Surveys: the questions worth asking before you bid

A survey is a standing set of questions the team answers about a project before and during estimating. Site conditions, scope boundaries, permit ownership, schedule constraints, long-lead items, the assumptions the number will rest on. It is deliberately boring, and that is the point: the questions do not change from job to job, so the answers become comparable, and the gaps become visible.

Most estimating groups already have this knowledge. It lives in the head of the estimator who has done fourteen of these, in a folder of notes from the last similar project, in a conversation someone remembers having with the superintendent. That works until the person is on another pursuit, or three bids close in the same week, or the job is unusual in a way nobody thought to ask about. The survey is what turns that knowledge into something the company owns rather than something individual people carry.

What it is, and what it is not

The pre-bid survey answers a narrower question than the reviews around it, and it is useful to keep the boundaries clear.

The survey feeds all four of the others. A question answered at pursuit start about who owns the utility permit is the reason the general conditions duration is right five weeks later, and the reason nobody discovers the problem during final bid review when there is no time left to price it.

It is also not a bid/no-bid scorecard. Those overlap, but a go/no-go decision weighs strategic fit, client relationship, backlog, and competition. The survey is about project facts, and it stays useful long after the pursuit decision is made.

What standardizing actually buys you

Consistency across estimators. Two estimators reviewing the same project find different things, so thoroughness ends up being a function of who had time that week. A standing question set puts a floor under that. It does not make a junior estimator into a senior one, but it does mean the junior estimator is asked the same questions the senior one would have thought to ask.

Risk surfaced while it is still cheap. Questions about existing utilities, permit timelines, single-source materials, or restricted access produce answers that change a number. Asked at pursuit start, that is a pricing decision. Asked after the bid is assembled, it is a scramble, and the treatment is usually a round number nobody can defend later.

A record of what you assumed. Six weeks after award, the useful artifact is not the estimate. It is the answer that says the survey asked about winter access on 14 July, the superintendent said the north drive would be available, and here is who said it. That trail is what makes an assumption reviewable instead of arguable.

A knowledge base that compounds. Each completed survey is a data point about a project type, a client, a region, a site condition. After twenty of them you stop guessing at questions like typical dewatering exposure on this kind of site, because you have twenty answers and the outcomes that followed.

Building the question set

Start from failures rather than from a template. Pull the last two years of jobs where something surprised you (a change order that should have been foreseen, a schedule assumption that broke, a permit that arrived late) and write the question that would have caught it. That list is more valuable than any generic checklist, because it is specific to your project types, your geography, and your delivery methods.

The usual categories, with the kind of question that earns its place:

Two rules keep this from bloating into a form nobody completes.

Every question has to be able to change something. If the answer would not move a price, a schedule, a risk treatment, or a go/no-go, it is documentation, not a survey question. Move it elsewhere.

Every question needs an expected answer format. "Any site issues?" produces "no." "Are geotechnical borings complete, and does the report identify groundwater above the lowest slab elevation?" produces something you can act on. Precision in the question is what produces precision in the answer.

Flag the answers, do not just collect them

A survey that returns a page of prose gets skimmed. Grade each answer so the problems are visible without reading.

The category people skip is the last one. Marking a question not applicable and saying why is what prevents the same question being reopened three times by three different reviewers.

An unanswered question is a red flag, not a blank cell. This is the same discipline that governs silence in a subcontractor proposal: the absence of an answer is a finding. A survey with fourteen blanks two days before close is telling you something specific about the state of the bid, and the blanks should be counted and reported rather than quietly ignored.

Use follow-ups instead of a longer list

The fastest way to kill a survey is to make everyone answer every question on every job. Conditional follow-ups keep it short for simple projects and deep where it matters: if the answer to a seismic upgrade question is yes, ask about structural peer review and delegated design responsibility; if the site is occupied, ask about phasing, noise restrictions, and after-hours access; if a material is single-source, ask about lead time, escalation exposure, and whether a substitution is permitted.

The base survey should be short enough that skipping it is never the faster option.

Who answers what

Sending the whole survey to the estimator guarantees the answers that require a superintendent or a project executive get guessed at.

Assigning ownership does two things at once. It gets more accurate answers, and it removes the ambiguity that otherwise surfaces after award, when the question of who was supposed to check the permit status has no recorded answer.

Keep it alive

A survey completed once at pursuit start and never touched is out of date the moment the first addendum lands. Three habits keep it current.

1. Re-run the affected sections when documents change. An addendum that moves a phasing requirement invalidates the schedule answers. Flag them for re-confirmation rather than assuming they still hold.
2. Attach the answers to the estimate, not to a folder. The survey should be reachable from the estimate it informed, and its red and yellow flags should appear in the pre-submission review as open items with owners.
3. Run a short postmortem after every bid that surprised you. One question: what did we learn that a question would have caught? If there is an answer, the question goes into the survey before anyone forgets. This is the entire mechanism by which the survey gets better, and it is the step teams skip.

The survey you start with will be mediocre. The survey after thirty projects and thirty postmortems is one of the more valuable documents your preconstruction group owns, and it is not something a competitor can copy, because it encodes what went wrong on your projects specifically.

About the savings numbers

You will see claims that preconstruction planning saves 15% to 20% of project cost. Treat those with the same caution as any other round industry figure: there is no authoritative benchmark for the cost impact of a pre-bid questionnaire, and studies that report savings from "better preconstruction" are measuring a bundle of practices, on a particular project population, with a particular definition of savings.

The honest version is more useful anyway, because it is measurable on your own work.

Track those for a year and you will have a number that belongs to your company rather than a vendor's marketing page.

Where teams get this wrong

Where Piper fits

The reason pre-bid surveys decay is not that anyone disagrees with them. It is that answering them well means reading a lot of project documents under time pressure, and that is the first thing to get compressed when several pursuits peak together.

Piper reads the bid set alongside your company's standards and answers what the documents can answer, with the clause or drawing behind each response, so the team spends its time on the questions that genuinely need a superintendent or a phone call. Flags, sources, and owners stay attached to the estimate, and what the survey establishes about the project feeds the scope and review work rather than sitting in a finished questionnaire. If you want the process this feeds into, start with the final bid review playbook.