AtlasView

    How to Defend a Pavement Budget Request

    AtlasView Team Dec 2025 6 min read

    The short answer

    A defensible pavement budget request shows four things: current network condition with distribution (using whatever condition score you collect, such as PCI), the projected condition under flat funding, the funding level required to hold or improve condition, and the added cost of deferring a year. Each comes directly from condition data run through deterioration and scenario modeling.

    Budget decisions are rarely lost on the merits. They are lost because the request arrived as a number without a consequence attached to refusing it.

    Show where the network is today

    Lead with the distribution, not the average. A chart of miles by condition band tells the story faster than any average condition score — PCI or otherwise — and it makes the failing tail visible.

    Show where flat funding leads

    Project the network forward five and ten years at current funding. If condition declines, that decline is the status quo option — and now it is on the record as a choice rather than a default.

    Show what the ask actually buys

    Convert the request into outcomes: miles treated, condition held or improved, backlog reduced. A dollar figure with an outcome attached is a fundamentally different conversation.

    Show the cost of waiting

    Quantify what one year of deferral adds to the total program cost as segments drop out of surface-treatment range into rehabilitation. This is usually the number that moves the vote.

    The City of Coachella used a defensible multi-year pavement plan built this way to justify a request for a fourfold increase in its pavement maintenance budget, and now communicates roadway conditions to residents through a public dashboard.

    Frequently asked questions

    What if our condition data is a few years old?
    Use it, and state the collection date. Modeled current condition from a dated survey is more defensible than an undocumented estimate — and it makes the case for re-collection.
    How many scenarios should we present?
    Three is usually right: do nothing, current funding, and the recommended level. More than that dilutes the recommendation.

    See how this works with your own data