Dodge Momentum Index falls 9% in October 2026: Rebuild the 2027 Revenue Forecast From the WIP Schedule, Signed Backlog, and a Sector-Weighted Pipeline

The Dodge Momentum Index fell 9.0% in October 2026, and because planning leads nonresidential spending by 12 to 18 months, the drop lands on late-2027 awards, not this quarter's billings. Don't cut the forecast by 9%. Rebuild it from the books as three dated streams: remaining revenue on active jobs, signed work not started, and a sector-weighted pipeline.

Why this matters now

Dodge Construction Network published its October 2026 Dodge Momentum Index on 2026-11-06. The index dropped 9.0% to 260.4 from a revised 286.0 in September and now sits 2.4% below October 2025. Commercial planning fell 13.4% in the month and is down 17.6% from a year earlier; strip out data centers and the commercial figure is down 16.6%. Institutional planning slipped 1.5% in the month but is up 33.4% year over year. Sarah Martin, Dodge's director of economic research, attributed the decline to weaker data center and warehouse planning. The index tracks nonresidential projects entering planning on a three-month moving average and leads construction spending by 12 to 18 months.

That timing is the whole point for a contractor. The jobs you bill in the first half of 2027 were planned in 2025 and early 2026 and are already in your backlog. The October drop is about what will be bidding in late 2027 and 2028. Set it against the other number on your desk: Associated Builders and Contractors reported on October 13 that its Construction Backlog Indicator rose to 8.7 months in September, with contractors doing data center work at 11.3 months and everyone else more than three months lower, while more than one in four contractors expect margins to contract over the next six months. Construction Dive summed up the pair on November 6 as less confidence despite more work. Backlog is high, planning just fell, and the two facts are about different years.

Industry coverage reports the index and quotes the economists. Backlog explainers give the formula for months of backlog and a healthy range. What neither does is show a contractor how to take one national planning index and one national backlog number and answer the question that actually matters in November: how much of my 2027 revenue plan is already in signed work, how much depends on bids I haven't won, and which sectors those bids sit in. That is this article.

The three streams behind a 2027 revenue forecast

A revenue forecast for a contractor is not one number. It is three streams with different source documents, different certainty, and different timing, and the Dodge index touches only one of them. Build each from its own document, schedule when it earns, and sum by quarter.

The first stream is remaining revenue on active jobs. It comes from the WIP schedule: contract value including approved change orders, minus revenue earned to date under the cost-to-cost method. That remainder is a point-in-time balance as of the schedule date, and it earns as the jobs burn. The second stream is signed contracts that haven't started. Their full contract value is backlog, and their start dates set when it earns. Those two streams together are backlog, and backlog divided by average monthly revenue over the trailing twelve months is months of backlog, the same arithmetic behind the ABC indicator.

The third stream is the pipeline: bids outstanding and pursuits, each weighted by your historical hit rate in that sector. A $2 million data center bid with a 20% hit rate contributes $400,000. This is the only stream a planning index speaks to, because the pipeline is where projects entering planning today will show up as bids in a year. If your pipeline is tagged by sector, the Dodge release tells you which tags just got riskier. If it isn't tagged, that is the first fix.

One contractor worked through, stream by stream

Call the company Harbor Line Builders: a commercial general contractor, calendar year-end, $42.0 million of revenue over the twelve months ended October 31, 2026, and a 2027 plan of $46.0 million that the owner set in September when backlog looked strong. The figures are illustrative. The question in November is whether that plan survives contact with the October index.

Step one: remaining revenue from the WIP schedule

Harbor Line's October WIP schedule carries fourteen active jobs with $58.0 million of contract value and $34.0 million of revenue earned to date under cost-to-cost, so $24.0 million remains. The earned figure tied to year-to-date contract revenue in the general ledger after the October over/under billing entry, which is the first tie-out and the one that makes the rest of the exercise credible. Tagged by sector, the remainder is $9.0 million of data center fit-out, $6.0 million of warehouse, $5.5 million of healthcare clinic, and $3.5 million of K-12. Project managers' schedules put $6.0 million of it in November and December and $18.0 million in 2027.

Step two: signed work not started

Two executed contracts haven't broken ground: a $7.0 million warehouse starting in January and a $4.0 million medical office starting in March. That is $11.0 million of backlog, of which $9.5 million earns in 2027 and $1.5 million spills into 2028. Backlog is therefore $35.0 million. Against trailing-twelve-month revenue of $42.0 million, or $3.5 million a month, that is 10.0 months, above ABC's 8.7 and close to the 11.3 months reported by contractors with data center work. By the industry's own yardstick, Harbor Line is in good shape.

Step three: the pipeline, weighted by sector

The bid log shows $40.0 million outstanding: $18.0 million of data center, $10.0 million of warehouse, $8.0 million of healthcare, and $4.0 million of education. Hit rates from the last three years of the same log are 20% on data center, 30% on warehouse, 35% on healthcare, and 25% on education. Weighted, the pipeline is worth $3.6 million plus $3.0 million plus $2.8 million plus $1.0 million, or $10.4 million. Allowing for award dates and job durations, about $7.3 million of that would earn in 2027.

Step four: coverage of the 2027 plan

Backlog covers $27.5 million of the $46.0 million plan. The weighted pipeline adds $7.3 million. That leaves $11.2 million, or 24% of the plan, that has to come from work Harbor Line hasn't bid yet. Ten months of backlog hid a plan that is 40% dependent on unwon work. And of the $10.4 million weighted pipeline, $6.6 million, nearly two-thirds, sits in data center and warehouse, the two categories Dodge named as the source of October's decline.

Step five: what the owner changes

The owner doesn't cut the plan 9%. The index says nothing about the $27.5 million already signed. Instead the plan becomes two numbers: $34.8 million of covered and weighted revenue, and an $11.2 million stretch that the estimating team has to win, with institutional pursuits moved to the front of the bid calendar because that is the planning category up 33.4% year over year. Overhead additions tied to the $46.0 million plan are deferred until the stretch is at least half covered by signed contracts. And the cash forecast is rebuilt on the covered number, because retainage on the warehouse and medical office jobs will hold back cash well after the revenue is earned. The 13-week cash flow forecast is where that timing lives.

The cost side gets its own pass. Construction Dive's November 6 roundup put construction input prices up 11.2% year over year in September, and the re-estimate of open jobs that follows from that changes the estimated cost at completion, which changes earned revenue, which changes the $24.0 million in step one. Run the re-estimate first, then the forecast.

Accounting controls for a backlog-based forecast

Source data: executed contracts and approved change orders for contract value; the WIP schedule for cost to date, estimated cost at completion, and earned revenue; general ledger contract revenue for the trailing twelve months; the bid log with sector tag, bid date, bid value, and outcome for at least three years; project schedules for start dates and durations; and the Dodge and ABC releases, dated, for the sector context.

Calculation: a fixed sequence anyone can rerun. Earned revenue equals contract value times cost to date divided by estimated cost at completion. Remaining revenue equals contract value minus earned revenue. Backlog equals remaining revenue plus signed-not-started contract value. Months of backlog equals backlog divided by trailing-twelve-month revenue divided by twelve. Weighted pipeline equals the sum of bid value times the sector hit rate, scheduled by expected award date and duration. Coverage equals plan minus the 2027 portion of backlog minus the 2027 portion of weighted pipeline. Keep the measures straight: revenue is a flow over a period, and backlog is a point-in-time balance at the schedule date, so a 2027 plan is only comparable to backlog after the backlog has been scheduled into 2027 months. Contract value and backlog never post to the ledger. What does post is costs and estimated earnings in excess of billings, a debit-balance asset, and billings in excess of costs and estimated earnings, a credit-balance liability, and those two balances tell you how much of the backlog has already been billed ahead of or behind the work.

Review: a second person ties every contract value to a signed document and the change order log; earned revenue on the WIP ties to ledger revenue after the over/under entry; hit rates are recomputed from the bid log rather than taken from memory; every pipeline line carries a sector tag and an expected award date; months of backlog is recalculated from the ledger, not from a dashboard; and the forecast shows covered, weighted, and uncovered revenue as three separate lines by quarter.

Decision: the owner or CFO owns the plan number and the choice to hold or stretch it. The estimating lead owns hit rates and the sector allocation of bid effort. Project managers own estimated cost at completion on their jobs. The controller owns the tie-out between the WIP schedule and the ledger and the scheduling of backlog into months. Whether to add overhead against uncovered revenue is the owner's decision, made against the covered number, not the plan.

Limits and common mistakes

The Dodge Momentum Index is national, nonresidential, and a planning measure. It doesn't track residential work, doesn't tell you when a planned project will bid, and doesn't know your market. A 9.0% monthly move in a three-month moving average is large, but it is one month, and the index is still only 2.4% below a year earlier. Treat it as a reason to look at sector exposure, not as a forecast input on its own.

Months of backlog is a ratio of a balance to a flow. It rises when revenue falls, so a contractor whose trailing revenue dropped can show more months of backlog while winning less work. Read it with the revenue trend.

Hit rates by sector need enough bids to mean something. Four education bids over three years don't give a 25% hit rate; they give one win. Where the sample is thin, use a blended rate and say so.

Earned revenue depends on estimated cost at completion. Stale estimates overstate or understate the remaining revenue in stream one, and the WIP schedule built from the books is only as current as its last re-estimate.

Harbor Line is illustrative. The index and backlog figures are from the Dodge release of November 6, 2026 and the ABC release of October 13, 2026, and both will be revised or superseded by the next monthly readings.

Frequently Asked Questions

What does the Dodge Momentum Index measure, and why does a 9% drop matter to a contractor?

The Dodge Momentum Index tracks nonresidential building projects entering the planning stage, on a three-month moving average, and leads construction spending by 12 to 18 months. A 9.0% drop in October 2026 means fewer projects started planning, which affects the bids a contractor will see in late 2027 and 2028, not the signed work already in backlog.

How do I calculate months of backlog from my books?

Add remaining revenue on active jobs from the WIP schedule, which is contract value minus revenue earned to date, to the full contract value of signed jobs that haven't started. Divide that backlog by trailing-twelve-month revenue from the general ledger divided by twelve. A $35.0 million backlog against $42.0 million of trailing revenue is 10.0 months.

Should I cut my 2027 revenue forecast because the index fell?

Not across the board. Split the plan into revenue covered by backlog, revenue expected from the sector-weighted pipeline, and the uncovered remainder. The index only bears on the pipeline and the remainder, and mostly on the second half of 2027. Decide overhead and hiring against the covered number and treat the rest as a stretch to be won.

How do I weight a bid pipeline by sector?

Tag every bid in the log with a sector, compute the hit rate per sector from at least three years of outcomes, and multiply each outstanding bid by its sector's rate. Then schedule the weighted amount by expected award date and job duration. Where a sector has too few bids for a reliable rate, use a blended rate and note it.

Does backlog appear on the balance sheet?

No. Contract value and backlog are off-ledger figures from signed documents and the WIP schedule. What the balance sheet carries is costs and estimated earnings in excess of billings, an asset, and billings in excess of costs and estimated earnings, a liability, which show how much of the work has been billed ahead of or behind the revenue earned.

Once the forecast is three streams built from the WIP schedule, the contract files, and the bid log, a national index stops being a headline to argue about and becomes a question with a dollar answer: how much of next year depends on sectors that just slowed. FinBoard reads every entity's QuickBooks file into one consolidated view, so a contractor's finance team can carry contract revenue, costs, and over/under billings by job and sector, compare the 2027 plan to covered and weighted revenue as budget versus actual through the year, and drill from any variance to the postings behind it. See how FinBoard puts backlog coverage and job-level revenue on one page.