Synergies are the easiest number in a merger to announce. They are the hardest one to check afterwards. I wanted a recent fintech deal where the buyer kept reporting progress, quarter after quarter, so I could compare the promise with what showed up.
Rocket Companies gave me one. I read the announcement, both 2026 quarterly releases and the August call. On 31 March 2025 it agreed to buy Mr. Cooper Group "in an all-stock transaction for $9.4 billion in equity value, based on an 11.0x exchange ratio." The release called Mr. Cooper America's largest servicer and Rocket the nation's largest lender.
Rocket put the synergy case in one line. The deal was "expected to generate annual run-rate revenue and cost synergies of approximately $500 million".
Then it split the number in two. Rocket expected "$100 million in additional pre-tax revenue from higher recapture rates" and from attaching its title, closing and appraisal services to Mr. Cooper's existing originations. It projected "$400 million in pre-tax cost savings from streamlining operations, corporate expense and technology investments."
I think the logic behind the $100m of revenue is simple. A servicer collects monthly payments on a mortgage it may not have made. When that borrower refinances or moves, the servicer has a head start on the next loan.
Rocket said the combined servicing book would exceed "$2.1 trillion in unpaid principal balance", covering "nearly 10 million clients, representing one in every six mortgages in America." Together the two companies "generated $4 billion of servicing fee revenue in 2024."
Rocket also said the merger was "expected to be accretive to Rocket's adjusted earnings per share immediately after closing." I noted the word adjusted. It comes back later.
Before reading Rocket's later reports, I went back to the research. The paper I use is "How Do Mergers Create Value?" by Erik Devos, Palani-Rajan Kadapakkam and Srinivasan Krishnamurthy, published in the Review of Financial Studies in 2009.
They used Value Line forecasts for 264 large mergers. "We estimate the average synergy gains in a broad sample of 264 large mergers to be 10.03% of the combined equity value of the merging firms."
Most of that was operating, not financial. Tax savings added "only 1.64% in additional value, while operating synergies account for the remaining 8.38%."
The finding I remember best is about where operating gains come from. "The operating synergies are generated primarily by cutbacks in investment expenditures rather than by increased operating profits."
That sentence made me reread Rocket's own wording. Its $400m of cost savings includes "technology investments". By the paper's logic, that is the kind of saving that tends to show up.
Closing came in October 2025. Rocket's release for the first quarter of 2026 reported that integration had "progressed ahead of schedule".
It gave one concrete milestone, "migrating more than half of the servicing portfolio to our unified servicing platform". Then it moved the date. "We are on track to realize our original $400 million expense synergy target and expect to realize the full amount by the end of 2026, one full year earlier than planned."
I had assumed synergy targets mostly slip. This one moved the other way. That is true at least on the company's own measure.
The second-quarter call on 6 August 2026 added detail. Management said: "We realized $100 million of annualized Mr. Cooper expense synergies in the quarter, in line with our expectations."
In the questions, the company said it was "about halfway through the realization of the $400 million goal as of the end of Q2". It expected "the other $200 million" in the second half.
Then came the raise. Rocket said it now had "line of sight into approximately $100 million of annualized expense savings above our original goal of $400 million", to be realized "in the first half of 2027."
So the cost target went from $400m to about $500m. I did not expect a raise this early. The revenue side was described only as "more than halfway" to its target "on an annualized run rate basis".
Here is the part that cuts against the tidy story. Every Rocket synergy figure above is an annualized run rate. The costs of getting there sit in a different table.
Rocket's second-quarter release reports GAAP net income of $229m and adjusted net income of $441m. The reconciliation between them includes two lines of deal costs.
"Acquisition-related expenses" were $99m in the quarter and $178m in the first half. A footnote says they "primarily" consist of transaction costs for the Redfin and Mr. Cooper deals and the Up-C collapse, a change to Rocket's share structure, "such as professional service fees (including integration costs), and severance expense."
"Amortization of acquired intangible assets" was $112m in the quarter and $225m in the half. That line "reflects amortization of intangible assets related to the Acquisitions", meaning both Redfin and Mr. Cooper.
Add the two half-year lines and you get $403m. Over the same six months, Rocket says it reached about $200m of annualized expense savings, which is a rate, not money already banked.
I cannot split the $403m between the two acquisitions from the release. I also cannot tell how much of the $200m run rate reached the first-half income statement. My guess is that it was well under half.
This has nothing to do with the synergy targets themselves. For the third quarter, management guided to expenses of about $2.35bn at the midpoint, including about $110m of intangible amortization and $100m of one-time acquisition costs.
I expect the amortization line to keep running for years, because acquired intangibles are written off over their useful lives, while the one-time costs should shrink as the integration work that produces them comes to an end. I find that asymmetry more interesting than the headline target. I keep thinking about it.
Back to the point.
I did one more piece of arithmetic. The result is rough. If Rocket reaches $500m of cost savings plus $100m of revenue synergies, the yearly total is $600m. Against the $9.4bn equity value of the deal, that is about 6.4% a year before tax.
That figure is not comparable with the 10.03% in the paper. Devos and his co-authors express total synergy gains as a share of both firms' combined equity value, while Rocket quotes one year's run rate. I am recording it only as a sense of scale. I would not quote it as a return.
On its own measure, Rocket is ahead. The $400m cost target was pulled forward a year. Then it was raised to about $500m.
The GAAP measure is slower. In the first half of 2026 the two deal-cost lines came to $403m. In the second quarter alone, GAAP net income was $229m against $441m adjusted, and those two lines were the largest items in the $212m between them. The paper reminds me of one more thing. Synergy value is measured against the whole deal, not one year's run rate.
I don't know when the cumulative savings overtake the cumulative costs. I looked for it. Rocket does not publish that bridge. Its adjusted figures remove the $178m and $225m lines I would need. That gap still bothers me more than any target in the original release.