Conference 7.286::digital

Something I read a couple of years ago...

    
                                                          Jim Seymour,
                                                          PC Magazine, 
                                                          September 12,
                                                          1989

    "It's just not true that all computer retailers and salespeople
    are terrible. Let me tell you about the best computer salesman
    I know.
    
    "He has better understanding of products he sell --- than many
    daily users of those products. He can do effectively hands-on
    demonstrations of them - and does, at length, to his prospects.
    
    "With every sale of a computer or major peripheral ---, he comes
    to the buyer's office or home and installs the equipment. When
    the process of delivery and installation is complex or time-
    consuming enough to require a technician, he still comes out and
    holds the buyer's hand while the technician hooks up the equip-
    ment. A week after a sale, he calls the buyer to make sure he's
    happy. Three months later he calls again.
    
    "If a new customer mentions that he was recommended by a previous
    customer, he sends the previous customer a thank-you note. And if
    the new customer actually buys a system, the previous customer
    finds a messenger at his door with a bag of cookies, or a box of
    chocolates, or a bunch of flowers.
    
    "When the customers' equipment comes in for service, he tracks its
    progress through the shop. If there are unexpected delays, the cus-
    tomer gets a call from him, not from some sleepy techinician. ---
    
    "In the buying process, he respects his customers' intelligence.
    He makes them comfortable by spending time with them, so that when 
    they reach a buying decision, they're happy - not uneasy - with
    their choice.
    
    "After the sale, he shows up on the customers' doorstep to help
    them each get the new computer or peripheral up and running. He 
    does that whether they're computer wizards or computer tyros.
    I don't need someone to help get an HP LaserJet Series II connected,
    but when the guy I just spent the better part of $2,000 with shows
    up with the printer, remembers the cable I forgot to ask for, and
    sitsand chats with me for a few minutes while I plug things together,
    I feel stroked.
    
    "Of course those installation visits let him size up the nature of a
    customer's business and possible needs. And his postsale calls let
    him make sure that things are still running right, as well as convin-
    cing his customers that he wants them to be satisfied with what they
    bought.
    
    "The whole process, of course, encourages referrals. And his acknow-
    ledgements of those referrals cement that bond: 'Thanks for the vote
    of confidence. What else can I do for you to earn that trust?'"
    

...petri

    
    
    
        In my 12 years at DEC I've worn a few hats, 2 of which have been
    sales and sale support (Currently sales support consultant for the G.E.
    AeroSpace team.) I can tell you first hand that these positions require
    2 very different sets of skills! In my opinion, there is an absolute
    need to have both sales and sales support.
    
       If you really understand the process, you would see this as well. In
    fact every vendor (IBM, HP, SUN, Amdal, Convex,) or large distributor 
    (Avnet, Pioneer) uses sales support throughout there sales cycle. In
    large SI programs, which is what I support, there's NO WAY a sales
    person could do it alone. It fact IBMs ratio of sales and sales support
    is almost 1 to 1. HPs is about 3 or 4 to one and DECs is about 3 or 4
    to one. (3 or 4 sales people for every sales support person to
    support.) SUNs ratio is higher, but they also do not have the product
    set or the installed base of IBM, DEC and HP. A good way to look at it is
    SUN has a sales force that is similar to DECs 15 years ago. 
    
       I do agree we need to do things smarter, use channels more
    effectively and get more done with less people. But this will not be
    something we can accomplish over night. Right now we have no focus, to
    many products across to many vertical markets. We have been trying to
    be "all things to all people all the time" for the last 6 to 7 years.
    If you look at SUN for instance, they focused on the desktop with open
    (UNIX?) systems. I seem to remember that we had this type of focus at
    one point. The major difference was it was the department level
    computing. We've made mistakes as a company, we missed judged the
    markets and it's tough playing catch up. But it's even tougher if you
    are under staffed and under skilled!
    
    
    
    
                                            Phil... 
                                                   

    
    
    
      regarding -1
    
      missed judged should be misjudged
    
    
      

RE: .0

    
>    1. The salesperson is self supporting - They don't need
>    sales support because they are technically literate and can
>    make phone calls/sendmail when they need an answer they cannot
>    supply.

This could be a reasonable goal to shoot toward.  As mentioned 
elsewhere, things are generally too complicated for this to occur.

Technical expertise is far overated (for a sales person)
if it is defined as...

Someone who understands Digital products and services well enough
to make them do useful work.

What the sales person MUST have is "technical" expertise like...

The ability to understand a customer's business better than the
customer does and apply proven solutions to improve that business.

The first definition is DEC centric and customers don't value it (at
a business level).  The second one is what customers value.

The first definition is needed for two reasons

1)  At some point someone has to actually wire things together etc.  
	this is valuable and needed

2)  Digital's culture requires the sales team to "devine" the value
	of our solutions and figure out how to put them together.  
	This is not valuable and is wasteful.  Rather than have 6000
	people figure out how to put together a network, why not have
	a few people create a tool that does it.  Currently, the 
	sales team must have someone who is technically good or they
	will fail.  Much of the reason for this is wasteful.

Also, the sales rep is required to do many things that 
should be automated.  These things should not require any questions
but generally take up 65% or more of a sales/support persons time.

These things are:

1) Defining/Configuring/Quoting what the solution should be...
	- What SW
	- Clusters or standalone
	- what type of network
  This is valuable stuff to do but there are no standards or tools
  to help this.  There are at least 6000 answers to every question.

	  336 VAXes in the price book
	X   3 types of configs (standalone and at least 2 types of cluster)
	X   3 types of SW (capacity, personal, concurrent)
	______
	3000 answers

Since this doesn't include things like mixed clusters and other
"creative" configurations, the number is easily double this.

No tools exist to help this...

I spend 25% of my time on this clerical task alone.

If a tool existed to automate the configuration process the sales team
would be much more self supporting.  800 DECSALE exist solely because
this is so difficult.  Also, the customer does not value this time
spent...they assume it is easy.  This is a separate discussion for 
another time.


On the delivery/installation side, similar issues exist.

RE: .1

This article provides a reasonable set up guidelines, I have the
following comments...

1)  It was written by a PC person and focuses on the end user and
the "computerland" sale

2)  Most of the stuff I sell gets installed hundreds or thousands of
miles away so I cannot provide the hands-on stuff that this article
asks for.  I must rely on Customer Services to do this.

Ken Slattery

    Which '90s were you talking about? Computers, particularly hardware,
    are bought like basic office supplies. A salesperson can not afford to
    come out and help you stock your shelves every time you order a few
    boxes of paper. Each salesperson needs to sell mega$s not k$ each
    quarter. The customers have demanded the low prices and therefore
    low proffit margins. To make up for the lack of proffit on each sale
    the sales person needs to sell more. To sell more they can not be
    involved with every other job (service manager, service tech,
    installer) that is needed.
    I'm sure most of us would like to buy from the type of salesperson in
    .0 but could we afford to? Some one has to pay for all the hand holding
    most customers will not. (and can't afford to)
    Jeff Filer

    If I have understood the speeches, our management is looking of having 
    different skill sets for different lines (They have not told where we 
    really want to be in 5 years time?)
    Commodity HW and SW -- Not very much handhelding with the prices
    			   in that market area in future. Good sales and 
    			   services points (via distributors?) and excellent 
    			   marketing. And products must be made for that 
    			   market.
    Computing utility  --  Solutions selling and respective support. End
    			   user equipment for utility users will be purchased 
    			   through commodity channels. (Will we be an utility
    			   provider or a systems provider for others?)
    Systems integration -  Similar skillsets as in utility business.
    			   Needs plenty of consultancy until more standard
    			   systems and packaged applications make things 
    			   easier. If IT usage goes towards utility
    			   service, it means strong standardisation in all
    			   levels. Look at the telecom business or broad-
    			   casting. (Will we be doing SI in large scale in 
    			   future?)
    Takes plenty of effort to be on the road to that. And the ship may sink
    before it is fixed. The investment for the new must be earned from the
    present business.
    
    Or may be that I have not understood anything :-(.
    
    --Ritva
    

re: .4

Could you please expand on this:

>The ability to understand a customer's business better than the
>customer does and apply proven solutions to improve that business.

How well must the sales person understand the solutions in order to apply
them?  What is the distinction between "applying proven solutions" and 
"understanding products well enough to make them do useful work"?

   Gary

re .0

>    1. The salesperson is self supporting - They don't need
>    sales support because they are technically literate and can
>    make phone calls/sendmail when they need an answer they cannot
>    supply.
 
I think of myself as rather technical, but if you want me to sell, you have to
give me some help.  I sell software systems to engineering, and so my customer
is very technical.  I can not keep up on the latest in data base technology as
well as stay current with my customers business problems.  I just don't have
time.  I emphasize understanding my customer, and staying just technically
knowledgable enough to engage them in discussion about their true problems. 

I will not spend the time to understand the ramifications of using
interuptable long transactions versus sequenced short transactions.  Nor do I
plan on learning the tradeoffs between single threaded servers and their
inherant security versus multi-threading and its performance.  I know what they
each are, and in general I can ask questions to get a customer to think about
the issues, however, I am not an expert nor do I think I can become a data
base expert.

After all, the data base is only one small part of the solution, I would 
rather understand the basics of a lot of things and have sales support 
available to back me up.  I do think the sales people trying to sell all the 
thousands of products DEC has are at a disadvantage with our sales appraoch 
and it's inherant lack of available systems engineers.
   
>    2. They have a history of success in previous accounts or job
>    positions. Don't just move bodies from position to position.

I agree.
    
>    3. They understand that to be successful they must understand
>    their customers business. They must partner with their customers. 

Carefull of the P word, many people use it to justify making poor business 
decisions and justifying them only on the long term "opportunity" where we are 
treating them like a partner, but there is no like commitment from the 
customer.  I all too often see DEC back down from enforcing the T & C's of the 
contracts we write based on "partnering."
    
>    4. Initiative should be encourage and rewarded. Good ideas should
>    be shared with the company immediately.

How can anyone disagree with this?  I only disagree in that spending time in 
meetings with DECies is time spent away from Custis.  Lets put the onus back 
on first level managers to require honest, complete call reports and then 
extract usefull information and pass it on.
    
>    5. Salespeople should partner with their counterparts throughout
>    the country. If a salesperson is assigned to manufacturing they
>    should talk on a regular basis to other salespeople with the same
>   account background. Learn from each others mistakes and victories.

Who pays?  We currently have sales people that are not comfortable calling on 
engineers and then they want excessive handholding.  Since we seem to have a 
sales person call on a wide variety of functions in an account or account set, 
there is no way to have them spend time with all the types of successful sales 
people and still have time to sell.  

I would rather see sales training be a rotation location where successful 
sales people spent a quarter every two or three years.  No training sales 
people allowed unless you have sold and made quota (budget) with in the 
last year.  

I took a rediculous training course (called BASE) where the first day was 
spent learning how to:  
	a) ask the customer how much time we have
	b) recap the last call
	c) tell the customer why you are there
This is was being offered (and required) for people that had been selling for 
years, but were new to the Digital Sales Way.  What a crock.  This is a fine 
10 minute reminder, but to spend a whole day on this was just stupid.
    
If we can't even train people that go to training, how can we expect general 
meetings to help people exchange meaningful information?

>    6. One base salary level for all salespeople. Commission should
>    be encouraged and those over budget rewarded accordingly. No
>    trips, DEC 100 etc. Cash rewards are great incentives.
    
How about 4 base salaries, one for each of the basic sales experience levels?  
I expect to make more than a person who has only sold for a year and didn't 
make quota.  I expect a larger base AND I expect that we could each make as 
much at the end of the year if she sells as much as I do.  How about using 
nonrecoverable draws as the differentiator?

Matt  

    re:  previous "salesman" characteristics
    
    I see one major ability missing from all the previous lists:  The
    ability of a salesman to also be a businessman.  The ability to know
    when he's doing something profitable and when he's not.  The ability
    to walk away from business that's not profitable, or to maximize the
    return on a sale by managing the effort required to get the sale.
    
    I work with a top salesman from one of our distributors.  At any point
    in a sales effort, he can tell you what it costs him to make the sale,
    and what the profit is that he will return to the company if he makes
    the sale.  That number directly affects his paycheck.  He still seems
    to achieve an acceptable level of customer satisfaction, and he gets
    lots of repeat business, so he can't be accused of skimming the cream
    on choice accounts.
    
    We need more of this kind of person, both as sales reps and as sales
    management in order to get our SGA numbers into line.  We need to focus
    on being a business again, not a "good ole boy" back-slapping club.
    
    Geoff Unland in Austin
    

re .9

This leads back to the whole metrics problem.  You said he gets affected 
(paycheck) by the effect of his business dealings.  No Digital sales rep is 
affected by the profit effects of their work.  There is a gret deal of noise 
about it, but nobody can tell me the cost to the company, net of any corporate 
taxes, of teh software and services I sell.

DEC sales people aren't dumb, just managed by dumb metrics.  Metrics set by 
people who have never really carried a bag in many cases.  Naturally dumb 
metrics are created by managers who have no concept of what they want, 
therefore they manage like their predecessors did.  Naturally that leads to 
the same results their predecessors got.  What a suprise.

I try not to write unprofitable business, but there is a software package used 
to evaluate profits called PET that ensures DEC will never know whether or not 
a project is profitable.  I worked (very shortly you can be sure) on a project 
where the people making pricing decisions did not understand the difference 
between margin and markup.  Now that is O.K. if all you do is run AQS (the 
assines quote system), but is unconscienable if you actually want business 
responsablity.  The people were IBU second level managers, a Program Manger, 
and several EIS consultants.

Needless to say, we lost the opportunity because while we were technically 
superior, we did not make the short list due to our pricing.

Matt 

    re. .10  don't get me going on metrics again. i'll just get in more
    trouble. tho metrics are still killing us big time.
    
    re. .0
    >The salesperson is self supporting - They don't need
    >sales support because they are technically literate and can
    >make phone calls/sendmail when they need an answer they cannot
    >supply.
    
    
    Absolutely not possible today. Salespersons, mostly don't specialize.
    They sell everything from DEREPs to mainframes. I doubt any
    salespersons even know of all of DECs products let alone anything about
    them. Not a criticism. Just Fact.

    re.0 - self-sufficient salesperson
    
    I will reiterate what some others have indicated.  If a salesperson has
    only a few products to sell, they can be very self-sufficient.  Digital
    salespeople have hundreds of products to sell, depending on what the
    customer is asking for, this time around.
    
    I have spent nearly 20 years in this business, focused on the technical
    side for the entire time.  What I have found in the last few years is
    that it is harder and harder to understand the individual products, let
    alone the inter-relationships among them.  If I am to be of assistance
    to my customers, I have to recognize when my knowledge stops and when I
    have to call on someone else with the requisite knowledge and
    expertise.  
    
    That is the beauty of the sales support function.  I can become and
    `expert' on several related products.  My compatriots do the same.  As
    a team, we can provide local support for the high-volume products and
    rely on other experts elsewhere for the other products.  Sales deal
    with the customer issues.  There is not enough time for an individual
    to be expert in everything that a customer will ask for.  I have a
    tough time keeping up with the technical stuff.  If a salesperson had
    to know how all the pieces went together, in addition to knowing how to
    deal with internal and external politics and competion and ... 
    
    	I'm not looking to justify my job, because I know that support is
    required both inside and outside Digital.  If the sales person could do
    everything, we wouldn't need any other job code.  8^)
    
    tgc

    
    
    
       Some great replies, and I could not agree more with Mr. Haag in .10!
    
       Something I left out of .3. I don't know about the rest of you, but
    I'm already seeing opportunities "slip though the cracks" because of
    layoffs. Reps being spread to thin. I've actually been making sales
    calls because the reps are comfortable with me, and know I have a
    background in sales. As far as I'm concerned if we cut our sales force
    by 2/3s, 6000 to 2000, we might as well just close the doors!  

    I think as we downsize people we will also downsize product lines.
    Perhaps we will get to the point where sales support can be derived 
    not locally but nationally from an 800 number similar to our
    help facilities. I use DEC Sell because they save me a lot
    of time looking for answers they have at their finger tips. They
    also provide me with timely answers. 

    re. Metrics and Good Salespeople
    
    As a salesperson, I know it is tough finding and closing new business
    (especially when you are taking it from the competition).It is much 
    easier to grow the installed base. Our metrics don't provide any
    real incentive for reps that seek out and close these new opportunities.
    
    Zereski, in his last DVN, mentioned that there will be a new incentive
    for reps to close new business. I hope this becomes reality, even
    though Zereski is now gone.
    
    Bottom line: Good salespeople are prospecting all the time. Good
    companies reward them for it when it pays off.
    
    CV

RE: .4
    
>How well must the sales person understand the solutions in order to apply
>them?  What is the distinction between "applying proven solutions" and 
>"understanding products well enough to make them do useful work"?
    
    Good question...
    
    Much of what I am about to discuss is in several books by Mack Hannon
    (not sure of spelling).  I just finished his book  Consultative
    Selling.  I would recommend this book to everyone.
    
    The way things work today is roughly...
    
    Customer describes a business problem to a Sales/Support person
    that does not understand their business.  Next, the sales team either
    pulls in some third party that does understand or locks themselves in a
    room and figure out (configure) a solution.
    
    This approach requires enormous skills in our products and how to
    relate them.  It also puts us at great exposure since every config
    is custom.
    
    This is the "understanding products well enough to make them do useful
    work approach".
    
    Another approach is...
    
    The sales rep/team is well trained on the industry of the customer.
    That person can look at what is normal for the industry and what the
    customer is currently doing.  
    
    An example of this in insurance would be something like...
    
    -Claims processing cost 20% of gross for medical insurance accross the
    	board
    -Company X is spending 30%
    
    This descrepancy is an oppurtunity.  
                
    The next step is to reach into your tool bag and pull out the proven
    solution that has worked 10 other places.
    
    This is the "applying proven solutions approach"
    
    This sell is done to business managers on the basis of reducing cost or
    improving profits.  It has nothing to do with technolgy.
    
    To sell this way, I believe industry marketing must...
    
    - Create training to make the sales team industry knowledgeable
    - Gather "norm" data about the industry
    - Create "packages" of solutions to solve problems.
    
    In reality, both forms will always be needed.  Today we do the first
    in most cases.  When we do the second it is becuase the local team
    has done it.  As we move toward the consultative selling approach
    are value added goes up (we are directly putting money on the
    customer's bottom line), we can proactivly go after deals instead of
    being told when they exist.
    
    Ken Slattery

    re: .1   "I believe industry marketing must..."
    
>    - Create training to make the sales team industry knowledgeable
>    - Gather "norm" data about the industry
>    - Create "packages" of solutions to solve problems.
    
    I don't disagree with your approach but:
    
    When I was in Sales Support, I got about 3 messages a day inviting me
    to some "Industry" event.  Kickoffs, "training", boondoggles of every
    shape and size.  The few that I went to usually ended up being a cry
    for help from marketing people looking for input from the field,
    followed by a "team-building" exercise.  Notably absent from any of
    these events were industry speakers from outside the company.  After
    awhile, I got tired of wasting my time, and shortly thereafter local
    management put the squeeze on expenses related to these events.
    
    The norm packages you speak of do exist, they contribute to the
    plethora of different part numbers and prices that we have for
    basically the same widget.  We don't need any more of these.
    
    So, where does the answer lie?  Are our marketing people doing the
    right thing, just not very well?  Or do we need to throw out the
    current stuff and look for a new approach?  The only thing I'm sure
    of is that we can't afford to stick with "ol' tried and true" methods
    in this day and age.  We have to innovate again, or the company is shot.
    Marketing is an excellent place to start, followed by Sales.
    
    Geoff Unland in Austin
    

    RE: .17
    
    Your analysis is a bull's eye...
    
    I have drawn the same conclusions as you.
    
    Industry marketing must change from an organization that reports data
    from the field to upper management (this can be done with a
    spreadsheet) to one that supplies value to the person on the street or
    the customer directly.
    
    This will require a HUGE change...
    
    If we don't do this we will only be a vendor, not a business partner.
    
    Vendors sell commodities on price (sound familiar) and thier margins
    approach zero.  Business partners get paid for the value they add.
    
    I was recently at an Insurance Idustry marketing gig.  This one was a
    very good first attempt at doing the sorts of things we need to do.  I
    don't know if other Industry marketing groups are doing the same.
    
    RE: comment about lots of packages for the same thing...
    
    You are absolutely right.  Instead of creating a new part number, I
    would suggest a system or person (a system would be better and cheaper)
    that embodied the rules to put together the standard part numbers to
    create a "custom package".  The value in a package isn't that it is a
    single part number (although in Digital this has value since our
    admin systems are soooooo bad).  The value is that it is the right
    combination of things to solve the business need.
    
    Ken Slattery
                                                     

    Re 2164.4
    
    I agree that SW complexity is one of the most confusing issues we face 
    today.
    
    However, we have existing sales tools that address the SW issue.  The
    Software License Configurator (SLC) guides the user through a selection
    process to help determine the correct License part number.
    
    SWIFT (Services & Warranty Inquiry Field Tool) enables the user to
    identify the environment (systems/networks/Multi-Vendor ) and tailor a 
    solution for that particular configuration.  Users can then
    automatically add the part numbers back to the AQS Quote.
    
    I am the business manager for SWIFT.  SWIFT is a 3 person team that
    simplifies the Software Product Services selling effort.  SWIFT
    enables a user to tell the system that a customer has a number of
    systems, with the appropriate software on each system and then asks the
    user what services are desired.  SWIFT eliminates the need to
    understand the QT part number structure, while still ensuring that the
    correct solution is built.
    
    SWIFT has been available for almost 3 years.  Customers are currently
    using SWIFT via the E-Store to order their own services.
    
    We as a corporation must understand the need to provide the selling
    team with easy to use, automated tools to assist in the selling
    process.
    
    Kevin Duhaime
     

And a typical question might be : "Why don't Lotus or Microsoft need such
tools".
								- Ian W.

    Re: -1 
    	
    	You raise a valid point.  I was merely responding to a claim that
        no automated tools exist to currently assist the selling team.
    
    	I fully agree that we need to simplify our portfolio of services. 
        However, I think it's realistic to accept the fact that due to our
        strategy of offering services on both DEC and other Vendor's products,
        the need for an automated tool will always be here.
    
    	Studies have shown that the average turnaround time for a DEC quote
        is close to 16 hours (including administrative/research/etc).  If
        existing automated tools can assist in lowering this cycle time until
        the portfolio can be simplified, we must utilize them.
    
    
Kevin Duhaime

    RE: -2
    
    Ian, there are two answers to your question.
    
    The first is that it should be a goal to make it so easy to understand
    this stuff that it is obvious to everyone.  Like MicroSoft.
    
    The second is that we are in a different business (more businesses)
    than MicroSoft so it will be more difficult for us.  In addition to
    this, as MS gets to more complex stuff, their configuration gets
    somewhat more difficult.  For example, configuring MS mail server (as I
    understand it) is not straightforward.  They are much better than us but 
    not always obvious.
    
    In the end, as with most things, the answer is somewhere in between.
    
    In a prior note, I stated that we had 6000 answers to everything.
    
    We have 336 variations of 
    
    over 25 basic models
    
    -4 3100s
    -6 4000s
    -7 6000s
    -4 7000s
    -4 10000s
    -?VAXft
    
    of 4 chips
    
    -CVAX
    -Rigel
    -Mariah
    -NVAX
    
    We could probably eliminate at least 2  of the above chips
    We could eliminate half of the basic models
    We could have one and only one variation of the basic models
    
    This would simplify things greatly (and probably not limit real choice)
    but we would still have 12 or so hardware platforms on which
    
    Services can be configured xxx ways
    Software can be configured 3 ways plus variations of those
    The pieces can be interconnected in infinite ways.
    
    This is still complex.
    
    RE: -1 and the study of 16 hours to complete a quote...
    
    This is great data!!!  I think the numbers sound fairly accurate.
    
    This 16 hours should approach 0.  Practically, anything more than
    30 minutes should get sales management on the warpath.  This is
    unproductive time and is a large cause of the fact that our cost of
    sales is about 35% when the industry average is about 24%.
    
    Some arithmetic...
    
    16 hours = 2 days
    There are about 22 days/month*12-20(holidays, vacation)=244days/year
    
    10%(approx. delta between our cost of sales and industry) of 244 = 24
    
    24 days = 12 quotes.
    
    The average rep does far more than 12 quotes/year
    
    If we got this down to 30 minutes, our cost of sales would get back to
    the industries.
    
    This arithmatic may be a little tough to follow, but I hope it makes
    some sense.
    
    In addition to this 16 hours there is probably an equal time spent
    "preparing" to sit down at AQS.
    
    RE: SLC and SWIFT
    
    Any tool is better than none.  I have used both of these in the past
    but have not recently.  The reason is that I drew the conclusion that
    as of the time I used them, they didn't really help.  It is my opinion
    that for whatever reason almost no one in my office uses these tools. 
    The reasons could be training, functionality or other things.
    
    Basically, they help if I already know what I need.  For example, how 
    does service effect the cost of the 6000 answers to my question (not 
    the one that I decided to sit down at AQS with).  My point is that, by
    the time I sit  down with these tools, I already have to have gone 
    through the info they provide on paper.  
    
    For example, SLC helps me decide whether to do Personal, Concurrent or
    Capacity licenses.  The answer to this question depends on the system
    you are configuring.  Capacity may be the best deal on a 3100,
    concurrent on a 4000 and personal on a 6000.  As I decide which way to
    go (before AQS since it would literally take forever to configure all
    the possibilities) I currently use "Rules of Thumb" since the hard data
    is not in a form that I can use it.  By the time I sit down to AQS I
    have committed (right or wrong) to one or a few systems.  I have
    probably done the arithmetic by hand on which license to use.  If I
    haven't, SLC can help some but not much (I have already done most of my
    work).
    
    This gets even worse when you have to make decision like...
    When is a NAS package a better deal than a la carte.
    
    If these tools currently do these things, then I should look at them
    again.
    
    By the way, I have created a prototype tool that does exactly the 
    things I am discussing here.  People in my office that have seen it
    absolutely think I am on the right track.  I have "presented" it to
    people responsible for these things and have gotten no real response. 
    The tool runs on MS-Windows on a standalone PC.
                                                                       
    Ken Slattery

    i had a friend here at DEC that was working on increasing productivity
    of the sales force.  Tools was a problem they wanted to address but
    before they got to that they found there were two areas that by in
    large had to be addressed first.  One was basic understanding of
    business and the other was basic understanding of computers, etc.
    I understand pilots in both areas were undertaken.

    I chaired a corporate sponsered "study" on improving sales
    productivity.  Number one goal was to spend more time with the
    customer.  Number one impediment was having to focus internally too
    much. 
    
    Sales people need sales tools, but more importantly, they need systems
    behind them that work.  

    RE: .24
    
    What was the outcome of the study?
    
    Is anyone acting on it?
    
    Is anyone taking suggestions for tools?
    
    Ken Slattery

    The outcome was a beautiful document.
    
    No one is acting on it.
    
    We really did not focus on tools, rather on the business/information
    needs of sales people.
    
    Like the need to spend more quality time with customers and not chasing
    logistical information about orders that should have shipped or will
    bew shipping or what happend to that BC16e-25 cable.... or digging
    around looking for competitive info....(should be shipped auto)
    
    

    Just doing a simple upgrade for a customer can take alot of time
    and internal systems effort.
    
    I am in Sales Support.  We just presented a proposal to a manufacturer
    for an upgrade to their VAXcluster.  We did all the right steps and
    here is how much time it took me:
    
    1) DECps and Capacity Plan - Done by EIS so other than a workstatement
    it took no time.  Confirmed, "Yup, gotta get bigger VAX systems".
    
    2) With 3 nodes, there were endless possibilities (lots of ways
    to get to more VUPs):
    
    	- Upgrade to more CPUs in for their 6000 class machines
    	- Get new single processor 6610s and trade 6410s
    	- Get one, two or three 7610s
    	- Lease, buy, buyout lease then trade in, or sell on open market
        - Will all their software work on a 7610 (new version of OpenVMS)?
    
    This took about 1 full week of time to narrow down the possibilites
    that we wanted to present.  We came up with three alternatives that
    would meet their needs.  We presented them, and left the meeting
    with three more alternatives they wanted us to price out (valid ones).
    
    4) Software licensing, and figuring out how much it would cost, was
    unbearable.  We researched:
    
    - Are they license-compliant as is?  (No, they were out of compliance)
    - What would it cost to get them compliant, then upgrade?
    - Should they do clusterwide, or newer user-based licensing -- for the
      20 or so layered products they have!  Which is most cost-effective?
    - Remember that this needs to be done across 6 scenarios now!
    - We brought in a licensing consultant, and he worked a couple of weeks
      on it.
    
    5) AQS quoting these scenarios also took a long time.  I had to
    get all the part numbers written out, for 6 scenarios, 3 machines
    and all the products, send it to our CAS person in ALL-IN-1, only
    to have it all re-typed in as a quote.  That took a day or two.
    The quotes, which were basically unreadable, were relegated to an
    appendix of supporting documents with a hand-done summary sheet
    presented to the customer.
    
    6) Finally, creating the proposal and presentation package took
    several more full days, each of the three times we did iterations.
    Numbers kept changing. "Oops! Forgot we need 600 point licenses on
    scenarios 3, 4 and 6 but not on 2. Unless they lease it."
    
    Now, this is a "simple" VAXcluster upgrade.  No systems integration,
    all-DEC shop, no third-party involvement.
    
    There is no way an 800-number sales support person could have done
    this!  And this was an easy one!

    Re .27 is a perfect example of why we are losing so much money. We have
    too many overlapping products, incredibly complex (and expensive!)
    licensing schemes, and entirely too much 'process' and 'processors'
    built into the task of selling to our customers. (Too much process and
    processors are also built into nearly every task at Digital. We can't
    keep supporting this level bureaucracy with our current margin levels.)

    I wonder how often sales folks spend this much time and effort on a
    single customer only to lose the sale? One only need to read .27 to
    figure out why our cost of sales is so high. Extrapolate that same
    level of red tape to the rest of the company and you can figure out why
    our general and admin costs are also so high. :-(
    
    Harry
    

The problem is about to get worse for those of you that still work mostly 
with Hardware as DEC finally makes some SW available on other peoples HW.

I sell PowerFrame which has only 4 types of licenses, running on 2 Sun 
Architectures, 3 HP architectures, IBM RS6000, NEC, and 2 DEC architectures 
with 2 OSs.  This combination has 234 part numbers as of now.  This is 
beyond stupid, it is laughable.  The Product Manager has been spending 
months trying to simplify this before we release support for 4 new 
architectures.  

How many part numbers does it take to define a product?

For profitability reasons we would like to break out some parts of the 
product, but making the current 4 types of licenses into 16 seems to be a 
nightmare.  Because of the archaic part numbering system that assumes 
people run all of our software on DEC boxes, we are limiting income.  

Matt

    right, and the day after it is announced some clown will ask for variant
    #235 and tell you they can't sell it without that one.  ;)

There are a lot of problems that make it hard for the customer to give
us money, including many that make it hard for the salesperson to know
what to recommend.  The number of part numbers is not one of the problems.
Minute differences that customers do not care about are problems.
Small differences that customers care about are features.  We need to
do a better job of identifying what customers want.  We need to make it
easier to introduce families of products, so adding a ninth language
or a seventh platform is adinistratively simple. We need regularity:
now models a, b, c, d, e, g, h, and i come with a power cord.
We need lots of different kinds of improvements, and I bet some of
them will reduce the size of the price book as a byproduct.
Treating the size as something to manage will cause even more problems.

Some examples:

In the U.S. income tax revisions of about 1985, the number of tax brackets
was drastically decreased.  Did that make it easier to prepare your
tax return or do any planning?

Does anyone here think sales would go up if we removed every n-th item from
the price list?

This reminds me of the common practice of "managing" the customer
satisfaction surveys, so we can insure that we can not learn from them.

Tirade over. Back to your regularly scheduled rathole.

    in response to .31
    
    Well, if we never sell something, it doesn't need a part number.  But,
    if a customer ever asks "Can I get sw XYZ for platform ABC?" we need to
    be able to give an answer.  Maybe we need to have a system that answers
    that question efficiently and then assigns a digital pn only when we
    actually get a customer's order?
    
    just dreaming,
    
    Dick

    that's how we do it in the EIC's.  Custom part numbers for each
    solution.  When the quote goes out it has a part number in it but that
    part number doesn't go up on the file until we're sure the customer is
    going to buy.  And after we deliver we take it off.

Have a two part number, say Part/Platform

Then you would have Parh XXYY0791 - FORTRAN 90/Parallel

And say
		Sxxx - For sun platforms
		Mxxx - For Mips
		Ixxx - I**
		VOxxx - For Vax/OpenVMS
		Axxx - For Alpha AXP (tm)


etc.

That way you have two lists, and a Matrix could tell you if it exists.
(Or a laptop could give a simple answer).

Why not allow us to have one part number for a license on any platform and 
then part numbers for Media and Docs?  We instead have growing part numbers 
that add complexity, but no value to the sales and collection functions.  I 
am not asking to remove sellable objects from the parts list, simply to 
rationalize it.  A simple quote took an experienced sales person 7 hours in 
AQS due to the complexity involved.  the complexity had nothing to do with 
the custi environment, simply our part numbering system. 

I am moving to bundling SW and Service together and selling it through a 
local Services manager( sorry, I can't keep straight the name of services 
this week).  He can bundle everything together, I can supply a commitment 
for engineering, and then deliver a fixed price quote to the custi saying 
it will cost $x, take y months, and result in a working system.

This is how I see quotes from Anderson, EDS, SCS, and other qualified 
service companies.  On a simpler level, it is actually what Microsoft does 
where the service included is rather limited.

This method also allows me to create a quote in front of the customer that 
is a rough order of magnitude, naturally done using resonable tools like 
EXCEL, Word and PowerPoint.

One price, one part number, two page quotes. 

Matt Haley

    RE: .31 and a few others
    
    The number of part numbers is a symptom of the problem, not the problem
    itself.  We have so many part numbers because things are too complex. 
    Not the reverse.
    
    I think this is basically .31's point...
    
    To clarify some of the things I have stated in prior notes...
    
    In a prior note I advocated knocking down the
    336 VAXes offered to about 12.  This was not to limit Part Numbers or
    choice.  It was to only offer choice where it was relevant and to
    simplify the process of configuring them.  I don't think that this
    would reduce our ability to solve customer's needs AT ALL.
    
    If I only have one base system to choose from on which I add all
    options I don't have to price out several base systems to see if their
    option package is a good deal or not.
    
    For example, at Wendy's you can get 3 types of hamburgers (single,
    double, triple) in 256 combinations each (Ketchup, mustard, pickles,
    lettuce,onion, tomato, mayo, cheese and all combinations of these)
    
    Does Wendy's menu have 3 * 256 = 788 lines for hamburgers...NO!!!!!
    
    It has...
    
    single	1.20
    double	2.30
    triple	3.05
    
    Free
    	Ketchup, mustard, pickles, onion, lettuce, mayo
    Additional Cost
    	tomato		.10
    	cheese		.15
    	
    Would you buy lunch at Wendy's if you had to understand a menu that had
    788 lines in it just for hamburgers?  Or, could Wendy's afford to be in
    business if they had to employ a "Sales Support" person to explain
    their menu to each customer.  
    
    
    Here's another thought...
    
    Why do we need part numbers anyway...what is wrong with the English (or
    whatever) language.
    
    Why can't I buy...
    
    A VAX 7000 with
    	256 meg of memory
    	4 2 GB disks
    	ALL-IN-1
    	Rdb
    	NAS 300
    
    Why do I have to translate English into the DECspeak of part numbers?
    
    The answer is because our internal systems need it and we insist on
    forcing our customers to solve our internal problems.  I can't begin to
    count how many times my customers complain about our impossible to read
    quotes (the document we FORCE customer's to use in order to do business
    with us).
    
    When you buy a car you buy it with...
    
    Air Conditioning not AC-34596-GP
    Air Bag          not AB-58837-PQ
    Sun Roof         not SR-ROOf93-OK
    
    I bought a PC about a year ago...
    
    I bought (from the invoice)...
    
    33 MHz 386
    4 MB
    85 MB Disk
    Color SVGA Monitor
    
    Ken Slattery
           

re .36

I agree that part numbers are the a symptom, but they are also an easy 
thing to fix.  WE are at the point where I must write an English version of 
the Service Description and the Statement of Work when I negotiate with 
custis.  This is stupid.  Everybody admits that we want the custi to 
understand what they are getting (we call this "Setting Expectations") and 
then set up systems that ensure the custi can't understand what they are 
about to get.
    
>    The number of part numbers is a symptom of the problem, not the problem
>    itself.  We have so many part numbers because things are too complex. 
>    Not the reverse.
 
I agree if I read this correcly, the "things are too complex" refers to the 
Digital "things", not the actual business problem we are tying to solve.

    
>    Why do we need part numbers anyway...what is wrong with the English (or
>    whatever) language.
   
>    Why can't I buy...
>    
>    A VAX 7000 with
>    	256 meg of memory
>    	4 2 GB disks
>    	ALL-IN-1
>    	Rdb
>    	NAS 300

Because this makes too much sense and therefore some group will say we need 
a common language (part numbers) for all parts of the world.  They forget 
that we must have a common description first or we could not have a common 
part numbering system.

matt haley

>    	256 meg of memory
    
    You mean '256 MB of memory', right ?
    
    ;-)
    
    karl

    
    simplification of the product mix has to occur.  We no longer achieve
    the margins necessary to afford allthe variants.  Choosing the right
    products to be part of the simplified list is not so easy a
    proposition. Aside from what's selling, theres the complicated job of
    untangling the internal politics and web of interdependencies, which if
    done incorrectly or haphazardly can cause much pain also.  I have no
    doubt that we will accomplish this, but we won't solve it in a day or
    a week.  Maybe a quarter or 2 though!
    
    bob

    RE: .39
    
    First cut at the new VAX hardware product mix
    
    MicroVAX 3100 Model 80   priced like a model 10e
    MicroVAX 3100-90/4000-100 same product priced like a model 90
    
    VAX 4000-600 priced like a 4000-500
    
    VAX 6000     empty cab you add as many boards as you want(this could
    			actually be dropped because of 7000)
    
    VAX 7000     empty cab you add as many boards as you want
    
    VAX 10000    does not exist an add-on to the 7000 is the extra stuff
    
    All systems are stripped (no memory, disk or software)
    You add these options 
    
    This just brought 336 part numbers to 6.  This has not limited anyone's
    choice.  It is simpler and therefore better/more cost effective for sales
    and the customer.
    
    
     Ken Slattery
	
                                                    

    
    Hardware products ain't the issue folks.  Most of my questions involve
    SOFTWARE.  I don't know how many times I've tried to explain the 
    differences between X, Motif and IXI X.desktop and what they do  
    and don't do to some braindead individual.
    
    Here's one for ya: just TRY to find a salerep who can explain the 
    features and benefits of the various NAS packages!
    
    -Ed_Sales_Support_SOFTWARE_Consultant
    
    
      

re .41

I will find a sales rep who can explain the features and benefits of NAS 
when you can find an architect who can explain teh architecture as anything 
but a string of loosely tied existing products.  NAS as a solution can help 
solve a custi business problem, but not before the custi says what the 
problem is.  No product has benes just because the vendor says it does.  A 
product can only have benes after the custi explain the problem she wants 
solved.  Benes respond to neeeds and wants, not to some silly brouchure.

matt haley

dumb_sales_rep_who_only_makes_quota_on_selling_software

    Re: .42
    
    Hey, nothing personal dude!  I am more critical of NAS than you are!
    
    Point is this: too many sales types cannot communicate benefits
    because they cannot grasp the technical aspects of the products 
    they sell.  You are obviously an exception to this rule.  
    
    -Ed
    

Sorry Ed, I obviously over reacted.

I will chill out!  I will chill out!  I will chill out...

matt haley

I can tell you one way to save this company thousands of person-hours of
field personnel time and much aggrivation:

Build a good quote system!  

Anyone who has had to use the AQS system can understand that.  It has a 
very poor user interface and response time, and it can take forever to make 
simple changes to a quote.

gary

    RE: various since 
    
    Ed Chaban re: software being the issue
    
    I agree...  It is at least as bad as hardware...
    If you want yet another nightmare, we can take up services...
    
    I usually use hardware in my examples because people can "feel" it.  I
    have equal concern about the three levels of VMS (Base, Trad,
    Advantage-Server), three forms of SW licensing (Capacity, Concurrent,
    Personal).
    
    When I discuss these, people's eyes usually glaze over.  Therefore, I
    start with hardware.
    
    RE: Sales Reps needing to understand NAS...
    
    Ingoring the other issues with NAS, I disagree that a Sales Rep needs
    to be able to articulate what NAS is/does...
    
    In an earlier note I discussed Value selling and having reps
    concentrate on how to solve customers business problems.  This does not 
    require knowledge of NAS.  What it requires is an understanding of
    the business, what leverage points exist for improving the customer's
    business and whether we have been successful doing this in the past. 
    
    I do think that marketing should create a few "sound bites" about what
    NAS and all our products do at a business level that all reps should
    know/understand.  This doesn't exist.
    
    Since this would not cover all situations (many are new problems)
    Sales Support should know this stuff cold.  A Sales Rep that can do 
    this and the other stuff is better than one that can't but I would
    concentrate on making Sales better at business first.
    
    RE: needing a better quote system...
    
    I believe that AQS is a resonable system when you accept it for what it
    is...
    
    It is a system that accepts part numbers, verifies that they exist (not
    correct), applies discounts and prints a report.  The report stinks but
    the rest of it does what it is supposed to do.
    
    Could the interface be better?  Yes...
    
    Instead of doing that, though, I would create a "Configuration
    Assistant" that had you enter data (in MS Windows) like
    
    1)  What software do you want
    2)  How many users or transactions will this system support
    3)  What systems exist and how much more capacity do you want
    4)  How many VUPS, Disk, Memory do you want
    
    Then give you a comprehensive list of all forms of systems and configs
    that meet those requirements with pricing.  This should do this in less
    than a few minutes.
    
    The output would look something like...
    
    VAX 7000-610	600,000	DSSI is least cost, then SDI, then Cluster
    VAX 6000-610	585,000                             "
    VAX 4000-600	367,000 Single node (dual host not needed)
    VAX 4000-500	300,000                          "   
    (2) VAX 4000-400	357,000 Dual Host needed to achieve performance
    (3) VAX 4000-100	257,000 Thi-Host needed to achieve performance
    
    Next, the user (Sales Rep) looks at all the answers and chooses the
    ones that look most interesting.  
    
    Various options exist now.  They are:
    
    Produce Quote  -  This would go to an MS-Word Template and be
    			customizable
    Produce ROI anqalysys - Excel spreadsheet. 
    
    The technology to do this exists, it is only a matter of spending a few
    months cleaning it up.  No one with funding is yet interested.
    
    The proto accurately configures all a la carte 4000s, 6000s and 9000s
    as they existed in December of 91 as standalone, DSSI or CI clusters.
    It also accurately configures 5 software products (ALL-IN-1, Rdb,
    DECforms and two others that I can't remember now).  The reason it
    doesn't do more is because data entry is a hassle.  It configures in
    about 30 different systems in about 5 seconds on a 16 MHz
    386.  Compare this to 16 hours to do an AQS quote that someone brought
    up before.
    
    It does not yet do cables but can easily be extended to.  It doesn't
    deal with Personal licensing because it was done before these existed. 
    As with cables, this is an easy add on.
    
    Ken Slattery

    
    NAS at the moment can only perceived as added value internally. It is
    an attempt, long over due, to force some systems interaction into a
    series of point products.  I see it more as a process isuue - try and
    force issues of interoperability and synchronization of delivery dates
    for our software packages.  Up until now, all we've ever done is
    release point products. Point products cannot compete functionally with
    Lotus Works or Microsoft Works etc.  
    
    I've called it GAS just as long as any of you.  But the idea of
    building system like solutions is not new, and this may be the best bet
    vehicle(even if it is a hot air balloon) to drive the needed process
    changes(ie: a goal you get measured and payed for) to get from here to
    there!
    
    Even if the package idea does not get very far, we may be able to get
    some milage out of forcing products to make use of each other,
    simplifying option selection.
    
    bob

    Re: 2164.45
    
    Ken,
    
    	I couldn't agree more with the concept you propose.
        I fully support the idea of qualifying the customers environment,
    	recommending equipment/product/services and presenting the solution
    	in a clear, consise format.  
    
    	The volume of SW part numbers is staggering.  I have been told that
    	over 60% of the current price file is SW products and services.
    
    	There are steps being taken to simplify the process.  However,
    	in the short term, I believe we need to communicate  the existing
    	tools we have to assist sales.
    
    	Here's an example;  If Sales wanted to sell All-in-1, there are 82
    	different model numbers that represent the license for All-in-1.
    
    	From a services standpoint, there are 204 part numbers.
    
    	Rather than rely on price books, Sales Updates and other manual
        tools, I would recommend using the Software License Configurator and
    	SWIFT.
    	
    	The qualification tool you discussed is a great idea.  If that tool
    	could be the front end of the cycle (qualifying needs) and the
    	existing tools we have (SWIFT, SLC etc) act behind the scenes to
    	translate the needs into DEC part numbers, we have provided the
    	selling team with a useful tool.
    
    Kevin Duhaime

    re .48
     Unfortunately we don't sell All-in-1.  We sell ALL-IN-1. 
    
    Which is part of the problem.  On any given day, Sales/Sales Support
    people incorrectly spell at least half of our trademarks; given
    the fact that additional salvos come in the form of product name changes,
    it's amazing that any customers know what the dickens we are trying
    to sell them (remember PCSA, no, DEC LanWORKS, no PATHWORKS?)
    
    Dan
    Official Self-Appointed Local Watchdog of
    the DIGITAL logo and other trademarks

    Kevin:
    
    You and I have spoken in the past about SWIFT and I have often
    mentioned it to sales reps who seemed totally frustrated when trying
    to configure SW licenses or service part numbers.  The answer is 
    always the same.......SWIFT is anything but......SWIFT is too slow!
    
    More than one sales rep has indicated it is faster for them to call
    us at 1-800-DEC-SALE than to try and use SWIFT.  The reps are not
    knocking the idea of SWIFT, but most feel it is too cumbersome to
    use.
    
    Karen
    

    Karen,
    
    	I appreciate your feedback.  However, in light of recent
    	enhancements that we've made to SWIFT, I wonder if SWIFT's
    	responsiveness is still an issue.
    
    	SWIFT used to require the user to enter a system and then configure
    	services associated with that particular system.  If there was another
    	system in the configuration this step would need to be repeated for as
    	many systems as the customer had.
    
    	In our latest version,  V2.3, SWIFT release the EXPert module. 
    	Simply put,  there are now three categories of information SWIFT
    	requires;  the first is to identify all systems in the configuration. 
    	The second is to identify the software running within the configuration
    	and the third is to identify the service requirements for all systems.
    
    	Using the EXPert module, the ammount of time needed to build a
    	configuration using SWIFT was cut by over 50%.  Keep in mind that in
    	most cases, using the "regular" SWIFT was still faster than the manual
    	tools provided to the selling teams.
    
    	SWIFT is built exclusively for field use.  I as the business
    	manager, encourage any feedback on the application.  If I receive field
    	feedback on SWIFT, it receives the highest priority.  If we provide
    	automated tools to help te selling teams, we better make sure the tools
    	meet the needs of the field.
    
    	Thanks again for your feedback and keep up the good work!
     
    
Kevin Duhaime

    Kevin,
    
    I had entered a rather detailed note, but lost it when the system
    bought the farm.
    
    I will try to pass on your information regarding the enhancements,
    but bottomline today is that most reps are going flat out; when they
    call us many of them are already in the midst of the quote and just
    need a couple of part numbers.  Since the name of the game is to
    get that quote out ASAP, if there is any way they can invoke SWIFT
    without backing out of the quote perhaps some will try it, but I
    wouldn't bet the farm on it :-(
    
    At one time I wanted to learn more about AQS myself, but since it is
    not my job to generate quotes my chances of getting the training are
    somewhere between slim and none.  Like so many others, I would *like*
    to take some spare time and fool around with SWIFT (assuming it will
    let me do so since I don't have quote capabilities); but time is a
    rare commodity for many of us.  I did obtain an AQS User Guide about
    6 months ago; are the newer enhancements included in it (should I
    ever get a chance to utilize it)?
    
    Karen
    

    Karen,
    
    	Thanks for the feedback.
    
    	Regarding your comment on Salespeople and backing out of the AQS
    	quote to use SWIFT:
    
    	SWIFT is integrated within AQS.  This enables the Sales person to
    	call SWIFT while they are in the middle of building the Quote, use
    	SWIFT to generate the service line items and then automatically pass
    	these line items back to the quote they are working on.  We are in the
    	midsts of an effort to provide the same functionality to users of the
    	Electronic Connection.
    
    	We have done 2 releases within the past 6 months.  If you (or
    	anyone else) would like a copy of the latest users guide, please
    	contact me and I'll arrange to have them sent to you.
    
    	The last release we did provides support for Open/VMS among other
    	enhancements.  SWIFT is the only automated configuration tool to
    	support building the Open/VMS services.
    
    	Our next release, V2.4, will include configuration support for both
    	Alpha and OSF1.
    
	The next step in the configuration/quoting process is to tie into a
    	tool like Ken Slattery described in an earlier reply.
    
    
    Regards,
    
    Kevin Duhaime