Estimating Jobs

[FileSquire]

Hello,

I've been a lurker for a while to this forum. I'd like to generate some response about good practices for quoting work. I currently am a boarderline fulltime developer working in a Mac consulting group. One of the owners is also a FileMaker developer who also has Mac centric clients besides FMP work. We've progressed over the last five years to the point where we now need to develop our estimating and presentation skills. We currently are meeting with other "developing and building" professionals to help in the process.

We know that there can be jobs of any size out there. From "can you create this report?" to larger "we would like accounting, payroll, data entry to tie in with our manufacturing process". A web building friend stated that his company only does original work. (This is a luxury we don't believe we can expect at this time). They prepare a proposal after about 3 interviews and charge $2 -$5k for it. Architechtural friends say they have three types of clients usually. Whole new construction, addition onto existing or renovation (and probably expansion / addition ) of existing. They also have different complexities of jobs: warehousing to hospitals. They generally don't get involved for less than a $2k fee.

So when you quote work do you have a minimum. Do you draw up a plan that you charge for. Do you find that on larger jobs you quote more specific needs or sections of their "business process" to get an idea how the client will be to work with? Are there any estimating aids available that you know of? I have Brian Dunning's "To Market" that covers a lot of info for budding developers and I would recommend it.

http://www.briandunning.com/

I thank you in advance for your comments,

FileSquire

[Kurt Knippel]

In general I will meet ONCE with the client for free. All other meetings will be charged for at my normal rate.

I avoided all fixed priced bids and only worked on a time and materials basis.

I gave every client a best guess estimate, with at least 3 phases of work: research & design, interface & core functionality and reporting. I told them that I could not generally estimate the charge for reporting, but that it was generally the same number of hours as the rest of the development.

Back in the old day of the early nineties, clients used to sketch out screens and reports on graph paper and write a more detailled specification. This was much easier to estimate. Now a days, and especiually with FileMaker, client do not want to do this. This is why I charge for design work and meetings seperately.

I also billed every 2 weeks, net 15 days, and stoped working when a payment was missed. This kept me from working for free and allow the client a great deal of control over costs. I also allowed them to pre-pay for hours, at a reduced rate.

[Vaughan]

"... avoided ... gave ... billed ... allowed ..."

Everything is in past tense Kurt -- have you got out of the development business?

[FileSquire]

Thanks Kurt

Did you find it tough for clients to tell you what they wanted or needed. Did you meet with all those doing data entry. Did you have different rates for different things. ie charge more for interface and core functionality than research and design? Also what kind of reduced rate would you give them for pre pay?

FileSquire

Cincinnati

[Kurt Knippel]

Some clients were very easy and told me more than I needed, some were very difficult and just wanted me to start coding right away. I met with as many people as the client allowed. I generally required that they showed me all existing forms and reports, excel matrixes, ect. That way I would not be surprised at the end.

I also REQUIRED that they assign 1 person who could make ALL decisions regarding the project, from development choices to payment.

I had one rate for all onsite work, one lower rate for all work done in my office and I offered a sliding scale of discount for prepayment.

For one client I even agreed to a very low rate, in return for a guarantee of 40+ hours per week and payment on the same schedule as thier normal payroll. This was an unusual case though as I was reworking thier entire system, including servers and acting as thier defacto IT manager.

[xochi]

I also echo the sentiment to "beware the fixed-bid". In the half-dozen or so projects I've done in the last decade, every single one has increased in complexity perhaps 2x to 10x over the original "estimate". There are numerous reasons for this, such as: Client not knowing what they want, client not knowing what databases are capable of, technology changes (i.e. upgrades & changes to deal with new versions of OS and FileMaker), client's employee(s) quitting or retiring and them deciding to outsource more labor to me, etc.

That being said, in most cases you will find that the client wants an initial "fixed bid" type of estimate. Often this is for bureacractic reasons only, i.e. it's hard to get approval for an open-ended contact. I'm happy to write such a bid, but I always structure it in the form of "this is an estimate only, you will be billed for hours worked, any changes to specifications will require additional time, etc." In some cases I'll put in language such as "you will be notified if any portion of the project will go 10% or more over budget" etc.

[mav]

Looking at this issue from a Creative/Graphic Design perspective, we estimate to perspective client in dollar terms how much a new item (let’s say a logo and packaging for this example) will cost them. The client will view our folio to see if previous work style is to their liking, and decide if the price is in the budget. This estimate has a ‘Concept’, ‘Artwork’ and even an ‘Assignment of Copyright Agreement’ components.

As Creative 'professionals' we have various levels of training, qualifications, experience and 'skill'. Skill can defined as god given talent - we are either born with it or we’re not !

The concept fee varies most from project to project, and this even takes into consideration the size of their company. Internally, the concept for the logo may take 2 hours or 20 hours, based on the designer’s ability and skill to meet the brief. There is no an hourly rate assigned to this component, as ‘ideas’ are not measured in hours. The better skilled the designer, the quicker they are able to deliver a new logo concept.

The final concept for the logo is still our Intellectual Property ‘IP’ until the client pays for the job, which includes the scope of future work, which is defined in the assignment of copyright. This can also be structured so that there is ongoing return to our company by a percentage of sales of the client’s product, as the better the packaging, the better the sales of the product . . .

To put this in FileMaker terms, clients will want to see your ‘folio’ of previous work, and will be inspired by and wish to ‘refine’ one of your previous projects, with consideration to the agreement you had with the previous client, as to whether the final job was exclusive.

I think there is a lot to be said for ‘Skill’ and ‘IP’ in the FileMaker industry, which will not only dramatically influence the final paid price, but the quality of the final product. This should be reflected in your hourly rate, if this is your method of charging.

Another method is to calculate how much a client will save for every ‘entry’ in the database, and charge an ongoing percentage of that saving.

An example of this is home loan processing, where a client may take 20 minutes using their old method, and 5 minutes using you new solution. Therefore the client can increase their commissions on processed home loans by up to 4x, and the developer receives a share of this, which leads to ongoing financial return, not just the initial hourly rate.

What ever you charge your client, ensure you have taken into consideration your god given skill and ability to complete the job in a timely manner, your ability to define and meet their brief, and the client’s ability to be more productive with your final solution . . .

[flapjacks]

I always say that it could take between 10 minutes and ten years. Been at this custom thing for about 15 years now and you better spend @ 40 hours doing Q&A with the troops before you even give a ball park.

John

[tjwett]

while i don't feel qualified to comment directly on pricing a FileMaker job specifically since i'm not a seasoned vet, i'd like to share what i've learned from other types of gigs that i do have a bunch of experience with such as custom AppleScript automation solutions, web design and general tech consulting. i am really surprised to see so many people around here so quickly point to charging an hourly fee. after about 5 years of working "freelance" my experience has been that hourly has been the least beneficial to me, and my clients.

right off the bat, you've got this idea of time attached to the bottom line hanging in the back of everyone's head. the client wants it done fast and you want it to take forever. not literally, but those feelings are there if even at subconscious level. i've walked away from more than a few jobs in the past that took me 5 hours to do and ended up saving the customer thousands. or websites for example...

let's say the auto mechanic shop and the antique store down the street both want me to build them a 3 page website. for which party is a website more valuable? most likely the mechanic will have his phone number and address up there, maybe a little info. not much money can be generated because of the site. but the antique store will likely use those 3 pages to showcase inventory, perhaps even some simple e-commerce. 2 sites, same amount of hours to create. i'd happily do both jobs but i'd not be at all ashamed to say that i'd be charging the antiques store a considerable amount more for the job. it's worth more to them and will make them money and is a worthy investment for them to make.

i've switched to a completely value-based pricing method and the results have been more money for me and happier clients. don't adjust your price to meet their budget. if you say "the price for what you want is $10,000." and they say "we only have $8,000 to spend." then you should answer "no problem, we can remove some features to fit your budget."

not only will you get the most for your talents but you'll likely have clients actually asking to give you more money. it also drums up repeat business. another perfect example is reusing your solutions. i've got really nice automation system for handling catalog assembly for a specific type of customer. it initially took me almost a month to complete and the original client paid me about $9,000 for it. they estimate it to be saving them about $8,000/month. i've gotten a few similar clients requesting similar solutions and i've been able to tweak that one system in a handful of hours just for them. if i was charging hourly i'd be really getting the shaft on that.

to each his own and this is all my opinion of course. btw, i got most of the info that inspired me to rethink my hourly ways from the book "Getting Started in Consulting" by Allen Weiss. it's the best $15 i've ever spent and his info is priceless for anyone working in any field trying to make a living on their skills. do check it out, you won't be sorry.

[xochi]

I think you bring up a good point -- there is a potentially vast difference between doing 100% custom work for a client (where the client owns the copyright / intellectual property rights), and "re-branding" an existing system where you own some or all of the IP. Essentially, this is the difference between "work for hire" and "licensing".

If you are re-tooling an existing system for a new client, you have to figure in the value of the software. In that case, you might even want to consider splitting it up into several parts, each of which is billed separately: An initial consultation to determine their needs, a fee to customize the database for their business, a yearly licensing fee, and perhaps an hourly fee for further technical support and modifications after it's installed.