The Specification Recipe for Construction Success Artwork

What the RFI?

Join Matt Brennan, Assoc. AIA as he discusses the day-to-day life in the Contract and Construction Administration world. This podcast bridges the gap between Architects, Designers, Engineers, consultants and General Contractors as they work through Construction Administration (CA) related items.

Each episode focuses on the challenges, techniques and technology to help navigate through the fast-paced construction industry.

How many RFIs did you get this week?

All Episodes

What the RFI?

The Specification Recipe for Construction Success

March 31, 2025 • Matt Brennan • Episode 25

In this episode of What the RFI, host Matt Brennan sits down with construction specification experts Steve Gantner and Dave Stutzman to discuss the critical role of specifications in architecture and construction. They explore how clear, well-structured specs reduce RFIs (Requests for Information), prevent costly mistakes, and ensure project success. The discussion covers common challenges in specification writing, best practices like early involvement, the use of Uniformat, and the importance of clear communication between architects, contractors, and owners. They also examine AI’s role in specification writing, the impact of precedence clauses in contracts, and key differences between public and private bidding. Plus, they introduce Conspectus, an innovative specification software that improves collaboration and streamlines the construction documentation process.

Takeaways

Specifications are essential for project success, bridging design and construction.
They should be clear, up-to-date, and tailored to each project.
Performance-based specifications can be challenging to write.
Collaboration among stakeholders ensures effective specifications.
Vague or outdated specifications lead to RFIs and disputes.
Specifications should align with project requirements and take precedence over drawings.
AI tools are not yet reliable for creating specifications.
Public bidding limits control over specifications.
Misconceptions, like reusing old specs, can hinder success.
Less is often more—concise, well-crafted specs improve outcomes.

Sound Bites

"Every job is the same."
"It's just like the last one."
"Two letters made the difference."
"Vague specs lead to RFIs."
"Tight specs eliminate RFIs."
"Less is more."
"Help us help you."

Chapters

00:00 - Introduction to the Podcast and Guests
03:03 - Celebrating Milestones and Personal Introductions
05:57 - Understanding Conspectus and Its Purpose
08:58 - The Importance of Specifications in Construction
12:04 - Types of Specifications: Prescriptive vs Performance
15:00 - Common Challenges in Specification Writing
18:02 - The Role of Specifications in Construction Administration
21:31 - The Importance of Specification Details
23:00 - Real-World Examples of RFIs and Specifications
26:56 - Best Practices in Specification Writing
30:00 - The Role of Specifications in Project Management
33:06 - The Future of Specifications and AI Integration
40:05 - Understanding Precedence in Specifications
46:16 - Navigating Material Specifications
49:01 - The Importance of Samples and Submittals
51:40 - Creative Uses for Samples
52:39 - Alternates vs. Unit Prices in Specifications
54:39 - Understanding Public vs. Private Bidding
56:10 - Misconceptions About Specifications
01:00:50 - The Role of Specifications in Project Success

🎙️Website - WhattheRFI.com

🍏Apple Podcast - What the RFI?
🎧Spotify - What the RFI?
🎥YouTube - @WhatTheRFI

1:00:00

Specs rule. Literally. Let's get into it.Welcome to What the RFI. I'm Matt Brennanand this is thepodcast all about CA. Today,as you guessed it, we're already talkingabout specs and this iswhat we're going to be goinginto. And we've talked about specs on theshow before, but I'veactually brought two amazingexperts to the show, both Steve and Dave.And before I get letyou guys do your officialintroduction, I just want to let you knowthat this episode is a milestone episode.This is episode number 25. It's a prettybig deal. And yeah, I'mthrilled because again,it started back in September and lookwhere we are. I, you know, the showscrest, you just breakingover almost, you know, 2,500 listens forthe whole, you know, theshow and the respect. It'skind of, it's kind of a cool thing to seethis grow over the last,you know, six months. Sothat's fantastic. It's great. But it'snot said I want the two ofyou to introduce yourself.Go for it, Dave. Who areyou pointing at, Steve? Yeah.Okay. I'm Dave Stutzman, president of theConspectus independentspecifications consultingfirm and software company. So softwarecompany is something new,but here we are. Lots ofexperience. We'll touch about that in aminute here. And I'm SteveGander. I'm executive vicepresident for Conspectus, Dave's businesspartner, future owner to be,and excited to be here withyou, Matt. Yeah. What do you guys do forfun when you're notwriting the, you know, the mostamazing things of specs and buildings?What are you guys doing? Oh,I'm here in Omaha today. I'mnot writing specs, but I'm in Omaha,Nebraska, because CSI isNorth Central region conferenceshere this week. And I just got finishedtalking to a bunch ofproduct representatives. Sospecifiers love talking to product repsbecause they are the peoplethat keep us going. They knowmore than we do about specific products,but they make us look goodand smart. But aside from that,I like to hang out with my family. I'vemarried four kids and oneof my oldest sons married,and he's got a grandson. So just lovehanging out with them.Yeah. Dave, what are you doingon the weekends? Hey, I'm missing it thisweekend, Matt. I'm abachelor. Oh, yes. The stories. Yeah,why? Well, this last week, my daughtergave birth to our seventhgrandchild. So now it's her first.Another grandson, he has me so confuseddue to lack of sleep. Idon't know what day of theweek it is. Honestly, it's Friday. It'sbeen crazy busy. Soyeah, my wife is helping outthis week as my daughter gets accustomedto someone else in theapartment besides herself.Well, that's awesome. And congratulationsto you, the wholefamily, and definitely to mom.Yes. Well, thank you. Very cool. So youmentioned Conspectus, and Ireally wanted to get into that.What is it? Tell the listeners that whatis this product? What does it do?The product, the ConspectusCloud? Yeah. Kind of what is it?It's a better way of actually writingspecifications. So wecreated it. I createdthis thing out of frustration mostly. Imean, what better way todevelop software? Trying tosolve a problem and trying to make itbetter for our team because initially wemade it only for ourteam. And we wanted to be able to make ita collaborative effortwith the entire project team,which is something that none of the otherspec writing softwarereally accomplishes,doesn't accomplish it well. So here weare on a new adventure, trying tobring all of the teams together by havingconversations in thespec writing software.Nice. Of course, architects, engineers,all using this.Contractors, suppliers, installers,owners. Everyone. Yeah. Yeah. That's thegoal. Yeah. On the flipside though, it's a testedsoftware because Conspectus, as theservice side of the company has beenusing it for over 10 yearsnow. And successfully for the past threeor four, right Dave? Wehad to beta some stuff.So our subscribers really aren't our betatesters at thispoint. They're more, "Hey,it'd be really nice if we could havethis. Can you do that?" So we're workingwith them now on that.Yeah. Yeah. Refining the product. ButSteve found a different opportunity alsofor using ConspectusCloud. We're using it as an educationaltool. I'm assuming that's what you'retalking about, Dave.We'll be actually offered to use it. I'mnot allowed to say thename of the university,but it's a large university in St. Louis.There's only two to choosefrom, three to choose from.But you might figure it out. They have anarchitecture school. And we've taught,I've taught a couple of classes downthere about specificationwriting. And we've offered it tothem to use as their final project or towrite a specification sectionand to do some TOCs and stufflike that. And I think it's successful.They've enjoyed workingwith the product and the endresult, we'll wait and see how theirfinal project turns out.But hopefully, hopefully well.Oh, very cool. And that must be like,from your guys'standpoint, it must be such aprivilege to see these firms, you know,because at the end of the day, they weredoing specs by word,you know, and that kind of thing andseeing the challenges thatcome with it. And then using thisproduct to deliver a much better endresult because things arelining up and RFI aren't,you know, being generated because ofuseless stuff, because the specs toolthat you've developedbasically complete. And we're going todive into all that throughout thisepisode about CA andspecs and RFIs and all that kind of jazz.But yeah, very kudos toyourself because I thinkthat's a very cool accomplishment to kindof sit back and go,yeah, we were part of that,even though you weren't directly in partof it, but you still were.Yeah. So think about it. You mentionedword and that was part ofthe frustration because we hadbuilt over a lot of years, because I'vebeen at this a long time,built a lot of macros to try tohelp us with some of the more mundanetasks and to be able to keepthe quality level consistentproject to project, specifier tospecifier. And so now thespec system that we developedreplicates or even does it better thanwhat Microsoft Wordcould. So we've replaced allthose macros. We don't needto even rely on them anymore.Exactly. Like simple things, changingheaders or connecting to differentproduct specs. And I'm,you know, suppliers are, you know, you'reconnecting with allthe different suppliers,hence why what you're doing right now,Steve. And then you'rebasically bringing those guys into thepackage. Correct. Yeah. One of the thingsthat I was doing this afternoon wastalking to some folkswho, you know, I was able to pull out mycomputer was talking toour decks, who is a flooringproduct below the floor tilemanufacturer. So think adhesives isgrout, things like that forceramic tile. And I pulled up our tilingsection, and I just kind of walkedthrough it with themstep by step, our products that we havelisted, and they noticedone that was incorrect. So Iwas able to jot a quick note down in thesoftware, and it goes off to Elias andElias will take a lookat that product and make the correctionsnecessary. And that can be pushed outinstantaneously as anupdate to our sections. Right. It's tomake sure that ourproducts are current and industrystandards. Right. Absolutely. No, that'sso brilliant. And that's,that really helps. Becauseagain, that's that's where the errorsare. Yeah, that's one of thebenefits too, is we try to,I don't want to say limit, but we try tomaintain our manufacturers that we cankeep up to date withwhat they're doing so that we don't havea long list of products andmanufacturers that we haveto continually check and make surethey're current. Because like you said,the RFIs you're generatedbecause a product may be discontinued, orit may have changedproperties a little bit or somethinglike that. We don't, we don't always knowthat we depend on theproduct reps to tell us.Yeah. So that brings us, and that's agreat segue tounderstanding the specifications.So what is your, you know, what arespecifications in context of theconstruction project,your take on specs, why are specs soimportant from you fromthat? I know what it is, but youknow, for the listener, some of that'sdoing, what is your take, why is specsthe holy grail for theproject? Shall I put it in really easyterms? You do mine. Okay. Sospecs are really the recipefor the buildings for the constructionproject. You think about it,you know, how many, how manytimes did you make Toll House cookies?Right? Did you follow a recipe?Sometimes. Only sometimes.That's right there on the package. Howcan you not follow therecipe? And that's the issue. Sospecifications are the recipe. If youdon't follow the recipe, it'slikely not going to turn outright. And if you forgot to put thebutter in the Toll House cookies, they'reprobably way too hard.That's probably why itdidn't taste so good there, Dave.Well, we use that as an example, too,when we're talking aboutspecifications, because they'rebroken into three parts, which ironicallyis who you work for. And we,if you think about a recipe,you have your instructions, right? Whichis part one, you have youringredients, which is part two,and then you have the bakinginstructions, which is part three, how todo it. So that's the, that'swhy recipes are the perfect analogy forit. The other way you canthink about it too, and it is,you play board games, right? So like,let's use Monopoly. Youdon't know how to play the gameuntil you have the rules, right? Well, wekind of write the rules tohow to build the building.Yeah. Yeah. Can't say I'm a fan ofMonopoly. It's just one ofthose games. You just don'tplay with friends because you won't befriends afterwards. It'salways a winner. And there's alot of losers. A dangerous one. Now withthe other type of specs,like there's performancebased perspective, a prescriptive. Whatdo you prefer in, youknow, in through this and that,again, going back to therecipe, you know, illustration?That's tough. It's going to bethe lawyer answer. It dependsreally on what the project requires. Asfar as the easiest specter, right, isprobably proprietary.Because really, you know what thearchitect wants on the project, and youjust list that one product.And that's all you have to do in parttwo. Let's say for instance,they wanted a certain type ofwindow, you can specify thatmanufacturer, that specific window, andyou're done. That's all youhave to do in part two. That's how Iprefer. Like, as you're designing thebuilding, you're goingthrough this and this process and youknow exactly what you want at the end.And so I do want thatnano wall. I don't want an alternate. Iwant this particular product. Becauseyou've done the homeworkand you've done the research. You lookedat the websites, maybeeven gone to the showroom.That's how I prefer it versus just doinga performance base where it's,you don't know what you're going to getat the end of the day. And I'm aparticular person. Ican't just go and buy a TV. I got toresearch the details ofthe TV. That's why I gotnot a performance base. So yeah,performance is actually the mostdifficult to write if it'strue performance. Because you have to setout the criteria. What is theperformance you're trying toachieve? How are you going to measure it?What's the criteria bywhich you're going to judgecompliance? And that list can get very,very long. And if you misssomething that is critical,you could end up with something that youreally don't want orsomething that doesn't satisfythe owner's project requirements. There'sonly really been, I shouldsay there was an experimentthat was successful, I believe. It was aCalifornia school projectwhere they actually did writeperformance specifications. And what cameout of that? There wasno product on the market,but as a result, we ended up withintegrated ceilings. Manyfolks today may not evenunderstand what that is, but it was theacoustics, it was the ceilingfinish, it was the lighting,it was the air distribution, it wasabsolutely everything thattouched the ceiling integratedinto a single system. And that was aninnovative solution to aperformance specificationbecause they didn't say that it was goingto be a certain acoustic ceiling tile.Right. And again, it goes back to, "Hey,I want this," but youkind of got a differentinterpretation of what you wanted. And alot of times you get acombination of the differenttypes of specifications. You can list aproduct and then you canlist the salient properties ofit that you want and allow equivalentproducts to be submitted intheir place. So you combinethis different styles of spec writing.Any particular spec canhave all different types.Yeah. So let's open up Pandora's boxhere. What are some common challenges andmisunderstanding about specifications?Every job is the same.It's just like the last one. We can justreuse the spec from that.Yes. It's just like the last one exceptit's in a different city, it's adifferent climate zone.Yeah, that's probably the most commonmisinterpretation, Iguess, of specifications.Yeah, using the one. And you're right, inthe way, Guilty is chargedbecause that's what we did withthe school. "Hey, just use the same one."At least we would gothrough the specs and correctwhat we got caught on. If there was achange order for thisitem or whatever, we actuallyspec'd... We ended up using a Pollock ACMon the building, butthe previous Leucobond orwhatever, there was a different type ofACM that we used. Andthen one of our GCs caught itduring tender and said, "Hey, thisdoesn't meet the fire testing based onall this." And it wasall due to that, I guess, the fire out inEurope or something whereit just rocks. Basically,you know what product... The Grenfellfire. Yeah, that was theone. Because they used anon-fire rated core. And basically, whathappens with non-fire ratedcore is the plastic and...Funny you mentioned that, I just had thatpresentation withGeorge. The plastic insidebetween the sandwich panel of aluminumwas dripping. And it wasdripping as it was on fire.So fire typically will not go down, butin the instance of theGrenfell project, because of theACM, the plastic was on fire and it felland it caught the lowerfloors on fell. And that's howthe fire really spread. Yeah. No. Andwhen that happened, when we got broughtup to our attention,we had, I think, eight schools and it allpulled. And then we went...We basically went to a Pollockbecause it was the next alternative andit didn't compromise thebudgets. And it was in line to whatwas priced and tender and everything. Andthat was the way we did it.But yeah, very frustrating.Lots of sleepless nights going, "Oh mygoodness, what have wedone?" And not realizing it, eventhough they keep... And even the languagewas saying, "Oh yeah,no, it's been fire tested."And everything. And you went into thefine details and you'relike, "No, it is not." So we justkept going on with that. But with specsand RFIs, now you're goingto transition here from...We've talked about what specs are and whythey are, but let'sget into the constructionadministration side and how specs arevery, very important into this world.Really the link betweendesign and construction. So the kickoffwith this question,"Inexperience, where do you see thebiggest disconnect between specificationsand how they'reinterpreted in the field and how canarchitects, contractors, even owners worktogether to bridge this gap?"Ooh, loaded question just foryou. That's a really loaded question. Oh,and Steve is the rightperson to ask since he was onthe CA side for this one. Don't you havea whole ton of examples that you can...I wish I had one particular that I couldpull up out of memory, but Ithink the biggest disconnectis reality versus wish. And Dave helpedme out here, but I... AndI'm not picking on architects.I'm an architect, but architects seethese things thatlook great in a magazine.They don't exactly know where the productcomes from, and then theywant us to specify it. Andit's kind of like, I can't specify what Idon't completely understand.And then when it comes intothe field, we've guessed at it, we'vedone our best, but it's not quite right.Does that make sense?I wanted to take it maybe a little bitdifferent direction, butI think the end result isvirtually the same. So there's been aphrase that's crept intothe design lexicon, and it'sdesign intent or architect's intent. Andif you're relying onthat, it's really difficultto try to explain an intent. And tospecify an intent ispractically impossible.So if you're trying to rely on thatphrase and thinking thatthe contractor is going tounderstand what the intent is without itbeing fully documented,there's going to be a bigdisconnect somewhere. Yeah. And that kindof ties into a clarity question too,why clear specs are so important. And I'msure you've got lots of examples aboutthat in the field too.Well, I'll go back to your ACM panel. Itwas only two lettersthat made the difference.So the core was either PE forpolyethylene, which will burn, or FR,which is fire resistant.So it seems like a very tiny detail, butyou get the two letters wrong or missing.And if it's missing, what will thecontractor price? Probably the leastexpensive. Now we'refortunate here in the US, I think all themanufacturers don'teven offer the polyethylenecore anymore. Right. Given thecircumstance and the reason history. Oh,sorry, go for it, Steve.No, I think after that fire, I don'tthink they make non FRfire rated cores anymore.Yeah, absolutely. And again, in your realworld scenario, any example of someRFIs that were caused by vague specs,anything that comes to mindwhere it wasn't quite rightor was unclear, and you could sit thereand argue back and forthwith the contractor and thearchitect and vice versa, or yourself.Probably myself.I want to go to the first one I think ofis I was asked to do aquality assurance reviewon a project, which was, it was a housingproject. And these, youwould think by listening to whatI'm going to say that I'm making all ofthis up, but this isabsolutely true. No names, because itwould be way too embarrassing. But thearchitect asked us to dothe review. And the very firstquestion I asked, because they were onlyinterested in us having to review thebuilding and closurespecifications. And I asked her why. Andthe response was, well, wecompleted the specificationsin about four hours. Now this is for amultifamily housing project would frameseveral hundred units.And what I found was that thespecification reallydidn't even match the project.So little things like the drawings mighthave shown aluminumwindows, the specificationsidentified the windows as a composite.But when I read the specification,neither aluminum or composite wereincluded. They were all vinyl. So nowwhat does the contractordo in a case like this? Because vinylwould be the least expensive. So does hejust rely on the specand say, hey, you said the spec was aboutcomposite. Youspecified vinyl. So I can giveyou vinyl. Or does he stop and ask aquestion? And notnecessarily ask the question becausethey're going to go down to the cheapestproduct typically.But I think that's whereis your take specs rule. There's whatgoverns the contract. Oh, I see Stevestarted kind of jumpingat this. He's ready to go for it. Sothere is thatmisconception. But to go back to yourprevious question, I think most RFIsgenerated because ofspecifications are because there'san inconsistency between the drawings andthe specifications. Andthere was a miscommunicationor noncommunication between the drawingperson and the specifier. Andwhat I mean is, they may havemade a late minute change to the drawingsand not told thespecifier. Or the specifier may haveassumed something that they didn'tconfirm with the drawings. So I thinkthat's where most RFIsbetween specs and drawings are generated.Just to follow on onthat, Steve, too, the otherthing that can cause the RFIs is if thefolks creating the drawingsbegin to label the productby a specific type. Because it may not bethe type that's specified. In this case,aluminum windows, instead of just callingit an aluminum, whichthey could have called it awindow instead of saying aluminum window.And then relied on thespec for what that window issupposed to be. So too much informationon the drawings can oftencreate conflicts as well.Or even worse, they named the windowproduct on the drawing.On the drawing, yeah. Because we wantthem to go back to the spec.Exactly. Exactly. No,I'm sorry. Go ahead.No, I don't keep going.All right, fine. I'll go.But is there something that other goodpractices like that, likethat you would offer out andsaying, "Hey, guys, don't do this. I'veseen this too many times onthe job." One thing I've learnedreally quickly on is don't put quantitiesin your specs. That's ano-no. Because the drawings inmy past will show that you have X amountof basketball hoops. But ifthe specs call a differentnumber or different quantity, you're introuble. Or same thing,bleachers call out so many numbersand counter provide 200 bleachers. Oh,that was from our oldlittle addition spec. This is abrand new school. Clearly, it's not 200seats. It's 1,000 seatsor something like that.Let's look at something that could have awide-ranging effect, doorhardware. You don't wantto tell them numbers of hinges in thespec. It would be in theschedule itself. And there areother things with door hardware thatcould have catastrophiceffects on the cost of a project.Yeah, exactly. And it's not just onelittle thing. That's a big extra. Andit's tough to justifyas an owner and everything too, you wantto see the project goahead. And to see that somethinglike that's missed, it's not going to bea couple dollar change. It'sgoing to be up there in thehundred thousand, maybe even a million,depends on where it is. Sogoing to the common of the specswere delivered in four hours. How manychange orders were onthat job? If they didn't fixwhat I found and found only because Ionly checked theenclosure. If they didn't fix that,there would have been, I don't know,several hundred questions.And there have been studiesdone. I mean, there was one that I foundthat was done by Navigantover 10 years ago, probably 15years now. And in that study, they lookedat projects around theworld, all sizes, all scales.And the conclusions they came to were 800RFIs per project average,10 RFIs per million construction cost.Now, do the math. And all of a suddentoo, they went back andthey asked the architects,how much time do you spend answeringRFIs? Now, this is only the architects.The average was eighthours per RFI. Times 800.Yeah, you clearly didn't make your profit on that one.Yeah, you clearly didn'tmake your profit on that one.Correct. That's a lot of man hours.Yeah. And it's unfortunate because ifthey just took the time to do it,then you wouldn't be in it. Like, youknow what I mean? It's all aboutcorrecting the drawings.Like I have dedicated checklists for alldifferent stages in the project. When itgoes out to building a permit, this isthe stuff that has tobe on there. You know,all of your code, your exiting, yourwashroom account, etcetera. And then it just kind ofevolves from there. And then you go outfor tender and it has,you have to go through that.When I would do a tender check, basicallyI would print off mydrawings in the 24 set,have my checklist and I'd redline bleedall over and go throughevery item. Even though I'm like,oh, I know it's on there. I know itdoesn't matter. Go double check it.Because it's easier now tofix it versus dealing with that RFIscenario that you justlisted. And same thing, you know,the last episode I was talking aboutdimensions. If, you know, it takes twoseconds, click, clickdimensions done. If you don't do that,take that extra effort. That, thatdimension is about ahalf an hour to do it later because ofthe process and thepaperwork and that follows.So yeah, spend your time. I guess that'sanything from this podcasttoday is just spend your time.Early.In the religions. Early.Early.Because.Well, that's a good question too. When doyou guys want to comein and do specs on a job?Well, as soon as possible. But.Like design development?That'd be fantastic. Schematic designwould be even better.Day one would be best.Day one. Yeah. And I just want to stepback a minute on thatRFI question. I think if,if there was a, there's a rush at the endof a project and itseems like every project isfacing that rush. If a lot of RFIs, Ithink could be prevented prior toconstruction even startingif the project team reviewed each other'sdrawings. I can't tell you how many timesI'm nearly 100% certain that our specswent out withoutanyone ever looking at them.Now, in my opinion, they're correct.However, you're the projectarchitect or project designer.You know what is correct and what is notcorrect. That RFI wasprobably generated because Imisunderstood something you told me. Oryou didn't verify that what I put in waswhat you really wanted.I put in what I heard you wanted. So Ithink a lot of RFIs can be prevented byjust spend that extraday. You know, don't tell me you wantyour specs on Friday when it's dueFriday. Tell me you wantthem on Wednesday. And then you look atthem Thursday. Tell meThursday night which you wantchanged and I'll get that changed out.Hopefully it's not a lot,but review it. And that's partof why I think our software, I'm going togo back to that for just aminute, but we encourage you tocomment on the specifications as they'rebeing developed so thatwe can prevent that RFI.And now jumping to the next part, when dowe want to get involved?Schematic design would be fantastic.And I'll let Dave talk a little bit abouthis favorite sayingwith schematic design.My favorite saying? Okay, but I said dayone. And the reason is the way that welook at specificationsare really two types. Okay, and we'repromoting the use of uniformat early on in a project, butuniform at the way it can be developedcan span the life of theproject. So it doesn't have to beonly early. But uniform at is a great wayto capture owner project requirements,capture system and assembliesdescriptions, start to build out whatthose assemblies areas you learn what they are as designdecisions are being made.Because Steve is getting reallyanxious for me to say it is we want tosay what you know when youknow it. Let's not make stuffup because as soon as you make stuff up,the estimator is going toprice it. And then we'regoing to be chasing bogus informationwith a bogus price and exceeding aproject budget. Nobody'sgoing to really know why. So if we cantreat the specifications much like youtreat a building modeland develop it in a level of development,just like you would inbuilding model. And we canactually define use the same kinds ofdefinitions when you're talking about aspec to say these arethe levels of information that we thinkare appropriate. And wecan do that in a uniformat base spec because hey, we don't haveto know that it's a brickcladding on the buildingto be able to specify an exterior wall.Right. But if we're goingto write a construction spec,we do need to know that it's brick to beable to select the rightsection to even be able to startthe spec. Right. So there's thatdisconnect too. And see,now you got me on a roll here.One of my passions, Matt. So bear withme. Uniform at has the distinctfeature that number one, itwas designed by estimators.Every family talking about Revit, everyfamily that you placein Revit is an assembly.Right. So now if you have the estimate, adescription, and themodel all aligned oneto one to one because they're all systemsand assemblies. Now theanalysis of the estimateand the comparison to the budget andbalancing the project costs tobe able to keep things in lineis far easier. You could say you havetrue value engineering at that point.Yeah. Because now you can make decisionsearly to keep everything in balanceand avoid what we know as valueengineering today, which isreally cost or scope cutting.Right. Now, one thing when you're sayinglike, oh, about the brickwhen we have this item and thenit applies. At the end of the day, that'syou kind of driving that because that'sall about experienceand being so many years and doing thisand going through thechallenges and the stories and allthat kind of stuff. You know, what's apodcast without the topicof AI you saw everywhere andyou see, but is there any kind of toolsthat you see today orwhere you see it going that canbasically go, oh, I've got this wall,therefore, bang, bang, bang,bang, bang, everything needs tobe applied to it. You know, this rightcoding, this rightcaulking, this air gap, whatever thecase is. Do you see that kind of wherethings may go? Maybe an opportunity?Funny you mentioned that.This was all improv too. So yeah. No, I,we've, we've done someexperiments and I know some otherspecifiers who have as well. And thestuff that comes out ofAI is not a specification.It's not, it's lacking in a lot ofinformation. And the othercomponent that could help usis we found out that what's the properway to say it, I guess it'schoking on the information.Because it can't read a drawing. It cansee the words in a drawing,but it can't reallyinterpret what those symbols are.Yeah. And this again, this is today. Youknow, it's not there. Weknow that back down the road.Maybe I don't, I don't know. It'd beinteresting. I've got, youknow, I've had the arguments,you know, a CSI conferences wheresomeone's like, well, you know, you'rejust going to model it andthen hit one button and it's going toproduce all the details. And I'm like,AI is not going to do that. It's going togo grab residentialdetails. And it's just not,it's, it would open up the door forextras, extra x-rays, becausethe details are, are just notthere. And best practices and climatechange and, you know, whereyou're, you know, being built.And, and one of the concerns too is doyou want to go out in apublic domain with AI,especially if you're dealing withprojects that aresensitive, that, that are under NDAbecause of what they are. And many of ourprojects today are underNDA, that we're not even,we can't talk about them by name. Wecan't in many cases eventell you who the architect is.So if, if that's the case, you're reallyforced to be using eitheryour own data set as part of AI,which is now somewhat limited because youonly have so muchexperience and AI is the advantage islet's take advantage of everybody'sexperience, but you, but youreally don't have that if you'rein a closed data set. Right. Yeah,exactly. It's only going to determinewhat it knows there and,and the data there. So if it's being fedwrong data, then that's, yeah, like,that's kind of like whatpart three's AI tool where view shopdrying, because it's justlooking at what's noted inthe specs versus when a shop drying comesin, just as a quick, youknow, cross comparison.And it's very cool because it's black andwhite, but in that case,what we're talking and whatyou're saying, David, is it's, it'ssubject, Mary is verysubjective to what it can be and all,you know, it's not one plus one, it's Aplus B equals C squaredand whatever the case is,it has a variety to it.There's my physics of the day.I think you have to be careful too, forusing it for the, whatyou're saying too, Matt, isbecause of the nature of shop drawings, alot of times when youget product data in,if you're looking for it to compare someitems that are in the specversus what's on the productdata sheet, a lot of product data sheetshave multiple products listed with them.So you may list a certain testing, I'mgoing to use ASTM, a certain ASTM number.And it may see that ASTM number in thatproduct data sheet, but itmay not apply to that productthat you are looking for. Right. So in away, you have to be kindof careful of what you'reasking you to do as well. Agreed. Likeyou, you're the one that's the driver,you're the one that,you're the professional. So therefore,yes, it can be a tool to help you. Yes.Yeah. And like, you know, when it comesto, you know, the podcast,like I use AI to make up myshow notes, you know, do the blog posts.I take all that and it'spretty good, but I still checkit and go through it. Cause sometimesyou're like, no, I don't, Idon't like that. I don't know.Just, just go back and you modify. It's atool, but it'll be reallyinteresting to see where itgoes. You know, post every day, peopleare commenting andseeing what, you know, are youscared or you're not, you know, adirector of CEO put a big post out there,lots of comments. Andyeah, just interesting to see what theworld is, you know, are youscared of it or you're wantingto embrace it? Is it a new tool that not,but at the end of the day,you're still the creator.You're the one that's driving the bus andit's not the machine. Themachine's a tool, you know,that's all it's there for. The hammer'snot going to make thehouse. You're going to do it.And the experience I have, I try to useAI for writing MicrosoftWord macros. After I just gotdone telling you earlier that we'retrying to avoid those. But it, it'sgenerating macros that are noterror free. So you still need to checkand prompt the AI to go backand correct the errors that itbuilt into the macros. And I discoveredone, I asked it to dosomething, it could not get it.And I, after I spent this, this, I wasdetermined that AI was going to fix this.And it was, it wasn't all of thatdifficult of a problem. So I probablyspent 45 minutes tryingto coax AI by different prompts to beable to fix this macro to actuallycomplete it correctly.And I finally gave up and I spent acouple of minutes and Iwas done, you know, so there,you know, where, where you think it maybe saving time and energy.And then, and in the end,in this one case, I couldn't even rely onit to finish it and get itright. So I had to take itover and do it myself.Curt Jaimungal Can we back up a minute,Matt? We never really finished adiscussion that we, we'vewent off on a tangent, which oftenhappens, which is good. But you hadmentioned earlier aboutprecedents and precedents clauses, as faras specs take precedentsover drawings. And that,that's a very, there used to be somethingin the AIA contractdocuments that had a precedents.And they fixed that a few years ago, Ithink it was 2017, with thenew issue of the A201 document.But there, there, there are no precedentsbetween drawings and specifications.They're to be viewedequally per the contract language nowwith AIA. So really, thereis no specs to take precedentsover this or this, whatthey instruct unless, go ahead.Matt BUnless it's a custom contract, and we seethat a lot in institutionaland public owners, where theyhave their own and they very, very wellmay have an order of precedents.Curt Jaimungal Yeah, right. But generallyspeaking, there's noprecedents. And that'san unfortunate side effect of that isthat generates an RFI,because there's a disconnect.And I mentioned that before, there's adisconnect between thedrawing and the specification.So that final review of both documents isvery important that often doesn't happen.Matt B So and here's, here's the side effect oforder of precedents. So if you set anorder of precedents,if specs do over roll drawings,which is the proper way it should beanyhow, because, you know,everybody knows specs aremore important. But and I probably willfind all the hate email in myinbox soon after we publish.Curt Jaimungal We'll link it to yourinbox. Absolutely. Forinquiries, contact Dave at this.Yes, exactly. So the side effect is thatwith an order of precedents,there really is no conflict. So if specstake precedence and thedrawings and the specs don'tmatch, the contractor has every right torely on the spec and noneed to ask a question.Because the order ofprecedence solved the problem.Matt BAnd that's how I would always kind oftalk to, you know, that'swhere you go. And that's what,yeah, they would look at that. And yeah,and they would eliminatethe RFI because they should bejust going to that as the final word. AndI'm like, said, ifyour specs aren't tight,that's where you'regonna get into problems.Curt Jaimungal Well, let's say your specsare tight, your specssay exactly what you thinkthey're supposed to say. Say I specify anEPDM roof, but theyreally wanted a TPO roof.Matt BYou got that that's a black and whitedifference in most instances.Curt Jaimungal Yeah. And notjust in the material color.Matt BRight. That's where I was going. Anyway,but there are two, those are twodifferent materials thathave two different costs andcould have be put down in two differentways. But I've had itspecified the opposite of whatyou really wanted. And then you have apotential change order in the next step.Curt Jaimungal Yeah,that's right. And like,when it comes to enforcement and shopdrawings, and you know, allin terms of that CA phase,list is no, I don't like alternates. Howdo you guys feel withthat? You know, like somethingcomes out on site, and the contractortries to slip somethingbeside you, you know, you've spentthis time writing these specs. You guyskind of prone tointroducing entertaining it, or it'sjust a flat out go away.Follow what I did. Matt BIt really depends on the architect or thedesigner of record whatthey want. And that's,that's a point of contention sometimesbecause I don't alwaysget direction on submittals.So if I'm not certain, I don't want tosay this in a negative way,but I could cause you a lot ofheartache in a way. Because I couldrequest a ton of submittalsthat you have to spend a lot oftime reviewing in CA. Right. And if I'munsure of the product, I'mgoing to ask for a few moresubmittals to make sure that's exactlywhat you want. But that'sanother point where the architectreally needs to review the specificationsbecause shop drawings are such animportant part of theprocess. Some of them, you know, if Ilist three products, andyou're okay with it as an architectfor those three products, if they submitthat product, why doyou need everything else?Just send me a note saying I'm using whatyou wanted me to use.This is the product I'm usingin this section. Done. You can just saveyourself hours of time, right?Matt B That's right. You don't need,I always love when you mention samples,the office would show up atthe door and we got a sample foryou. Okay, and it'd be a big piece oflike metal standing seam roofand you're like, I don't needa piece of this. We know what it lookslike. Thank you forsending. Yes, I know the colors,iron gray. I know thecolor. Thank you. What do we do?Matt B That's look at it from ourperspective. We wanted youto look at that piece of metalbecause that was the coloryou told us when you told us.Matt B That's right.Matt B Did your mindchange or did the ownerchange their mind between when you toldus and when it was let off or bit?Matt B Yeah.Matt B Or tender. Sorry.I'm coming to terms with thedifferences between northand south here. Bid tenders.Matt B Yeah.Matt B So let's go the other way though.Matt B Oh boy.Matt B Yes, because Imade a trip over to anarchitect's office. They were actuallyworking on at then at thattime. It was the Trump Taj Mahalin Atlantic City. Okay.Matt B Yep.Matt B Okay. And Iwalk into the architect'soffice and there's gigantic crate sittingout just outside thelobby door. So I'm asking, sowhat's in the crate? Oh, that's a sampleof the curtain wall for the building.Matt B Yeah.Matt B Right. I said, well,the building's already under constructionat this point. It's like,you're going to look at it?Because the crate wasn't even open. Itwas all nailed still, youknow, it was the shipping crate.Now, we're not going to open it. I waslike, wait a minute. Youknow, you've got this full-scalesample that you asked me to specify andyou're not even going to openthe crate to look at it? No.So they didn't. I took the sample. We hadit sitting in our office.Matt B Nice. I justimagine, like as you said,the crate and the, you know, it wasthere. The flashes, youknow, the Christmas story,right? The old classicChristmas story, the lamp.Matt B Yes. Yes.Matt B I just imagine you just opening itup and, you know, it says fragile.Matt B Just this, thisimagine, you know, but yeah,that's quite interesting. I got one ofthose samples too. And what the heck?Matt B But I mean, that sample, you justthink about it's onesample. This probably costssomething like $10,000 to build thissample. It was not a small sample.Matt B It was not small.Matt B Oh, geez.Matt B It was like six-inch deep curtainwall with full glazing.Matt B Wow.Matt B It was probably about a four footby four foot sample.Matt B That's a good, wow.Wow. That is a really decent size.Matt B Yeah.Matt B So where is it now?Matt B Dumpster.Matt B Recycle.Matt B Yeah.Matt B But think about it.Matt B You don't get all the samples andstart buildingsomething with them, do you?Matt B No.Matt B No.Matt B Actually, you can because a friendof mine actually took abunch of those 12 by 12insulated glass samples and incorporatedthat into the exterior wall of his house.Matt B Nice.Matt B So he built it as a bunch oflittle 12 by 12 windows.Matt B That's cool.Matt B Pretty neat little idea, but onlyan architect would do that.Matt B Yeah, exactly.Sometimes I look, I'm like,there's quite a bit of material that wecan make a doghouseout of this or something.Matt B It would be the most creativedoghouse of samples andthis product, this product,this product, and why not?So you need to be careful what you'respecifying, what you'reasking. If you don't need it,let's not torture the contractor and thesubcontractors. Let's not have the ownerpay for something that's reallyunnecessary. And that goes with thesubmittals, the samples,anything else.Matt B And as far as alternates go, Imean, if you're really indoubt, yeah, the alternates aregreat and can be very beneficial. And Ithink that's where youwere coming from at one point.But also on the flip side too, you canuse unit prices to do the same thing,for different types of materials that,hey, we want it, this isreally what we want. But if itcomes in too much money, can we swap itout for this other thing that we know isa lot less expensivethat we approve of.Matt B Yep. And I wantedto touch on a point aboutenforcing that and you kind of justanswered right thereSteve, it was like, you know,you as the architect is enforcing it, youknow, the spec writers onthe job that you're pushing,if the contractor is saying, well, no,I'm going to do this one.No, you guys are, you know,specs. This is what, like I said, this iswhy we designed, we'vedone this. And therefore,we're just going to stick to the plan.And then we're not going togo to some other third-partyapplication, alternate, whatever the caseis. At least that's, yeah,that's kind of the way I'mtaking it too, right? So, but if you doget that pushback, how do you handle it?Matt B And pushback fromwhom? Because right now,we're only dealing withthe architects really.Steve That's true.Matt B We don't dealwith the contractors,unless the architect brings us in to ask.Steve Yeah.Matt B But the pushback is likely todepend on whether or notit's privately bid projector publicly bid project. So privatelybid, the architects, theowners can do whatever they want.They don't have to accept anything as analternative or a substitution. Publiclybid, you don't have the same kind ofcontrol because it has to beopen. It has to be availableto anybody that can prove that they canmeet the spec. So it'sfar, far harder to enforceto get exactly what it is that youspecify on a publicly bid project.Matt B Unless there's aspecific set of criteriathat will make or break that material andthe other person companycan't meet that criteria.Steve Right. But thenyou have to be careful too,that you haven't created a proprietaryspec that'smasquerading as a performance ordescriptive spec because then thatbecomes subject of apotential bid challenge.Matt B That becomesintent or letter versusyour letter of the spec or what's yourintention at the end of the day?Steve Yeah. And that's up tointerpretation, right?Matt B That's right.Hence going back to thecookie recipe and it was, do you believeeverything you read? I don't know.Steve I don't likebutter. I like margarine.Matt B Oh boy. Nowyou're opened up the date.Steve Yeah, sorry.Matt B What were you using in the recipewhen you're making it?The butter or margarine?Steve M Did you follow the directions?Matt B There's the question.Matt B There is the question.Steve Oh, I was supposed tomake those? I just ate them raw.Matt B You said chocolate chip, right?Matt B Yeah, chocolate chip, absolutely.Steve M When I put itin my vanilla ice cream.Matt B Yeah.Matt B And you just did it all together.Steve M And you didn't have enough leftfor the cookie. So theywere just whatever they werewithout chocolate chips. Matt B In theend, the audience was satisfied.And we all lived from there. So here'syour, as we're kind ofcoming up to the hour here,let's wrap this up, lots of goodtakeaways, you know,contradict points. This is awesome.And thank you for that. Thanks for beingon the show for this.So your soapbox question,right? If you could change one industrywide misconception about specifications,what would it be? And why? Let's gofirst. Who's brave? Who's the brave onethat wants to go first?Matt B You want it, Steve? Go ahead.Steve M I'll take a crack at it. I thinkthat the biggest misconceptionsis that specifications areinterchangeable. And they can be used onmultiple different projectsat multiple different timeswithout any review or anything byanybody. If I couldchange it, I would say that weas specifiers, we don't like to bemushrooms. There's a commonjoke around the specifyingcommunity that we're treated likemushrooms. We're kept in the dark andwe're shed a lot of.Bring us out of the dark. Treat us likean equal on the team. Wewill help you if you help us.You know, we can we can help you makeyour project much better.Maybe not much better. We can help youmake your project better byhelping you be informed ofmaterials. Like I was saying earlier,before we started that,you know, I might have you'revisiting with product reps. I havecontacts that you might nothave. That if you don't know whatyou want, I can put you in touch withsomebody that can get youwhat you want. And it can makeyour project a little bit better foryourself and for the owner. Ultimately,we're all working forthe owner, right? We want the owner tohave the project they wantat the end. Let us help you.There's my soapbox.Yeah, well said.I have a different take.Go for it.All right. So misconceptions. You thinkabout constructionspecifications. And the first thingeverybody thinks about is CSI format,master format, sixdigit section numbering and50 divisions. And I'm saying that's notall that's available.So I'm going back to uniform at that canbe a specification to dowe need the traditionalconstruction specifications foreverything? My answer isno, we do not. So we can writean entire project specification usinguniform at if we really need thetraditional constructionspecs, let's plug them in where we needthem, but only where we needthem. And I think if we lookedat projects in this kind of a way that wemight see the specificationvolume shrink dramaticallyand if that's the case, I really believethat enforcement andunderstanding will be much simpler.So if you had and we proved this on oneproject where I delivered auniform at based architecturalspec only for a cancer infusion center,25 pages was thetotal architectural spec.It was built from that. I had noquestions about thespec during construction.It's possible.Very interesting. I'm usedto books about very large.Yes. And why? Why do we have them? Whyare they so large? Isit really necessary?Yeah, just to cover your basis, don theeyes, crossing the T's. One example,and I don't want to take away from yourmoment, your momentand everything, but likeit was a curtain wall. We didn't have theanti twistingblocking in there. It wasn'tcalled up in the specs and it was anextra. So less is more.Maybe that's kind of the theoryto kind of say with what you're gettingat. Yeah. So yeah, following on for me,so yeah, less is more. I'll quote anotherfamous person, Jerry McGuire.Help us help you.That's awesome. That's the best. There wego. Mike drop from Steve.He got the award for that.You know, less is more. Let me help youout. Exactly. Where canthe audience find you guys?I know you already told what you were.You're located right now with our Steve,but where can they findyou on the social side?You know, don't call me at Omaha. I'm nothere after tomorrow.Oh, no. Okay. All right. But LinkedIn,you know, this is yourplug. Where can they find you?Where can they reach you? Well, they canalways find us on ourwebsite, conspectusinc.com.It has all of our team members listedthere. Just go to theabout us page and you'll beable to find all of us with all of ourcontact information. Yeah.And please reach out. We'dlove to talk to you. Dave and I both lovetalking specs. We'rejust kind of geeks that way,but other people in the firm are in thesame way. Even if it's asimple question, hey, how would Ido this? Even if you're not a client,call us up, ask us a question. I'll takea few minutes to helpexplain specs to somebody. Yeah. Andspeaking of testament oftruth, like, you know, even you'reat the conferences and on top of that,you're leading classes.Like I really, really enjoyedthe design build spec class that we didwith the Lego that youbrought to Houston last year. Andagain, that was awesome. It's in my heartfor life and everything.Because I so enjoyed that.It was a good analogy and a differentperspective and a take on that type ofworld and how to kindof wrap that up and tying all back to thespecs and the differenttypes that you can do. Yeah.That was really fun to kind of pulltogether. That was acollaborative effort between Dave andmyself and Charlie Dunn, anothergentleman that you shouldprobably talk to at some point.I'll get you in touch with him. Yeah.He'd be great for you. Butyeah, it's always fun to teachabout specs. Yeah. But yeah, thank you somuch for being on the show.This was an absolute pleasure.It was a blast. Can't wait to be inperson again at the next conference. Andthere's a lot of lotsof them are coming up. So I'm sure we'llbe there and do itagain. Absolutely. Yeah. Well,forward to it. Yeah. And you're comingback on our podcast eventually too.Yes. Yeah. Let's do it. Yeah. And deliverwords. Let's do it.Absolutely. So after I'm off toEurope in the next couple of weeks andthen I'll be back and then I,yeah, let's schedule it in andput in the books because I can't wait. Ihope it's a pleasuretrip. It is. It's going to beexploring everything. It's been 22 yearson the book. You know, it'sbeen now I've been planningthis one for a long time. So it's time tokick it off as the bucket list item.Are you taking the Jeep with you to dosome trail lazing? That would be cool,but not in this case.Okay. But we will be going to the Omaha,Juneau beaches and doingthe whole D-Day landing.Oh, nice. Very good. Big passion of mine.100%. So we'll talk aboutthat sometime. Oh yeah. We'lldefinitely bring that. We'll do it.We'll, we'll connect intoyour show and then I'll get backfrom the trip and we'll go into that. Butyeah, well that wraps upthis week's episode. Architectskeep designing and contractors keepmaking those blueprints into reality.We'll see on the next one.

People on this episode

Matt Brennan

Host