Metal Roofing Specifications: "No Visible Oil Canning" equals Mission Impossible.

On my first visit to the job site after the roofing contractor had started installing standing seam metal roofing, the Clerk called me aside and said, "I think we've got oil canning."

I looked at the newly installed metal. It looked fine.

"Wait a while", he said. As the sky changed and the angle of the sunlight changed during the day, oil canning appeared, then disappeared, then reappeared differently. "The specs say 'no visible oil canning'," the Clerk accurately noted.

The spec writer's best intentions, possibly influenced by the designer's best intentions, were that from Day One and under any and all light conditions the metal roofing would exhibit no visible distortion. Seeing the situation in the field, a part of my mental faculties that were still intact suggested we had a problem that was more perceived than real. Over the next few days I visited and photographed a number of other projects with metal roofing - from afar, up close, and under different light conditions. These appeared to be carefully crafted installations with little or no distortion related to the installation. Yet under some light conditions and viewing angles, they all exhibited 'visible oil canning'. I put the findings into a little presentation for the client, who had been alerted to the oil canning condition by the Clerk. Following the presentation, the client was satisfied that the metal roofing was normal in terms of visible oil canning. The spec writer? Well, that's another story.

(Visit http://www.smacna.org for an October 2005 newsletter article on oil canning and information about how to minimize it.)

Construction Tolerance

Where codes or regulations establish minimum or maximum values and allow no tolerance outside the limit, industry standard construction tolerance should be incorporated into the design. For example, concrete ramps that must comply with "zero tolerance" maximum slope requirements for wheelchair accessibility should be designed at less than the maximum slope, such that allowable construction tolerance for concrete surfaces will not result in a slope that exceeds the maximum for wheelchair accessibility. If the maximum slope for wheelchair accessibility is 1:12, a design slope of 1:13 (or 0.95:12) may provide sufficient tolerance for construction to meet the slope regulations.

Masonry Design: Separate hung masonry from foundation-supported masonry

When drawing up elevations for masonry walls, in addition to locating control joints (arguably "expansion joints") to accommodate expansion and contraction related to temperature and moisture, it is important to consider how masonry supported on lintels hung directly from deflecting structural steel will move differently from masonry that is supported directly by the building foundation. While "loose" lintels supporting masonry over openings usually rest on foundation-supported masonry at the ends of the spanned opening, hung lintels should be isolated from foundation-supported masonry in a way that allows for independent vertical movement related to deflection of structural steel. If the hung lintel overlaps the foundation-supported masonry, a soft joint similar to that below a relieving angle may be appropriate below the end of the hung lintel. And the control joint above the lintel should also be located to allow for independent movement of the two masonry conditions. Determining the location of the ends of the hung lintel is important for construction that will properly respect the different needs of the two conditions. It gets more complicated for hung lintel assemblies suspended from beams that are interrupted by columns. In any case, careful consideration should be given to the locations of control joints within the wall area that is supported by the hung lintels.

In some cases the best control joint locations for wall performance will not be the best locations for aesthetic intent. But, then, cracks are not usually attractive in appearance.

Roof Drainage Design: Foresight or Afterthought?

If we could just get it a little further away or get it to go over there, everything would be okay.

Purposeful chaos (the 2-minute design offense)

While throwing staff at a project at the last minute is likely to be counterproductive (see Staffing a design project), an imminent project deadline can ignite productive and purposeful chaos that is otherwise hard to generate. Consider the possible accomplishments of the "2-minute offense" or "no huddle offense" that are so well known and widely used during the last minutes of a football game. A scripted sequence of plays may produce effective results in a short time. Deadlines and completion milestones are important for efficiency, because work tends to fill the time available, and it will likely take more hours (and therefore cost more for labor) to do a task without an imminent deadline. Interim or "partial completion" deadlines with well-defined scopes of work and timely, attentive reviews can be helpful in keeping a project on schedule and also on target in terms of design intent. Architecture is like other work when it comes to efficient production.

Mike (see Paint the Color Chip and Mike's Abridged Edition) offered another idea for managing labor costs on a design project. "If you know you are going to have to work overtime on a project," he said, "do it in the beginning. It costs less to do overtime when there are only one or two people working on a project."

Architectural Design: CAD doesn't care

CAD (computer aided design) is a wonderful tool, but it is often credited with attributes that it does not possess. CAD does not care if a picture drawn by an operator is incomplete or nonsensical. It does not care if structural plans and architectural plans for the same building are based on different dimensions that should instead be the same. It does not care if a column is missing or if a mechanical duct encroaches into building occupants' headroom. It does not care if the roof slopes away from the roof drains. It does not care if there is no flashing or if the building leaks. It does not care if exterior grade is above a window sill. It does not care if the building is uncomfortable or if mechanical equipment has no electrical connections. It does not care if wheelchair users cannot get through a door or down a hallway or if building occupants cannot safely exit the building. It does not care if a building is energy efficient. CAD is a wonderful tool, but "caring" is not one of its attributes.

Masonry Construction: A hodgepodge of reinforcing

A recent obituary (darn those things!) reminded me of an event years ago during the construction of additions to a local school. As the architect's project manager, I attended weekly meetings at the construction site. During the meetings we talked about the status of various parts of the work. I recall asking during one early meeting when the mason would start work on the new gym, which would be located on the back side of the existing building, out of sight from the contractor's construction office where we were meeting.

"He's almost done with it," the superintendent answered.

"Really? When did he start?" I asked.

"Yesterday," said the superintendent.

"Wow! I guess I'd better have a look," I said.

We walked around the end of the existing building to the location of the new gym, and, sure enough, the block back-up walls were full height, and the masons were installing brick veneer. I climbed the scaffolding to the level where the masons were working. Looking closely at the construction, I noticed that a variety of horizontal reinforcing had been used, and none of it was the type or dimension specified or approved through the submittal process. It was apparent that the masonry subcontractor had made use of a lot of scrap material left over from other projects. On the surface, the brickwork would look good, but the construction was clearly unacceptable. Although the masons had obviously invested a lot of effort in producing the work that was nearly complete, it would have to be redone.

As the architect, I could not direct the contractor to tear down the walls. Instead, I told the contractor's superintendent that I would not be able to approve any payment for the walls due to the use of incorrect reinforcing. The superintendent translated the message to the masonry subcontractor, who shouted obscenities at me but quickly took down the walls and soon rebuilt them with the correct reinforcing. The masonry work went very well on the rest of the project.

Masonry Design: A subtle difference in brick sizes

It may come as no surprise that bricks are available in different shapes and sizes. But, in at least one situation, the difference in size may be too subtle to notice until you see an area of infill that does not match the rest in coursing or joint thickness.

"Hey! Look at that small area of bricks with the fat mortar joints. What happened?"

Many U.S. buildings that are decades old were constructed of "standard" brick, each unit with an actual length of 8 inches. Today's "standard modular" brick of similar height and depth typically has a nominal length of 8 inches including the mortar head joint and an actual length of 7.625 inches, allowing for a .375 inch mortar head joint. (An actual length of 7.5 inches may also be available to accommodate a .5 inch head joint.) If you are renovating or adding to an existing brick building, it's a good idea to measure the existing brick. And, if you can't get a matching size, you may want to consider that in designing areas of infill.

(Similar issues can apply to matching existing brick color and bond. It may take a close look to spot a Flemish bond, a possible indication of underlying construction that differs from today's typical cavity wall or brick veneer construction.)

Productive coordination meetings on the go

When an architect assembles a conference room full of consultants for a coordination meeting, power and status can sometimes interfere with productive coordination. Some of the consultants may have traveled for hours, may feel uncomfortable or out of their element, or they may be unnerved by having to make a presentation in the architect's office. They may not be at their best when attending such a meeting. And they may tune out much of the conversation, feeling like they are wasting time waiting for their turn to speak or to go home. Coordination points may be lost during the meeting or on a long drive home. "We talked about that," recalled the architect. "I'm sorry, but I don't remember that," replied the consultant.

Some of the most productive coordination meetings can take place when the architect goes to the consultants' offices. The consultants are on their own turf and may be more comfortable with the conversation. They may also have better access to coordination information and materials. Individual members of the consultant's team can be on call and participate only as needed, while continuing to be productive with other tasks in their office outside the meeting. A meeting in the consultant's office can be more economical for the consultant, and it can also be beneficial for the architect to see where and how the consultant's business is organized.

Building Thermal Movement: Everything in a building moves

We tend to think of buildings as static creations.

Sure, doors and windows may move, elevators and escalators go up and down, water runs through pipes, air through ducts, and fans turn on and off. But we tend to think of the structure, the walls, the floors, and the roof as fixed components that do not move. In fact, they all do move. They expand and contract at different rates in response to changes in temperature and/or humidity. Bricks and wood swell with moisture absorption; concrete shrinks as it cures; aluminum expands at a much higher rate than steel under the same temperature increase. Some movement is reversible; other movement is not. Much of this movement is imperceptible until the forces of expansion or contraction result in a visible scar like a crack in a wall or broken glass or water damage. Alas, the rewards for successfully accommodating building movement are not usually published.

Value Engineering can add value

Architects and engineers may be used to thinking of Value Engineering as a form of design torture, where value is stripped out of a project design in order to save money. But Value Engineering can also provide an opportunity to consider values that may enhance the performance of a building over time. It is a good time to consider or revisit questions of maintainability, access for maintenance, and flexibility for changing uses over time. One use change that is common but rarely considered is the conversion of non-storage spaces to storage use; this is often done informally by building users and without consideration for the capacity of the floor or other supporting structure that is being loaded. In some cases the structural capacity of a floor can be increased by including additional floor joists and reducing the spacing between them, and the initial additional cost may be minimal for a substantial gain in flexibility. (Storage uses may also bring other requirements that should be considered in making the decision.)

It takes timely money to make a project go

A project owner was feeling frustrated that the contractor, who was doing work of good quality, was not going fast enough. Considering the contractor's monthly application for payment, the owner decided to not pay the contractor for the previous month's work in the hopes that the decision would cause the contractor to work faster. Instead, the contractor's work slowed to a crawl as the contractor's employees and subcontractors stayed away from the project. Subsequently, the contractor submitted a delay claim against the owner. The owner also heard that work on another project in the area had sped up during the same period of time.

A different owner on a different project was not satisfied with the quality of part of the work done by the project's contractor, but chose to pay the contractor for that work anyway, hoping the contractor would correct it. Instead, the contractor proceeded with other work, making correction of the unsatisfactory work less practical and more costly. When the owner complained to the contractor about the contractor's failure to correct the unsatisfactory work, the contractor pointed out that the owner had paid for it, so the contractor considered it complete and thought the owner did, too. The following month, the owner withheld part of the contractor's payment to cover the cost of correcting the previous unsatisfactory work. Then construction progress was slowed by a delay in delivery of materials of approximately the same value as the payment amount withheld by the owner. The material supplier had furnished similar materials for several other projects done by the same contractor, and delivery for this project was contingent on the supplier's receipt of payment for a previous project.

During a very busy construction boom, a project owner decided to make advance payment to the project's contractor as an incentive to get construction started on his project. Then several weeks passed without any construction activity on the project. When the owner complained about the situation to a friend who had been a project owner during a previous recession, the friend told him he had done the same thing to help a contractor get that project started during the recession - and the result was the same.

It takes money to make a project go, but payment at the wrong time can be counterproductive.

So, you want to count trucks?

It seems there is always a lively discussion about how to quantify additional earthwork on a project. While established standards may stipulate that payment will be based on compacted, 'in place' volume determined by survey, someone always wants to use a different method for determining quantities of material removed or imported, loose or compacted - and it seems there is always someone who wants to count truckloads.

There are several problems with counting truckloads. First, of course, someone has to be present whenever the trucks are coming and going in order to count them ("Hey! No bathroom breaks!"). Then, the counter needs to be able to differentiate between a truck he has just seen and another one that is in the same vicinity ("Hey! That looks like the same truck. The driver just went around behind the building and came out again. That's cheating!"). In some cases the counter may need to determine how to consider material that is transferred from one truck to another truck or trucks ("Dude! That truck dumped its load on a pile over there, and the material was picked up and taken away by two other trucks. I counted 3. Should that really be 1? 2?"). Next come the questions about how much material is in the truck: "Is this a 7-yarder or a 10-yarder? Did you fill it even with the top of the dump box? Are the corners filled?" And, oh yeah, "How loose is the material? How should I determine a 'legitimate' cubic yard when I see all these spaces between the chunks in the truck?" What is the actual 'fluff' factor? And, finally, "This hole looks like it should have taken ten truckloads to remove the material, but I counted fifty! And it's looking like it will take 42 to fill it in."

These truck counting problems may be some of the reasons why authorities who establish such standards prefer to use 'in place' volume measurement to determine quantities.

But, hey, go ahead and count the trucks, if you want to. Don't forget to count that blue one over there. Is that coming or going?

The Schedule of Values deserves close attention

On public building construction projects (and, possibly, on most private building projects) the contractor develops a Schedule of Values that forms the basis for monthly applications for payment (also known as "requisitions") that will be submitted as construction proceeds. The Schedule of Values is typically submitted for architect and owner review at the beginning of the construction project and well before the first regular application for payment will be submitted. When reviewing the contractor's proposed Schedule of Values, it is important to consider the meaning of the specific line item descriptions on the schedule and to also consider the architect's and owner's ability to assess progress on those discrete parts of the project as applications for payment are submitted by the contractor.

The meanings of the line items may seem simple enough at first glance, but they have been known to be interpreted differently by subcontractors, contractors, architects, engineers, and owners as construction proceeds. Site work items seem to be uniquely subject to varying interpretation. The intent of terms like "Excavation", "Cuts and Fills", "Fill", and others are good agenda items for a meeting to discuss a proposed Schedule of Values. For the architect and owner who will be assessing the value of periodic progress on a given item, it is especially important to understand what is included in the line item and what is not included - in part to avoid a possible disagreement as construction proceeds. "I think the owner already paid for that in this item," says the architect. "Oh, no," says the contractor or the sub. "That was never in that item. It was in this other item."

The architect's and owner's ability to assess progress on the listed work items also depends on the monetary amount of the particular item. It is common to require a contractor to break down items that are too large (e.g., Site Work $2,000,000) into smaller items (e.g., Rough Grade south yard: $15,000, Fine Grade west field: $7,500, etc.) that are more easily assessed and add up to the total value of site work. For some, the rule has been to break down the schedule into items no greater than $20,000 in value. That may not be a practical rule to follow in every case, but it is important to consider the ability of the observer to assess progress that is claimed on a given application for payment while avoiding confusion with other work that may be similar or within the same trade. Contractors and subcontractors are usually quite interested in receiving prompt payment for completed work, so it follows that they should be willing to invest time in the beginning of a project to establish a common understanding of the items in the Schedule of Values.

Another rule, sometimes challenged, is that once the Schedule of Values is established, it should not be changed. That's a good rule (with a notable exception that sometimes a more detailed breakdown of an established item may become necessary in order to assess progress and determine appropriate payment). And there are other rules....

Form suddenly meets function

One Monday morning many years ago, the boss came into the office and announced that we were going to enter a design competition for the West Hollywood Civic Center, which would be located on a block across the street from Cesar Pelli's Pacific Design Center. He had seen an ad for the competition on Saturday evening while he was looking through an architecture magazine, and he had come up with a concept that he was sure would win the competition. Soon after telling us about the idea, he sent in a check to get the program coming through the mail, and he called in an old friend who was also a space planner. The boss had already drawn up a building plan before the entry requirements and program information arrived in the mail. His plan - a giant bow tie - would be symbolic of a Hollywood he knew.

When the program and related information arrived in the mail, the space planner went to work, and we also enlisted help from a cost estimator. The program information stressed the importance of the budget.

After several days, the space planner and the cost estimator met quietly with the boss. Then they both left the office, and the boss presented the rest of us with a dilemma.

The space planner had found that the program spaces could not fill the immense bow tie plan that the boss had created. All of the departments and their discrete spaces could be sprinkled along the perimeter wall of the bow tie plan at each floor, but that left a vast amount of unoccupied space in the interior of the building. The construction cost estimate exceeded the project budget severalfold. We were also running out of time; the submission deadline was only a few days away. We would have to abandon the bow tie, cut off part of it, or reduce it. Considering the boss's situation (and ours), we ruled out the first two options.

We determined an ideal efficiency for the building, and we photographically reduced a copy of the building perimeter plan by a percentage that would accommodate the program area at each floor while maintaining a high net to gross efficiency. Then we shuffled and fit the program areas into the reduced bow tie plan. It would not have been the best plan for building occupants and furnishings, but it was mathematically efficient. Also, the cost estimate shrank to fit the budget.

Alas, the firm's entry did not win. The entry's call for knocking down all adjacent buildings on the block - some apparently occupied by competition jurors - may have influenced the outcome and kept the bow tie design from receiving an award. However, there were stranger entries among the hundreds that found their way into the display tent. One, perhaps given more serious consideration, was a billboard-like building featuring a giant Marilyn Monroe.

Pragmatism trumps engineering

An industrial client company's in-house engineers had spent weeks studying the expected loads and determining the structural design criteria for the new industrial building's floor, a concrete slab on grade. They anticipated stacks of steel plate, and the resulting loads (approximately 1,000 psf) exceeded those normally experienced in the area. Push came to shove when the developer's structural engineer determined the number and spacing of piles that would be necessary to support the anticipated loads..........and when the contractor determined the cost of the support system. "The cost of the foundation will exceed the cost of the building several times," the contractor reported in a meeting with the client company's managers and engineers.

The client company's engineers were not fazed by the cost information, but the client company's managers were more interested. A senior manager asked what loads were normally considered when designing a slab on grade for an industrial building, and the developer's engineer replied that 250 psf was a conservative norm. "That will be fine," replied the senior manager. And so it was.

Ask the right person or get the wrong answer

We were working on a fast-track office building project. The project duration was 6 months and 4 days from the start of design to owner occupancy. It was a 4-story, 125,000 sq. ft. office building to be built on deep friction piles in very deep silt. Construction started while we were sharpening our pencils. We expected questions from the contractor, and the contractor's personnel were aggressive in pursuing the information they needed to advance the project toward the ambitious substantial completion date. Contractors may learn from some school of practice that they should call the highest authority in order to get the information flowing (a.k.a. rattle the highest cage). In this case, the contractor's junior project manager made it a practice to bypass the architect and structural engineer and to call the developer's representative for the smallest detail of information, like embedment length of anchor bolts.........important, for sure, but an annoying subject for a developer's representative who has just negotiated a $7 million deal for his employer and his client. The developer's representative quickly tired of receiving such phone calls from the contractor's junior project manager. One day, in response to the project manager's phone call, he said, "Don't call me again..........for the rest of your life."

Monitoring construction progress

One effective way to monitor progress on a construction project is to consider cumulative payments to the contractor or claimed percent complete against an estimated payment schedule that should be established at the start of the project. In most cases the estimated schedule of cumulative payments or percent complete should look like an 'S' curve, similar to the one in the diagram above. It is likely to be relatively flat at start-up, due in part to the limited number of trades that can work on the project at first. As the project advances, progress allows the addition of trades (and workforce), indicated by the steeper line in the middle of the 'S' curve. The flatter tail end of the project represents a reduced workforce, fewer trades, finishing details, and punch-list work.

If actual progress looks much different than the 'S', the project may be headed for trouble. For example, a prolonged flat period at the beginning (or a shallower slope on the steeper part of the 'S' as shown by a dashed tangent line above) may forecast a late project completion. Alternatively, if requested payments suggest a steeper curve earlier in the project, the requested payments could be exaggerating actual progress, and one bad result could be releasing too much money too early and not holding enough funds to complete the project. The 'S' curve should not be considered an absolute measure of progress, but it can be an effective tool for comparison and forecasting.

[Note that a similar approach can be used for looking at design progress. See Staffing a design project.]

Staffing a design project

I remember (about 30 years ago) being enlisted to help finish construction documents for a hospital addition and renovation project. The project out-to-bid date was fast approaching, and many of the drawings had not yet been started. I was assigned to do one of the reflected ceiling plans first. When I started asking questions about ceiling systems and materials in this room and that, it became apparent that no decisions had been made about those things. We were up against the deadline, and more staff was being added to the project team every day. As staff were added, new questions surfaced. The engineers' work and coordination would have to follow our initial architectural work on these drawings. What a mess!

Managing staff level through the life of a design project is very important to a successful outcome. Staff levels should be increased gradually but early enough to allow for a project learning curve and also to allow for the tail end of the schedule to be spent tying up loose ends and putting the finishing touches on details (Curve A in the diagram above). A project staffing approach that follows Curve B in the diagram above suggests a combination of under-staffing and/or a need for a longer time-line.

[Another approach to this is to consider an 'S' curve, where the X axis is time (as in the diagram above), but the Y axis is cumulative staff hours or staff costs or percent complete. With that diagram, the steepest part of the 'S' should be in the middle of the diagram, and it should be close to flat at the beginning and at the end that represents project completion. This same 'S' curve approach can also be helpful in monitoring construction progress on a project. See Monitoring construction progress.]

Zero Slump Concrete

After a few of us had taken the architect license exam, we were talking about the questions. One of the multiple choice questions asked how long to wait before removing the cone during a concrete slump test. My colleague had chosen answer "c. 3 days".

Why we need to manage architectural projects

Everyone has an agenda. If we don't manage the projects, they may be managed by individual interests like the following.

1. "I already know what I'm going to do. Just tell me when to start."
2. "I'm here to draw. I'm not interested in any program information."
3. "I'm only going to do what I'm told, nothing else."
4. "Finally, a chance to demonstrate my design prowess!"
5. "I think we should use those windows that were presented to us at the last Lunch-n-Learn. Wasn't that a great tart?"
6. "This is not a linear process. It takes as long as it takes."
7. "I'm not going to let Jim ruin another project."
8. "We never meet the budget, and we never finish on time. It's not my fault. That's just the way it is. Why can't you get that?"
9. "Someone has to tell the owner to butt out. We already know what they need."
10. "I don't mind attending a few meetings, but I'm not actually going to start doing any engineering work until I know the architect has stopped screwing around with the plan. I'm not getting enough fee to spend engineering time on design refinements."
11. "We're going to do the whole project with our standard details."
12. "Hey! It will be really cool if we tilt the walls over like this so they look like a roof but will still be made of wall materials. I wonder if it will keep the water out. Maybe it won't rain much. We can specify really good sealant. Maybe it will be okay this time."
13. "Green? I'm not going to say it in these words out loud, but I refuse to be held accountable for any guinea pig design experiments or systems. Therefore, I'm going to do exactly the same thing I've done for the last 20 years. It's all going to be off the shelf. I don't want any call-backs on this project. That's what I call efficiency."
14. "I will be calling all the shots. Period."
    

Construction, Toledo, Spain

Design Review

Presenting the windows

"The mullions in Option B are about 1/16 inch wider than in Option A, and in Option C the horizontals are also 1/16 inch wider. We wanted to make sure you had a lot of choices. Do you have a preference?"

We usually locate the roof drains near the low point

We were working on the layout of the roof plan for a large, irregularly shaped building, designed with the structure gently warped to shed water to the edges. A drafter was adding roof drains to the plan, and I noticed that he was locating some of the drains about half way up the slope between the low point and the high elevation. I asked him why he was doing that, and he replied that he thought the drains at mid-slope would collect half the runoff on its way to the low point.

"Oh, I forgot to tell you," I said. "We usually locate the roof drains near the low point." He looked skeptical until I added, "It's office policy."

Value Engineering gone awry?

Want to know where you stand?

A notable roof warranty exclusion

Manufacturers of single ply roofing membranes offer warranties that vary in duration and conditions of coverage. The warranties typically include a lengthy list of requirements and exclusions. One of the most notable exclusions, which appears in all caps and/or boldface type on some warranties, is a statement that excludes representations or promises made by a representative of the manufacturer which differ from the provisions of the published warranty.

The 'sales rep' exclusion highlights a common dilemma. The manufacturer's sales rep typically has to please several "masters". The rep has to please the architect and project owner in order to get the manufacturer's product or system specified. Then the rep has to please the roofing contractor, the first-line customer who will actually purchase and pay the manufacturer (or distributor) for the product. And, of course, the rep has to please the manufacturer with productive sales. In construction, full payment usually follows installation, so the manufacturer's rep may be involved in site visits, inspections, and warranty processing related to project completion. In order to get paid upon project completion, the rep has to pursue the satisfaction of all these parties - the roofing contractor, the architect, the owner, the manufacturer, and maybe even a few others. That may be a tall order under circumstances where the interests of those parties are not all aligned, and one outcome may be a promise that cannot be kept. The 'sales rep' exclusion suggests that manufacturers have had experience with reps making promises that are contrary to the terms of a published warranty, and the exclusion protects the manufacturers from that problem.

Considering again the question "Who is the customer?", I recall a conversation with a manufacturer's rep several years ago. We were concerned with the quality of work that was being done by a roofing contractor, and we thought the manufacturer's rep might be able to help persuade the roofing contractor to do a better job. The rep's response to our request was very enlightening: "I can't really help you out with that," he said. "The roofing contractor is my customer."

Manage your consultants' communication with your client

"Where am I supposed to plug in my computer?" asked the owner's Food Service Director as we surveyed the room.

"Oh, we did not realize you would be using a computer in the storage room," I said. I thought she had confused this storage room with her new office, which was located in another area.

"This is not a storage room," she replied. "It's my office. I told the kitchen consultant that I wanted this room to be my office. I did not want my office in the other area. He said he would take care of it."

When I looked at the food service equipment drawings, I realized the kitchen consultant really had changed this space from a storage room to an office, just as the Food Service Director requested. Unfortunately, he did not share the information about the request with anyone else, so the room had architectural, mechanical, and electrical features that were appropriate for storage use and not for office use.

Consultants should report back to the architect any owner requests to them for design changes. The changes could lead to other changes, and the overall costs and consequences may be such that the owner - once aware of them - would make a different decision. Consulting agreements can include provisions for reporting separate meetings and communications with the owner. It is also good practice to cover communication protocols in project kick-off meetings and to issue periodic reminders, especially when you are aware of a possible upcoming meeting between a consultant and a representative of your client. The clients may also benefit from reminders about communication protocols and communication channels for requesting changes.