http://unixwiz.net/techtips/be-consultant.html
Does this site look plain?
This site uses advanced css techniques
[Steve Friedl Logo]
Steve Friedl's Unixwiz.net Tech Tips
So you want to be a consultant...?
* Home
* Contact
* About
* TechTips
* Tools&Source
* Evo Payroll
* CmdLetters
* Research
* AT&T 3B2
* Advisories
* News/Pubs
* Literacy
* Calif.Voting
* Personal
* Tech Blog
* Evo Blog
[Consultant Graphic]
Or: Why work 8 hours/day for someone else when you can work 16 hours/
day for yourself?
I've been a consultant of one form or another since 1985 when I
started my old company, V-Systems, with a friend from college, and
actually did bits and pieces of consulting as early as 1982. I have
been asked often about the business, and I decided to write this up.
Please note that I am providing observations from my own personal
experience, but I am not providing tax or legal advice. You need to
pay somebody for that, and I'm not qualified.
Furthermore, I am not even attempting to make this a comprehensive
guide for everything required by one in or contemplating the
consulting business. I am purposely omitting whole areas, such as
licensure, insurance, and negotiating -- there are other books for
that, and this isn't trying to be one of them.
These sections (except the last) aren't in any particular order.
Different flavors of Consultants
There are many ways of structuring a self-employed practice, and I'll
touch on two that are at different ends of the spectrum. I only have
my own experience to draw on, so these should be considered broad
generalities rather than pigeonhole-type definitions. I have gotten
substantial pushback on these definitions, so I urge you to take them
only generally.
Contracting
A "contractor" is typically hired for one project (via a
"contract"), and the relationship is often more technical than
personal. One is hired to write a communications controller,
build a website, or modify some software, and at the end of the
project: you're done and you move on.
Most contractors work on one project at a time, which surely
allows a great deal of focus, and though there is a often a bit
of hanger-on work after the project has finished (say, helping
the in-house staff integrate your work), once the contractor has
moved on, the focus goes with him.
This is not to say that you won't be back: doing a good job on
the first project certainly means you'll be given more
consideration for a subsequent project, but it's just as often a
new project as opposed to working on the old one.
Consulting
A "consultant" typically has multiple customers at a time, and
it's more about a long-term relationship than it is about a
specific project. There will certainly be projects in the course
of a consulting relationship, sometimes big ones, but the general
idea here is that you're an always-available resource they can
call on for big matters or small. In a few cases, I am the IT
department for several customers, and they use me so they don't
have to hire a dedicated person for it.
Unlike the contractor, this involves a lot of juggling when the
fluctuating demands of multiple customers comes into play. It's
harder to get a deeper focus because of all the context swapping
going on, but long term relationships are worth it in the long
run.
I mainly operate as a traditional consultant, and though from time to
time have done one-time projects on a contract basis, this paper
mainly addresses the consulting relationship.
What's it take?
The single biggest surprise to prospective consultants is when I
suggest that their technical skills will not be their biggest asset.
Those skills are required, of course, and often help get you in the
door, but the long-term customer relationships on which a consultant
depends are built on one thing only:
Consulting maxim:
You must give the customer The Warm Fuzzy Feeling(tm)
Your customer certainly has to believe you can do the job, but they
cannot wonder if you're going to get back to them, or if you're going
to do something stupid (again?), or offend one of their customers.
Your practice is never more stable than when your customers trust you
completely to take care of them.
I still have my first customer from 1985, plus several others since
before 1990, and I've been asked to do work I was clearly not the
best for, simply because my customer liked doing business with me.
Anecdote:
Many moons ago I knew a gal who was a salesperson for the computer
vendor I did business with, and she was not even close to being as
technically competent as I was. When she left to do consulting, I
figured she'd never make it because of the "lack" of those skills.
But she did make it, and did a damn good job, too. She was good
enough with the subject matter to take care of what was in front of
her, but she was ferociously dedicated to taking care of her
customers, and this was where I learned that The Warm Fuzzy Feeling(tm)
is central to a good customer relationship.
It didn't take long before her subject-matter skills were as good as
her customer-service skills, which made her a gift from God to her
customers: within a wide range of problem space, they would rather
pay her to figure something out than hire somebody else who already
knew how. It wasn't until later in my career that I found this out
for myself.
Thank you, Cindy, for showing me what makes a good consultant.
Consulting maxim:
"Trust" is your best job security
But it takes more than good customer service skills: one must be a
self starter, require very little supervision, and the ability to
keep on task without a boss looking over your shoulder.
Working at home can be a wonderfully comfortable and productive work
environment, but it's not meant for everybody. Distractions abound --
spouse/children, the television, that really comfy couch -- and it's
really easy to burn tons of time without getting anything done.
Working by yourself requires substantial time-management discipline.
Though some consultants (like me) prefer a solitary work-at-home
existance, it doesn't have to be this way. Many consultants spend
much of their time on the road, with plenty of human contact, and you
can tailor your practice to have whatever mix you like.
Job security
Notwithstanding the prior maxim, the short answer is that you don't
have any.
Consulting maxim:
You have no job security, even if you think you do
Consultants -- even good consultants -- are often considered a
necessary evil by customers who use them, and it's exceptionally easy
to stop using one. It doesn't require firing or confrontation, just
"we don't need you any more" or even just not calling.
This has happened to me several times: a sure thing with many years
of history stopped on a dime when a customer's customer canceled a
project, or when another customer was bought by a bigger enterprise.
This is not personal to your customer -- it's just business -- but when
your gravy train evaporates without notice, it's very personal. And
terrifying.
The best way to approach this is to diversify -- if you have only one
ongoing project, you can find yourself out of work with no more
notice than the time it takes the phone to ring. By being aggressive
about finding more than one customer, and even making it a point to
value a backlog of work, you have a fighting chance at dodging the
vaporizing-customer bullet.
Cash Flow
Beyond having regular ongoing work lies matter of cash flow. Even if
you regularly invoice at the start of every month, customers have
their own schedule for paying, and this can be nerve-wracking to deal
with.
For many years I operated on net-30 terms -- payment due within 30
days of the invoice -- but when you add postal delays, waiting for
manager approval, being put on the next regular check run, and
sitting on the president's desk for a signature, it's not hard for
invoices from even good customers to take 45 days to arrive in your
hand.
I moved to net 10 terms, and this made an enormous difference in cash
flow: a few customers simply cannot accommodate that fast of an
invoice turnaround, but it's been a big win. This is a soft 10: I
tell my customers that the first regular check run on or after net 10
is timely, and this usually means 10-15 days. What a huge difference
this made in cash flow.
Where this still gets tricky is when a customer is a bit late for
whatever reason: your creditors don't really care. But a consultant
can never do anything other than perform routine collections with a
customer. Asking for a stale invoice to be paid is fine, but it's
exceptionally bad business to give the customer any hint that you're
in a bad place.
Consulting maxim:
A financially-struggling consultant does not give a customer The Warm
Fuzzy Feeling(tm)
No matter how tight your bind, your customers can't find out.
The Customer Relationship
I've mentioned several times The Warm Fuzzy Feeling(tm), and I'll do so
again: this is absolutely central to how you manage your practice.
Have "customers", not "clients"
This is a minor semantic point, but one I've stuck with for many
years. A client implies that the consultant is superior, while
customer suggests that the consultant is beholden.
It's virtually impossible for a consultant to forget the
technical nature of the relationship -- you solve problems all day
long -- but it's very easy to forget the customer service aspect.
We want to avoid this as much as possible.
Medical doctors are notorious for getting this wrong: we've all
had a doctor we really loved (The Warm Fuzzy Feeling(tm)), but the
office staff didn't measure up. We are a patient to the doctor,
but a customer to her medical practice.
I've never heard of a doctor or a lawyer use the word "customer",
but as a consultant I am proud to. I exist at the pleasure of my
customers, and that means being easy to do business with.
Consulting maxim:
You are primarily in the customer service business, not the
technical business
Be exceptionally easy to find
When I go into the bank and find a long line to reach a teller,
it's of course frustrating. Mentally, I start a timer in my head,
and the longer the timer goes the worse of an experience it is.
What stops the timer? Leaving the bank?
No. -- It's reaching the teller.
When your customer pages you, his timer starts: return his call
immediately. If the customer has an emergency, it's much better
for him to hear your voice 5 minutes after paging saying "I am on
the road right now, I can jump right on this in an hour" than to
just sit around waiting, wondering if you got the page.
Consulting maxim:
For a good consultant, your voice is comforting: Be very easy to
find
When the customer knows you're going to be on the case, he can
stop worrying and attend to what's in front of him. In my entire
career, I have never dodged a page or screened a call even though
I was tempted more than once.
Billing
[Billing icon]
There are all kinds of arrangements for getting paid, and none is
really superior to any other. I have virtually always operated on a
strictly hourly basis, invoicing once or twice a month, but it's also
possible to do monthly retainer, fixed-bid contract, or some other
combination. And no method of billing can avoid talking about
churning, which I'll define also.
Hourly billing
This is the easiest to manage: you work an hour, you invoice the
customer for a hour. For occasional or ill-defined work, it's
hard to use anything but hourly billing. The customer bears the
brunt of projects that get out of hand, and the customer is
really at the mercy of the consultant for being fair. This makes
many customers nervous for fear of project-creep and "churning"
(defined below).
Consulting maxim:
Hourly arrangements of any substantial magnitude require that you
have earned your customer's trust
Retainer
There are many variations of retainer arrangements, but one
version is where the customer purchases a guaranteed minimum
number of hours per month (at a substantial discount), and they
are worked off as the month goes on. These are particularly
popular for outsourced IT departments, where you form a remote
help desk, servicing issues as they come up.
They are good for the customer, because they get a significant
discount for a consultant, and good for the consultant, who has a
predictable income stream every month.
Fixed-bid projects
When a customer has a specific project, you provide a fixed
quote: "I can do that job for $X". Customers often prefer these,
because they know in advance what they're going to pay, but they
are the most problematic for the consultant.
These require highly detailed specifications that list exactly
what is expected of the consultant. If it's not entirely clear
what the requirements are, there will be endless disagreements
over whether this or that is in the scope of work: The customer
will think it is (you work for free), and you will think it isn't
(customer pay me more).
I have relatively little experience with fixed-bid contracts, but
I do know that your estimate should include a substantial fudge
factor. This is not gaming the customer but allowing for
surprises. All programmers are optimists, and all projects run
into unexpected roadblocks: your fixed bid means you allowed for
that.
Consulting maxim:
The best way to appreciate the value of a good spec is to do a
project without one
Generally speaking, ambiguities in a specification must be
resolved in favor of the customer (you're the consultant, you
should know how to do this), though it's certainly possible for a
customer to try to take advantage of you. A good customer will
take into account extraordinary roadblocks that nobody could have
anticipated (say, a critical vendor goes out of business), but it
has to be much more Earth shattering than "it's harder than I
thought".
It's not unheard of for a consultant to go back to the well on a
fixed-bid project, laying out a case why the project has gone
over and why you need to be paid more. This is always
uncomfortable, and the only way to manage this is if the customer
has been in the loop in fairly substantial detail throughout the
project.
If you've been telling the customer "it's going great" for six
weeks, you're going to have a hard time showing up on the due
date explaining why you're not finished and also asking for more
money. Instead, if you have been making regular detailed progress
reports, even including your own mis-steps, it's more likely that
problems will be headed off at the pass, and the customer will be
more amenable to sharing the cost of surprises.
Consulting maxim:
Customers hate unhappy surprises much more than timely bad news
A customer recently made a point that's essentially the same
thing:
Consulting maxim:
Bad news does not get better with age
Churning
This is the all-too-common phenomenon of a consultant running the
clock simply to run the clock whether the customer actually needs
the work or not. This could be outright padding (aka "lying") on
an invoice, or recommending/doing work that's not really
required.
Consulting maxim:
Churning by dishonest consultants is the single worst thing that
has ever happened to honest consultants
Every consultant has projects that run over, and sometimes in
retrospect you will realize that your time was not spent in the
most beneficial way: this is inevitable. The question is whether
your efforts were made in good faith, and a healthy dose of no
charge time (see below) is usually a fairly effective cure.
I wish to repeat that I am not expert in the finer details of
structuring a business deal or how to best negotiate with your
customer. There are so many ways that a deal can be structured, so
many different kinds of projects, and so many different personal
styles, that no simple summary could possibly capture even a fraction
of them.
But what they all have in common is that you are completely up front
with the customer, deal in good faith, and fully intend for this
customer to provide you with a great reference when the project is
over. Not surprisingly, these all contribute to The Warm Fuzzy
Feeling(tm) that is so important.
Touching on some more specific points:
Never, ever lie or fudge on an invoice
If you are ever caught -- or even suspected -- of funny business on
the financial front, you will not be trusted anywhere else. It is
impossible to give a customer The Warm Fuzzy Feeling(tm) if they are
wondering about the legitimacy of your invoices, and this is
fatal to a customer relationship and to ever getting a good
reference.
This is not to say that mistakes on an invoice won't happen, but
how you deal with them will tell a customer a lot about how you
do business. Your goal should be to overwhelm them with
integrity.
Anecdote:
Once, I accidentally double-included an 8-hour charge on a
customer project: hit the wrong key on my time-billing system and
didn't catch it. One of them was completely legitimate -- and the
customer knew it -- but the fact that two charges in a row had the
same time and detailed description were a dead giveaway to a
mistake.
My response was to write off both line items even though I was
entitled to one of them. By making the penalty for error high, I
made it clear to the customer that I don't play the "see what I
can get away with" game, only writing off the mistakes. Eight
hours was real money, but nothing compared to the ongoing
business I got from these folks.
It sometimes hurts to "eat" time, but the alternative (burning a
customer) can be much worse.
Consulting maxim:
Ongoing business is much more important than maximizing every
billable hour
Consulting maxim:
It's better to give away some time than to throw away your
reputation
Provide time-billing transparency
Anybody who's dealt with a lawyer has gotten a bill that said
Review documents ....... 40 hours
and for most of us, the lack of visibility into work done is
maddening.
Your customers should never think this about you: list your work
in detail to say what you did, and -- if necessary -- who asked you
to do it. Which of these two line item details would you rather
see from somebody you're paying real money to?
Consultant #1 Consultant #2
Email system administration ... Investigated email system
3.25 hours. problems per Martin; found that
MS Exchange required a patch to
deal with the latest Outlook
IMAP queries; re-optimized all
the message stores and tested
with several email clients. ...
3.25 hours.
Your detail should ring a strong bell with the customer, with him
nodding his head on every item while he reviews your invoice.
This gets you paid sooner and gives the customer The Warm Fuzzy
Feeling(tm).
Consulting maxim:
Detail is comforting to a customer
Give away some free time, but make it visible
Most honest consultants spend a certain amount of time for
customers that's off the clock, and this can be for several
reasons. It could be that it's for something you're not strong in
(and cannot justify charging full consultant rates), it could be
a pet interest of yours where you are willing to give away the
time, or it could be that your time on the task was so
unproductive that you don't feel right about charging for it.
These are all fine reasons.
But don't do it quietly. If the customer asked you about email
systems and you felt like you wanted to research this on your
own, it's important to list this on an invoice as nocharge time:
Researched email systems per Frank; investigated MDaemon, Exim
and Sendmail for the Windows platform; note to Frank summarizing
the findings ... 2.6 hours (n/c)
This shows the customer the level of effort you've exerted on his
behalf, and it puts you in the "trusted advisor who's in our
corner" category, and not a "consultant only thinking about
cha-ching".
Consulting maxim:
If the customer doesn't know you did work off the clock, you
don't get credit for it
It also makes it harder for a customer to object to the time you
did bill for, but it's important not to use the nocharge time as
a ploy: the goal is to maximize the customer goodwill, not to
maximize the revenue on any given invoice.
It's good for business to make your customers comfortable with
your work for them: it maximizes The Warm Fuzzy Feeling(tm).
Admit your mistakes
Every consultant pulls the boneheaded move now and then: deleted
the wrong file, forgot to do the backup, left the firewall off
for the weekend. It's contrary to human nature to expect that a
consultant will never make a mistake, but it's also not
reasonable to expect the customer to pay for them.
For a surprising number of your mistakes, your personal
culpability won't be known to the customer unless you tell them,
and in most cases you should do exactly that. Though you're
likely to get away with deflecting the blame as often as not, if
you get caught your reputation will be positively destroyed far
worse than the mere responsibility for the mistake.
Anecdote:
I performed the same software upgrade for two unrelated
customers, and some weeks later both had the same problem with a
little-used feature. This issue was getting increasingly
important, as it was affecting their customers, and after several
days of intensive research it turned out to be due to a small but
significant oversight I had made during those earlier upgrades.
When we finished, it would have been really easy to (a) blame it
on the software vendor, and (b) get paid for all my time. Neither
one would have been right, so I wrote a letter to both customers
explaining that the problem was fixed, that it was due to my
error, that they didn't have to pay for any of it, and extending
my apologies.
This was an expensive adventure for me, but both customers were
grateful for getting a straight story about something that was
impacting their business, and this kind of candid admission
contributes much more to The Warm Fuzzy Feeling(tm) than you'd
expect from "I made a mistake".
It's embarrassing to admit mistakes, but you owe this to your
customer. The very reason you're a consultant is that you're the
subject-matter expert and are supposed to know what you're doing.
Everybody makes mistakes now and then, and I suppose that even
some honest mistakes will mark the end of a customer
relationship. But it's never happened to me.
Consulting maxim:
If you routinely take ownership for your own mistakes, you're
much more likely to be believed when you claim something is NOT
your doing
Housekeeping and Paperwork
In the late 1980s I almost left consulting because the administration
and paperwork was such a burden, but my savior was Quicken,
Timeslips, and Turbo Tax. If you don't have a mechanism for dealing
with invoices, your checkbook, and your taxes, you'll surely go out
of your mind.
Time Billing
You need a way to capture your time, invoice customers, and
record their payments: this is hard to do well with anything home
grown. Many moons ago I had written a time-billing system using
the Informix database, and I hated invoicing every month. Then I
discovered TimeSlips and have been using it ever since. This is
very popular with lawyers, but it does a fine job for consultants
as well.
The disadvantage of TimeSlips is that it's "a different program"
from Quicken, and it means double entry of invoices and payments.
This is a serious drag, especially when Accounts Receivable
doesn't reconcile.
The Quickbooks accounting system has a time billing system that's
very substantially inferior to that of Timeslips, but it's all
integrated with the rest of the accounting system: I have
recently migrated to it. Those comfortable with real,
double-entry accounting may wish to give it a try, but it's not
for those who have never run their checkbook electronically.
Finances / Your Checkbook
For some time I operated with two checkbooks -- one business, one
personal -- but I gave this up more than ten years ago. Once I
started using Quicken, the need for that simply evaporated as I
was able to properly categorize and class all my transactions to
obviate the need for two accounts.
It seems natural to separate accounts on a "purpose" basis, but
in practice this is difficult. If you go to Home Depot to get
some stuff for your office and some stuff for your house, which
credit card do you use? Are you really going to write two checks
at the cashier's stand? With the line waiting behind you?
I've found it much easier to use Quicken's "class" mechanism to
make this work perfectly. Every transaction has a "category"
(office supplies, home repair, telephone charges, etc.), but this
is not sufficient to split your expenses. By creating a class to
each item, you can make that Home Depot receipt go where it
belongs:
Description Category/class
Home Depot -- light bulbs for kitchen Home Repair/personal
Home Depot -- light bulbs for the office Home Repair/business
One account, one receipt, two purposes. This works really well
and takes much of the burden of separate recordkeeping off your
shoulders.
However -- many accountants and CPAs will strongly advise against
this, especially if you're anything other than a sole proprietor
working out of your house. It's said that IRS dealings are easier
with separate accounts, but I believe that having meticulous
electronic records alleviates that substantially. I suppose I'll
find out during my first audit.
Taxes
Taxes are much more complicated for consultants, because Schedule
C (Income from Self Employment) is much more complicated than
Schedule A (Itemized Deductions). Every expense you have that
contributes to earning income is deductible -- software, auto use
when traveling to customers, a good portion of your internet use
-- but this requires substantial recordkeeping.
It's tempting to just get a good tax guy, but the taxes are not
the hard part: it's the recordkeeping that categorizes which of
your expenses are properly business expenses. It's not fair -- or
at least a bad idea -- to drop off a box of receipts to your tax
guy and have him try to read your mind. Good tax guys are not
cheap, and you want to pay him to prepare your taxes, not do your
bookkeeping.
If you're using Quicken, getting into good habits of category and
class will go a long way to taking care of this: having detailed
reports of your properly-organized expenses mean that you don't
need receipts for your tax guy, and he can file your return
properly.
For many years, I used Turbo Tax to compile and file my own
taxes, and it was helpful to really understand how it all works.
But I can hardly emphasize enough that you should get your own
tax guy. My tax guy -- who walks on water as far as I am concerned
-- saved me an enormous amount of in-my-pocket money the first
year I used him.
It's one thing to know the mechanics of taxes -- as I do -- but
it's another thing entirely to know what's a supportable
deduction and where the gray areas are. This is a minefield for
those who haven't gone down this road, and having a good tax
advisor will pay for himself in about the first fifteen minutes.
Thank you, Tom: you rock.
Filing your receipts
Most new consultants attempt to separate receipts as well as they
do accounts: separate folders for business receipts and personal
receipts, but this quickly falls apart unless you're unbelievably
rigorous in maintaining this separation. When you get your credit
card bill every month, which folder does it go in? What if it has
some personal and some business expenses? This gets messy in the
first month, and it's not necessary.
Years ago I adopted the practice of using 15" x 9" accordion
files with a separate tab for every letter of the alphabet. I use
a new file every year, and file all receipts -- for whatever
purpose -- alphabetically by vendor. This is optimized for the
only kind of search you'll ever do: fetch a specific receipt from
a specific vendor.
There are other financial queries you'll do (how much did I spend
in March?), but you'll never need to gather all receipts from
March. Having your receipts ordered by vendor name is perfect.
Promotion and Advertising
[Self promotion] I've never done any explicit advertising or
promotion: no Yellow Page ads, business card on bulletin boards, etc.
For years I was fortunate enough that word of mouth was sufficient
for me without having to put on my sales hat, but the internet has
changed this substantially. Some may consider banner/Google ads, but
I believe this is likely to be expensive and of limited effectiveness
(though I have not tried either one).
It turns out there is a very effective method of promotion that does
not involve spending any money or putting on a sales hat:
Consulting maxim:
Your best advertisement is publishing original, technical content
In my case, I write Tech Tips on my website - this paper is one of
them -- and publish them on my website. I usually do one or two per
month, but it's not on any kind of schedule. Once I've solved a
problem that I think others might have trouble with, I'll write it
up. Or, if I believe that I can explain a topic better than other
resources (say, An Illustrated Guide to Cryptographic Hashes), I'll
produce one as well.
In some cases I'll proactively promote them, such as submitting
Malware Analysis to a narrowly-targeted venue such as Bugtraq, and I
certainly announce them in my weblog, but by and large I allow Google
to do all of my promotion for me simply by indexing them for people
to find.
These do take a long time to write (and it's non-billable time), and
certainly those who are not strong writers will have a harder time of
it, but I believe these to be very, very effective methods of
promotion. Most people who benefit from your Tips won't be candidates
for customer status, but it does help establish a reputation for
competence in the area you're writing about. It's also nice to get a
thank-you from random people on the internet who are grateful for
your efforts.
But when prospective customers are investigating a potential
consultant, a website that is more technical than marketing may very
well be seen as a positive: with clear, strictly-technical content,
the customer can find out for himself that you're qualified rather
than rely on you to simply claim that you are.
Consulting maxim:
It's a huge asset to communicate well -- cultivate this skill
vigorously
Writing for a magazine is also a great way to raise your status in
the community: most magazines are always looking for good writers,
and getting published is typically a relatively informal process.
Find an area you're competent in, make sure you understand the
publication's target audience, and email the publisher with a
proposal.
You will be expected to produce an article of so many words, and this
number is chosen by the publisher based on how much space he has in
the issue in question. If he asks for 3000 words, don't submit 2500
or 3500. And you must be on time: there is no surer way to burn a
relationship with a publisher than to miss your deadlines.
This promotional approach does take time -- a reputation is lost
quickly but gained slowly -- but I believe that these efforts gather a
kind of momentum: as your track record of writing and publication
grows, you'll have an easier time being accepted into the next one.
All of this contributes to a larger and larger body of work that
precedes you to potential customers.
References
[References icon] If you conduct yourself properly, you will
accumulate a list of customers who will speak well about you: in many
cases these will be your best source of new business, and in any case
they have a name: your references. No matter how renowned you are for
your technical skills, it is hard to overemphasize just how important
your references are to a successful consulting practice.
Think what you would ask if you were checking up on somebody you were
about to hire: you'll ask about his skills, of course, but that's not
all:
* How was he to work with?
* How much supervision did he require?
* Did he show up when he said he would?
* Did he deliver what he said he would?
* How were his invoices?
* Would you recommend him to others?
Though some of these questions touch on the technical, most are about
the relationship. Many of us know engineers who are technically very
strong but nevertheless hard to work with, and some are so difficult
that the customer will work with a lesser consultant rather than deal
with that primadonna.
Consulting maxim:
Your references are your reputation in the consulting world
Obviously, one can get a friend to pretend to be a reference, and I'm
sure this happens (much like lying on a resume), but I believe this
to be a terrible way to start a new relationship.
Likewise, it's obvious that you get to pick from your customer base,
and you'll not choose those that are less than happy with you. But
your prospective customers know how this works, and they can read a
lot into what your references say (and they can smell a lukewarm
reference a mile away). If you're doing a good job taking care of
your customers, they are usually more than happy to say good things
about you to others.
This doesn't always materialize, of course: from time to time there
will be that relationship that simply doesn't work. A personality
conflict, internal politics that you get caught up in, or a non-work
life crisis can all conspire to create an engagement that you don't
care to talk about.
Almost all long-term consultants have customers who will report a bad
experience -- I have a few -- and sometimes these will be your fault.
Only if the problem is due to overt dishonesty or incompetence should
you worry about word getting around.
On Being Objective
[Objectivity icon] Some consultants choose to include product sales
in their practices, and though this is a legitimate aspect of
business, the question of objectivity always arises with the
customer. They will (rightfully) wonder whether you're recommending
the product because it's the best or because they have higher
margins, and there is not really any way to completely remove this if
you have a vested interest in the transaction.
Make no mistake: it's possible to be objective, and many consultants
do choose their product lines strictly on what's the best value. What
remains is the perception in the customer's eyes, and all you can do
is conduct yourself honorable and transparently so there is no
question about your vested interest.
Consulting maxim:
Your customers cannot wonder where your interests are
This means transparency.
I have long believed that computer-products sales do not have enough
profit to be worth the trouble of dealing with sales tax (and sales
tax authorities) or warranty returns. Maybe you'll make $100 on the
sale of that printer to a customer, but if it's dead on arrival, you
have to eat the time required to box it up and send it back.
But this is a business decision, not an integrity issue.
Personally, I made it a practice of selling nothing but my time, and
to never accept referral/finder's fees, kickbacks, or commissions
from vendors (though there is nothing wrong with a vendor taking you
to dinner now and then). Not only does this avoid the warranty/sales
tax issues, but the customer won't ever wonder if I have a vested
interest in the transaction or if that interest is influencing my
recommendations.
From time to time, a customer will require something -- software, a
cable, etc. -- and I'll purchase it on their behalf, but this is
strictly as a convenience to the customer. I include the item on my
invoice at exactly what I paid for it, and I include the receipt.
Those who use this pass-through arrangement may be wise to include
the phrase "Total includes sales tax collected by merchant" with the
item to make it clear that you're not a reseller and are not
responsible for collecting sales tax on your own account.
Be Easy to Fire
There are several reasons why a customer might find it difficult to
fire a consultant:
* He does great work and is a tremendous asset
* He has passwords for everything: will he give us the list?
* He has all the source code: will we get it back?
* He has access to everything: will he do bad stuff to us?
* He hosts our DNS: will he screw with us?
Only the first is a valid reason to keep somebody around, and
customers are unnerved by the subtle fear of "what happens if...?"
for the others. Consultants ought to structure their arrangements to
remove as many of these as possible, and instead rely on their own
good work to be their best job security.
From time to time, I give customers "How to fire me" instructions:
"Here are all your important passwords", "Your contact at the hosting
center is who", "Source code is here", "Revoke my network access by
(A), (B) and (C)", etc.
Customers should ask for this from all of their consultants, not only
to make them easier to fire, but as a contingency in case the
consultant is hit by a bus. Sometimes customers do ask, and it must
be met with immediate, full disclosure. It's not your data, you're
not allowed to hoard it.
Consulting maxim:
Customers are comforted by consultants who don't act entitled to
their engagements
If a customer wants to fire you, hoarding this data will buy you a
little time, but this breeds resentment in the customer that
positively does you no good.
And sometimes you will be fired -- it could be that the project has
been cut, they cannot afford you any longer, or they may simply not
like you very much. These things happen, but what marks you as a
consultant of integrity is how you behave on the way out.
Find everything -- physical and otherwise -- that conceivably belongs
to your soon-to-be-former customer and return it with a cover letter
stating that you have done so (this partly for your own protection).
Even if they have covered their bases well, it's common for them to
have missed something (e.g. "I'm still the registered contact for
your DNS management, I recommend that you transfer this your own
staff. Password is hello").
You may never hear from them again, but you'll know that you did the
right thing.
Firing your customers
[Firing customers icon] Many years ago, I read a column in Electrical
Engineering Times about small-company-makes-good, and the
now-successful owner was asked "How did you know you'd made it?". The
answer has always stuck with me:
"When I didn't have to do business with people I didn't like"
I am fortunate that the overwhelming majority of my engagements have
been pleasant and productive, and I've made some very good friends
out of them, but not every one is like this. For one reason or
another, the relationship can sour or a customer may start taking
advantage of you, at which point you can consider just walking away.
I once had a customer that I generally liked a lot, but for whatever
reason, they positively could not pay me on time (and they even had
net 30 terms at the time). I believe they were in an industry where
taking your time was just part of how they were allowed to operate,
and in spite of talking to my customer about it every single month,
nothing changed.
I believed that I was performing a valuable service to this customer,
but apparently not valuable enough to be paid on time, so I completed
all outstanding work and invited him to find another consultant
(i.e., I fired him). Not surprisingly, he's not among my references.
Consulting maxim:
The customer is NOT always right
Not all consultants in the same position would have made the same
choice, but everyone is allowed to choose whom to do business with.
But even when you fire a customer, make sure you do so with
integrity: the customer may not speak well of you, but you should
give him no grounds to claim that you acted dishonorably.
Google is not always your friend
These days, many customers evaluating a consultant will do the same
thing you do before going on a first date: Google them. This is where
the other side of a net presence comes into play: though your website
may be professional and full of outstanding technical content, this
can all be undone by your own words found elsewhere.
If a search finds a potty-mouthed political flame, a public trashing
of your ex-wife, or open advocacy of illegal activity, this can go a
long way to telling a prospective customer what kind of person you
really are. It might not impact their view of your technical skills,
but it may give them a good reason to look elsewhere for a
consultant.
Consulting maxim:
The internet never forgets: don't provide dirt for your future
A good rule of thumb: if a search engine can index the venue, don't
say anything you would not want your customers or your family to
read. This certainly leaves you open to private venues, such as a
members-only forum, where you can tell that dirty joke or vent about
the ex, but even here it's wise to be concerned about inadvertent
leakage.
A colleague has noted a countermeasure for this: have a ridiculously
common name.
How much do I charge?
[Consulting Advice 5 cents icon]
This is the single most common question that prospective consultants
ask, and one of the hardest to answer. The short answer is a mix of
"whatever the market will bear" and "how busy you wish to be". Some
consultants operate on a strictly hourly basis, others on a retainer,
and yet others on fixed-bid contracts -- none is inherently superior
to any other (though some are more suited for some kind of
arrangements than others).
I operate almost entirely on a by-the-hour basis, with only
occasional forays into retainer or fixed-bid arrangements, and my
rates were initially set when I started out in 1985. Those rates have
risen steadily over the years to reflect my growing experience,
though the end of the Internet boom did put a damper on them.
New consultants typically ought to charge a bit less than the going
rate for other consultants in the same area, which allows for their
lesser experience and wanting to get a foot in the door. I'll note
that "lesser experience" refers as much to "conducting a consulting
practice" as it is to "knowledge of technical issues". Many very good
engineers leave big companies with outstanding subject-matter
expertise but nevertheless have only corporate experience: much of
this does not directly translate into consulting.
Raising rates is usually difficult for new consultants: you've built
a good personal relationship with your customers, and now you're
asking for more money. Raises should be modest, and with plenty of
notice (at least one full billing cycle, perhaps even two).
Raising rates is usually done when demand for your services is high,
and there's a good chance your customers know when this is: you're
juggling more work and they see scheduling farther and farther out.
Some customers will use less of your time at the new rate, but in
some sense this is intended.
Consulting maxim:
If you're booked up solid, your rates are too low
One must be careful that one doesn't take this too literally: the
concept "booked up solid" is over a fairly long time period -- many
months at least, perhaps even years -- and it doesn't mean that rates
should be jacked up just because you're having a busy week. I can't
remember ever raising rates more often than once a year, and even
then not for all customers at once.
It's also common to have different rates for different customers.
Long-term customers, upon whom your business is likely based, should
probably pay a bit less (or at least reduce or defer the increases):
they are your bread and butter, and consistent ongoing work is much
more important than maximizing your hourly rate.
Consulting maxim:
Your long-term customers are your best customers
Furthermore, different kinds of work deserve different rates. For
some of my customers, I am essentially the on-call MIS staff, so I am
at the mercy of day-to-day whims. This makes it hard to schedule the
work, so it is normally full fare. But for other customers I do
project-based software development, where I might be writing code for
several months.
This is work that I can do when I'm not solving
problem-of-the-moment, and in that respect is better work because I
get to pick my own schedule. Long-term, at-my-own-schedule work
warrants a lower rate to encourage this kind of business. Background
work makes your time more liquid, and consultants (like me) who
prefer to mainly work at home love having a backlog of things I can
work on in my slack time.
A similar principle applies to offering lower rates for work you
simply enjoy more than other kinds of work.
I don't typically offer multiple rates to the same customer, though
this is certainly possible. I could imagine a consultant offering web
design for one rate and network diagnostics at another, but this gets
difficult to manage, especially when there is some legitimate
question over which category the particular work should be in.
When offering a discounted rate to a customer, consider doing so not
by lowering the rate itself, but by offering/increasing the
timely-payment discount. If your going rate is $100/hour and you wish
to offer a customer a $90/hour rate, make it a 10% discount if their
invoice is paid within account terms.
This way you have given more teeth to your due-by date, and a
customer that decides to use his consultant for float will have a
hard time defending taking the discount after paying late.
Happily, though: consultants are generally earning the most when they
are thinking about money the least: By focussing on your customers
and your skills, you make yourself the most valuable.
Consulting maxim:
The best way to make a lot of money is to make your customers a lot
of money
Time and Project Management
This is a very hard problem for a true consultant, but less so for a
contractor. Your schedule is largely at the whim of your customers,
and many consultants find that it's a feast-or-famine existence. You
may have a nice comfortable level of work, but then two customers
come with large projects: how can you do it all?
Unlike corporate employment, where your boss gets to make these
priority decisions, one customer rarely cares that some other
customer also needs your attention. In many cases you can simply
schedule the work ("I can do that for you by next Tuesday"), but
consultants involved in day-to-day work cannot really expect a
customer whose whole business is down due to a crash to just wait a
week for you to show up.
Though your customer may very well understand that you're busy, are
dealing with some other customer disaster, or you have to stay home
with a sick child, one cannot escape the reality that a consultant
who is not available has less value than one who is. I've heard many
customers speak well of previous consultants:
Said by customers:
"He was a great guy, but he was never available"
To some extent this can be partly addressed by higher rates -- which
reduces your demand -- but this only works over the long term. In the
short term, you do not want your good, long-term customers going
elsewhere because you're not around, so the only solution may be to
just work harder.
Consulting maxim:
As long as you're sleeping, you still have inventory
You have to decide how much of your life you're willing to be
consumed by your consulting practice, and you may decide that the
always-available treadmill is just too stressful -- this is a
completely legitimate decision. But you're going to be less valuable
to customers, and therefore justify lower rates.
Many consultants have a common fear -- will the work dry up? -- and
this can be quite pervasive. You have no real job security and are at
the mercy of customer projects, budgets, and personalities, and in
many cases you don't know that you will have work two months from
now. This creates an incentive to take more work than you can handle
on the assumption that it might not be there in the future.
I was booked up solid for 17 years (yet always with the "what if...?"
fear) before the bottom fell out in late 2001, and an empty pipeline
is a very sobering experience.
Your pipeline is the amount of committed work you have in front of
you, and longer pipelines are much more comforting. You can't do much
to avoid or predict the routine day-to-day interruptions, but you can
schedule the larger projects. Taking on a new project that you can
start in two weeks means that you now have a certain amount of work
you can count on.
Consulting maxim:
The fear of an empty pipeline is with most consultants constantly,
even if they're consistently very busy
Unfortunately, consultants worried about their pipeline have a
natural tendency to squeeze in new projects. Though the new project
is done at the moment, the time taken adds to the total pipeline
because the other projects remain. This is not terribly fair to the
existing customers who are getting pushed back (probably without
notice), but it's hard to overemphasize how powerful the
empty-pipeline fear is.
Even good consultants do this now and then, though they typically
manage it by cutting into non-work time rather than outright
shortchanging existing projects.
A final note on time management: with ongoing customers, there will
often be low-priority background projects that can be handled more or
less at your convenience -- e.g., anytime in the next few months -- and
these are great to have in your back pocket during a slack time.
But many of these long-term customers have a kind of unspoken limit
on how much you can bill them in any given month, so there is only so
much of that back pocket time you can expend all at once. Of course,
if there is a disaster and you have to dive in, it takes what it
takes, but most customers will call a time out when the invoice level
reaches a certain amount.
You're likely to run into a circumstance where you have things you
could do for this customer, have nothing on your plate for anybody
else, but have already reached that unspoken limit: if you go over,
you'll probably get paid, but it means you're not going to be trusted
to keep it reasonable in the future and will be subject to heightened
scrutiny.
Locating this unspoken point is difficult, but it's important not to
see it as gaming the customer. This all presumes that the customer
legitimately needs the work done: otherwise it's just churning, which
is dishonest and besmirches the whole consulting profession (and this
is a close cousin to "value billing" used by accountants).
Instead, you're providing the customer with a service by properly
prioritizing the tasks in front of you, and in some cases you are
more qualified to do this than the customer is: "Does that server
need to be replaced right away, or can we get six more months out of
it?" or "Is our firewall secure enough, or must we upgrade the
software now?".
Consulting maxim:
You must know how to read your customer
The Projects you don't take
[No Thanks]
It should come as no surprise that not every consultant is qualified
to take every project, what's not obvious is that a consultant should
not take every project for which he is qualified. There are
circumstances when telling the customer 'no' is giving them good
service even if you don't get paid for it.
Anecdote:
A customer wanted me to "just start programming" on a large software
system, but over their objections I insisted on creating a formal
specification first.
After creating a 100-page document (which in retrospect was
nevertheless wholly inadequate), Microsoft Project told us that it
was going to take far more time and money than the customer had
available.
The project was shelved.
It would have been easier (and certainly much more fun) to "just
start programming", and I took serious pushback from the customer
during the process, but it's absolutely clear that this was the right
thing to do.
Consulting maxim:
Your customers are buying your judgment, not just your time
To be fair, I did get paid for writing the specification, but it was
far, far less than what the programming would have been.
Anecdote:
I was brought in by a new customer to evaluate another consultant's
work. I found him to be technically competent, but the customer felt
he was too secretive and generally difficult to work with.
So they could get rid of him, they asked me for a proposal to provide
around 20 hours/week of consulting, and even at a discounted rate it
would have been good work: it was a very slow time for me.
But during my evaluation I met a low-level guy on the help desk, and
it was apparent fairly quickly that he was the perfect guy for the
job. He was young, but he had good training, good experience, and had
simply been smothered by the other consultant and not allowed to grow
in his position.
During the meeting with the customer, I told them they did not need
me: they would get much better service for much less money by using
their own staff. They were flabbergasted to hear this, but it turned
out to be exactly the right advice for them.
I understand they have been thrilled with Joe.
I lost the project, but gained a reference and helped a very nice
young man take a big step early in his career. And you never know
when one of those managers may go elsewhere and need the services of
a consultant who's demonstrated that he looks out for his customers.
Consulting maxim:
Being known for your integrity is the Holy Grail of consulting
Consultants should always act in the best interests of their
customers even if it's not so good for you. It may be expensive, but
this is how a reputation for honesty, integrity, and professionalism
is built.
Intellectual Property
Unless you are providing nothing but services, your engagements are
likely to involve some intellectual property (software,
documentations, designs, patents), and the open question is "Who owns
it?"
My policy has always been "The customer owns everything".
It's possible to structure an arrangement where you provide software
in binary form, with source code being extra or put in escrow, but I
believe these to be counterproductive to a good customer
relationship. If you're seen as holding back or trying to create a
device for extra billing, the customer often feels like you're trying
to manufacture job security: this never engenders good feelings.
Consulting maxim:
An open customer relationship cultivates The Warm Fuzzy Feeling(tm)
By telling the customer "Everything is yours", you are making it
clear that you are relying on nothing but your performance for
ongoing work, and there are no questions about what's in and what's
out.
One question that arises often: "can I reuse software from one
customer to another?", and the answer is "sometimes". Generally, you
cannot reuse a whole project because it represents customer-specific
functionality. This is their property and you can't repackage it
without their express permission.
But once it gets down to the building-block level, it's perfectly
reasonable to reuse library code that performs generic,
non-proprietary functions (say, a networking library or a database
abstraction layer). I have a substantial body of C/C++ code that I
draw upon for every project, and as long as nothing is tied to the
original customer (either by comments in the code or proprietary
functionality), this allows you to do your project in less time.
Some consultants actually charge for their building-block libraries
(so called "toolkits") and though I suppose this can be reasonable if
it's a really substantial infrastructure, I have never done so. I
decided that part of my value is everything I bring to the table, and
it's what allows me to do a project faster and better, justifying my
rates.
The other reason is that customers are usually suspicious of these
toolkits, viewing them as an excuse to charge more money. This is
less of a concern if the toolkit is available on the open market and
not strictly as a pitch to consulting customers, but my libraries do
not warrant that kind of packaging.
Many of your projects will be one-time custom jobs, but sometimes
you'll be asked to build an actual product that your customer will
sell (more likely on a contract basis). In these circumstances, other
arrangements are possible, such as offering a discounted rate in
return for a royalty on each unit sold.
This reduces the hit that the customer takes up front and gives you a
vested interest in the long-term success of the product: it's better
for everybody when your interests are largely aligned with that of
your customer.
But it does require more recordkeeping, and it create incentives for
the customer to underreport sales. These arrangements require much
more negotiation skill on your part, especially since the customer
usually knows his own market space better than you know his space. I
have never undertaken a royalty engagement.
Another issue that comes up from time to time is patent rights, and
this often generates very strong feelings. I have always believed
that software patents are terrible for this industry and have opposed
them vigorously. My previous position had been that I would not
participate in software patents in any way, but I have reversed
myself.
I have several customers who patent everything they can find, but
strictly as defensive measure. Often, the only defense to a claim of
patent infringement from another party is a patent portfolio of their
own: they will trade and call a truce.
My customers positively hate this -- it's expensive and an enormous
distraction -- but they have to play in the same game as their
competitors. Refusing on principle to patent anything means the
customer may be defenseless when the Cease and Desist letter arrives.
I have not yet been involved in any patent proceedings, but the
general sentiment of the "Customer owns everything" consultant is
that the consultant assigns the patent rights to the customer while
perhaps retaining the role of Inventor.
Making the Leap
[Making the Leap]
I have been a consultant for essentially my entire adult life -- I've
never held a "real" job -- and the question of making the transition
from corporate life to consulting is something I simply have no real
experience with. But I can touch on a few issues that I imagine are
probably important.
It's important to handle the obvious financial considerations first:
have a nest egg in the bank, arrange for health insurance, create an
office to work in. This all takes more money than you think, and
cashflow issues are always more difficult than you expect.
Many consultants get their start not by diving in, but by dipping a
toe in the water. A bit of work on the side is a great way to get a
feel for the business and see if it's really right for you.
Though the extra income will be nice (and is often the main draw),
the primary benefits of these early contracts are usually not
financial. Do you have the ability to manage your time without
supervision? Are you able to treat customers like customers? Are you
able to think well on your feet without a boss to fall back on? Can
you earn a reference?
Most beginning consultants don't realize how important these
non-technical factors are, and some have taken the full plunge only
to find out that they were simply not cut out for consulting for
whatever reason. Failure is usually a very painful experience, and
starting slow may help avoid this.
Your early contracts will almost always be at a less-than-market
rate: if you're not generally available during working hours, and if
you take a day or two to respond to queries, this makes you less
valuable to the customer. But it's extremely valuable for you to find
out how you work in this arena. This is an investment.
But I'd like to wave you off on a common trouble area: customer
theft. When your corporate job includes servicing of customers, it
can be tempting to go solo and take those customers with you, but
this is an enormous mistake if it's not done with the approval of the
company. In addition to potentially running afoul of
non-compete-agreements with your employer, it sends the message that
you cannot be entrusted with a customer relationship.
Consulting maxim:
If you have a reputation for stealing customers, you'll never be
trusted by other professionals
But once you've gotten a few projects under your belt, you'll
hopefully have a taste for how the business operates, have some great
references, and have enough work lined up to create a comfortable
pipeline. If so: make the leap!
Technical Skills and Certification
I purposely put the technical part of this Tech Tip last, to
reinforce the notion that "customer service", not "computer science"
skills are the biggest factors in a successful consulting practice.
But it's foolish to think that technical skills don't matter: you
don't have a business unless you can offer a service that a customer
is willing to buy.
I have a Bachelor's Degree in Computer Science, but this only rarely
comes up and I have never been asked for my GPA (it wasn't anything
spectacular): this is my only certification. Though for employment,
certifications (degree, MCSE, CISSP, etc.) tend to matter a lot, they
are much less important for consultants.
Consulting maxim:
Your references and your experience are far more important than your
certifications
What counts here is truly learning the subject matter, and there is
no harm in obtaining the certificate in the process. But if the goal
is just to collect some paper, it leads to the prototypical computer
jockey with lots of alphabets after his name but limited power in the
driver's seat.
Where the skills question gets tricky is when getting outside your
comfort zone: a customer will ask you about a project that you are
almost, but not quite, qualified for. Surprisingly, this happens a
lot: if you have conducted yourself well, your customer would rather
find a way to use you -- a known quantity -- than find somebody else.
This occurs over a fairly wide range of skills.
When considering one of these projects, the first rule is: never lie
to your customer about your skills. Be completely candid with your
customer about what you know and how you would address the project.
This would likely include substantial off-the-clock time as you got
up to speed on the technology in question.
Consulting maxim:
Do not BS your customers
But it's important to weigh not only how you think you could handle
the job in question, but how much you could leverage this for some
next project, perhaps with a different customer.
For instance, if you've been a Visual Basic programmer for a long
time and are invited to consider an ASP.NET project on the web, this
may be a stretch for you technically. If this is an area that you see
yourself wanting to branch into, tell the customer you will get up to
speed on your own time and may even offer a discount to make your
services more attractive.
What you get at the end of the road is not only a new skill, but a
new reference. The time you spent on this project makes you more
valuable -- more full service -- to this and other customers in the
future.
Consulting maxim:
"Education" is one of the best investments a consultant can make
The term "education" is used broadly here: formal training,
heat-of-battle experience, previous consulting projects, and even
personal projects at home that interest you.
---------------------------------------------------------------------
This Tech Tip has been many months in the making, and it's important
to emphasize again that it's based strictly on my own experience and
observations, and I have no special business-school training to back
it up. Most of this was learned on the job, and I can only assume
that my still being in business after 20 some years means that it's
not entirely without merit.
But there are a lot of ways to conduct a consulting practice, some of
which bear no resemblance to how I have conducted mine: you are
strongly urged to seek real counsel before setting out on your own.
Consulting maxim:
Don't quit your day job solely based on what you read here :-)
The very best of luck to you.
Note: The (tm) in The Warm Fuzzy Feeling(tm) is a presentation device, not
a real trademark.
For another look at the life of consulting, please consider Spare
Room Tycoon Succeeding Independently (subtitled The 70 Lessons of
Sane Self-Employment), by Dr. James Chan, which is less about how-to
and more about what you're going to go through. It's an easy read,
and it gives a great insight to the experiences of self employment in
a wide variety of industries. This may give you clues as to whether
you have the personality for going it alone or not.
Special thanks to Chris Mospaw for his graphic-design help.
More Tech Tips
Home Stephen J. Friedl Software Consultant Orange County, CA
USA [Steve's Email] [RSS Feed available]