt can cause system instability. I've seen quite a few supplies that a power supply tester thought was ok, but were the cause of an issue.

When I test supplies I put them under as much of a load as I can (maybe load up a live os and start a hardware stress test program) and then check with a digital volt meter.

Of course, all of that IMO is a waste when you can plug in a spare bench PSU and find out if that's your problem. I'll agree that is a quick problem to diagnose/estimate.

However, I've seen systems that will not post but do get lights, some fans spin but the screen stays blank. Not a video problem, not an issue with an add on card. Re seating the memory usually fixes this problem.

I've seen very, very few machines that have ever had a cpu go bad.

With the right amount of parts available/handy I can see a certain percentage of repairs being fairly quick to diagnose. I can also understand, to a point, not wanting to test every piece of hardware in the machine. Some people do full diagnostics, some do not. I can easily see problems that would take much longer to diagnose, like a hard drive that's starting to go flakey or caps that are failing but not swelling or leaking or even a PSU thats starting to crap the bed.

Laptops are an entirely different story. Even the fastest tech's here who do them all the time are looking at a good half an hour just into disassembly / reassembly or more and then the diagnostic/fix on top of it.

 

[Abet Computers]

I do it like this, I charge a cheap $10 Diagnostics - 90% of the time I can make the diagnosis over the phone before they even bring it in.

I get them to get the machine to me, I run a quick diagnosis call them with the results - if they agree with the repair / Diagnostics fee gets waived and I proceed to fix it. ( if its to much to repair it costs $10 and they can come pick up the computer ) My success rate from doing that is about 88% and for the ones that don't come pick it up i get free parts to use as bench test and recycle scrap.

I have each person that drops off the computer sign a work order that states if the computer is not picked up within 30 days............... they give up all rights to the property.

( I still hold each item for 6 months ) required by law in case it was a stolen item.

 

[brandonkick]

I do it like this, I charge a cheap $10 Diagnostics - 90% of the time I can make the diagnosis over the phone before they even bring it in.

I get them to get the machine to me, I run a quick diagnosis call them with the results - if they agree with the repair / Diagnostics fee gets waived and I proceed to fix it. ( if its to much to repair it costs $10 and they can come pick up the computer ) My success rate from doing that is about 88% and for the ones that don't come pick it up i get free parts to use as bench test and recycle scrap.

I have each person that drops off the computer sign a work order that states if the computer is not picked up within 30 days............... they give up all rights to the property.

( I still hold each item for 6 months ) required by law in case it was a stolen item.

You cannot make a diagnostic over the phone.

You cannot know the problem until you have the machine in front of you and you do some testing. You can make a few really good educated guesses, but they are guesses none the less.

For example: Customer rings in that his machine won't turn on at all. You tell him, oh that's easy. It's probably the power supply. A new one is X and the install will cost X.

They bring it in and the machine fires up in your shop. Now you have no idea what's wrong. Anything from an intermittent problem with the hardware to a fault with the outlet / wiring where the customer had the machine.