Advice to “never bill by the hour”

Is this true? I usually bill by the hour, but I have to say that there’s always some awkwardness about this aspect of consulting. Compared the typical hourly rates charged by statistical consultants, my impression is that I charge more but that I bill for far fewer hours–partly because I do consulting as an extra, not as my main job, so I’m typically trying to keep the hours limited.

Maybe a fixed charge would be better, but the trouble is that it’s not always clear to me what exactly is needed. Or maybe there should be two stages: first a fixed charge where the product is an assessment of the problem, then another fixed charge for the main projects. Or maybe charge an hourly rate for the little problems and a fixed rate for the big ones. It’s something to think about. It would be great to get enough money from consulting to really support some of my research efforts.

18 thoughts on “Advice to “never bill by the hour”

  1. I spent a year doing various machine learning consulting jobs, and (though I'm no expert) I had success with the following strategy:

    My policy is to charge by the hour until a clearly defined project takes shape. At that point, I prefer to work on a per project basis, doing whatever it takes to produce the deliverable. I can quote an approximate number of hours, then give a bit of a discount for doing the work in bulk. If the quote turns out to be significantly over or significantly under, then we can meet and discuss why things are easier or harder than expected, then revise the quote. Otherwise, we don't worry about plus or minus a couple hours (which can be somewhat arbitrary anyway, depending on what your definition of billable hours is).

    This probably only makes sense for larger jobs, though. If you're just doing a few hours of work here and there, the hourly model works fine in my experience.

    I think his points about setting it up so you only have to deal with red tape once or twice are quite good, though. From what I can tell, most of his points are just about making sure that it's a team effort rather than a you vs. the client situation.

  2. I think Danny's model works well, but once you have a scoped project you might offer a client the option of per-hour or a value-based formula. Then you can quote a high hourly rate and still forthrightly say: "I'll make more money on the value-based formula, but if you really want to I'll let you pay by the hour instead."

  3. When you charge by the hour, you'll get less of those "Oh, one more thing . . ." emails.

    When you charge by the job, the client usually asks for of your time than you might have expected.

    If you do decide to charge by the job, make sure to factor in the time creep by calculating how long you expect the job to take and then tacking on an additional 30-50% for those "unforeseen issues".

  4. Breaking the consulting contract into problem definition and problem solving certainly makes sense. The only challenge is that clients are unlikely to pay for "merely" defining the problem.

    Overall, for consulting contracts that are of the problem solving nature, I am more of a fan of the Percent of Value created argument. This approach, IMO, puts a natural premium on the solving of problems where there is value (net positive cash flows) to be created. (Instead of trying to solve problems just because someone thinks it is a good idea.) Also, this structure works well with the amorphous nature of the effort to solve problems. There is enough flexibility to try different approaches but at the same time, pressure to make sure that you get to a workable solution pretty quickly. Also, the client is happy that he is not paying just to watch the clock go around.

    However, for maintenance-type consulting projects, it makes more sense to have a per hour contract. Here, the onus is more on execution than a creative problem articulation and solution set-up.

  5. I can understand why clients might want that. It appears to reduce their risk. However, value-based pricing basically means that the consultant is trying to make sure the client never gets a good deal. If the client's problem is big for them but easy for the consultant, then they may pay an awful lot per hour.

    It also creates some difficult incentives. If the consultant is billing a fixed amount no matter how much they work, then it's in their interest to get done as quickly as possible, just as long as the client never notices the corners being cut. Since part of the reason you hire a consultant is that they know something better than you, that's a little odd. It also encourages the client to pile things on, stretching scope as much as possible. Preventing that means a lot more time spent on specifying scope up front to prevent surprises.

    Still, I think it's good for consultants, or anybody in business, to focus on customer value, so I really like that part of it.

  6. I used to work for a CRM consulting group (for the marketing analytics side) and we didn't bill out our projects by the hour. Our theory was that if you can't set a price on it, you don't know the domain well enough or you don't know your client's problem well enough. We kept certain things as time and materials (especially anything IT related on their systems, since that is fraught with peril) and had an hour-based service contract after the project was completed following our warranty period, but for the base configuration and consulting we would charge a flat fee. Travel expenses were, of course, billed as materials but we didn't charge for time of transit.

    One key advantage that probably doesn't apply in your case was that clients tended to care less about where we worked. Since they didn't feel like every hour was costing them money, we could pretty much work from anywhere and spend as much or as little time as we neede so long as we got the job done.

  7. Most people charge at least double a comparable hourly industrial rate for consulting. Someone who knows stats, can write code, and implement end-to-end customer solutions should make around $150K/year in industry, so we'd pay them $150/hour. But there are a huge number of variables, such as length-of-contract, stability of company, cost of negotiation, how much you like the job, etc.

    If you look at the productivity of someone like Andrew, you should be able to justify paying him multiples of what you'd pay an "average" statistician. Unfortunately, hourly rates don't typically go up to $500/hour for stats consultants.

    Almost every time we charge time and materials for consulting, the customer winds up wanting us to do way more than originally planned. I think this arises in part because corporate budgets are so gerrymandered — they often have slush funds for consultants but not permanent hires kicking around in their budgets.

    Expert witnessing is a great business if you can get into it and don't mind working with lawyers — going rates I've seen are around $300/hour, and often the lawyers want lots of time to do things like prior art reviews for patents.

    I have a friend in Pittsburgh who had a two-week consulting gig to sort out some problems in financial reporting software. He fixed the software in about 10 minutes (they were using their own buggy sort routine rather than the language's libraries). I don't know how much he wound up charging them.

  8. Both per job and by the hour have distinct disadvantages.

    Another possibility is to charge per job up to some given number of hours (with an implied per hour rate below the normal per hour rate) and then per hourly after that.

  9. Andrew,

    You should let me know if you link to my blog []. In short, billing hourly is inherently unethical, and rewards inefficiency. The client must make an investment decision every time they want to engage you. Also, there is a HUGE distinction between "fixed fee" and Value-Based fees. I'd be happy to extend this conversation to a different thread if you like.

    Best wishes for a happy new year!

    Mark Richman

  10. Hi,

    my wife runs a mediation business (180 chf/hr is the standard rate). She softens this with a free set-up session and then the real thing.

    We came up with the idea to have a declining rate – so the longer she took to fix the 'problem' the less she would get per hour. This is a transparent way of saying – I will not extend your problem – always an issue with conflicts or psychiatry ;-)

    However the customers so far are happy with the fixed rate.

    Andrew has to charge more than $150 otherwise there is no space for us mortals, not to mention India! In the Pharma industry 250 chf/hr for SAS programming is not unknown. (Although these days more like chf 90 is more usual).


  11. Thanks for the hourly rate information, Bob. I suppose in the long run it's cheaper for companies to consult as opposed to hiring a permanent statistician?

    DaveG, I like the declining compensation rate idea. I guess the key question is how to formulate the "decay rate" for your wife's declining hourly rate. If you use an inverse squared relationship (=1/x^2, where x=hours worked on the job beyond some time limit), then total compensation for the job will approach a fixed limit as x -> infinity. If you use an inverse relationship (=1/x), then total compensation for the job will approach infinity(!) as x -> infinity, even as your wife's hourly rate declines. :)

  12. I can only speak as a lawyer and as a client. Lawyers wish they can charge for the job because that usually means a percentage fee. This is normally done in estates – big moneymakers for little work – and by firms that are close to the flow of deal funds (only a handful, mostly in NYC). We charged a flat fee of cents psf for leases for mall developers but that only worked because we'd lose money on small leases and make it hand over fist when we'd do 29 renewals for a chain all at the same time.

    As a client, the reality is that you should know what your charges should be and we always watch them and discuss them as we go. If a project looks weird, we may negotiate a deal with a regular law firm but then only because we have a course of dealing and we both understand that no one is supposed to lose here.

  13. Wouldn't an agent make the most sense?
    I sense time spent thinking about this is time wasted for you.

    I did a quick google, I'm surprised no Jerry McGuire JD, MBA type turned up, but an operation like this should be able to handle the client flow and billing for you:

    Just don't let them take too high a percentage of your fees (I'm guessing it should be somewhere between 5% and 33% depending on how much income you'd be generating).

  14. Regarding Hopefully Anonymous's comment about agents, I've never found a true agency that makes sense for most independent consultants. They seem to be ok at matching relatively long contracts (3-12 months) if the position is well understood, a known sort of contract labor. But for actual consulting, or even short contracts, the ability to identify and negotiate the work seems too tightly bound up with the skill that is being sold.

    I have had luck outsourcing my back-office work (contracts, invoicing, taxes, benefits) to a third party. I use (and like) MBO Partners, and they charge 5% of gross. Since I hate paperwork, it's well worth it to me.

    Another option is to package your material as classes. People really like the clear, discrete nature of them.

  15. I always bill by the hour.

    My clients' needs have never been well-defined enough to allow me to bill by the job.

    Sometimes things are much simpler than I first think they will be; more often, they are more complex.

  16. I always bill by the hour as well. When you don't, the client owns you until the job is done (and "done" by their definition, which might change). For budgeting purposes I see no reason not to help the client out with estimates, but the actually billing is done per-hour just as my work is.

    I help soften to blow a bit by pointing out that I accomplish more per-hour than anyone else in my field, so the end cost for the project (which is really what should matter, yes?) is often less than they would expect. My experience, in other words, is the opposite of Mark's point #7.

    One of Mark's other points is that billing by the hour (or, put another way, billing whenever a client freshly requests my time) will act as a barrier to work. But what it actually does is prevent me from giving out free work if I don't want to. New work comes as a result of a history of good work, not one's billing scheme.

    Billing by the hour eliminates project creep as well as the necessity of determining when a project goes beyond the scope of the original work order.

Comments are closed.