Quick answer
A deposit growth strategy for 2026 starts from one fact: deposits now move for convenience and intent at least as much as for rate. A broad promotion mostly repays balances you already hold; the cheapest new money is the rate-shopper already on your website who abandons the form or never funds the account. Win the deposit at the moment of intent, fund it inside the session, and move maturing balances into a relationship before the rate sheet has to move.
What changed: the end of deposit inertia
For most of the last decade a CFO could treat the core deposit base as sticky. Curinos' Q1 2026 review says that assumption has expired: half of checking switchers held four or more checking relationships in 2024, against 7% in 2019, and balance churn now runs about 25% higher than it did in 2019 (Curinos, March 2026).
J.D. Power's 2026 U.S. Retail Banking Satisfaction Study (March 2026) shows the same behavior from the accountholder's side. The average customer holds three deposit accounts across institutions, and 20% moved money away from their primary bank in the past three months, up from 17% a year earlier. Under-40s (23%) and affluent customers (25%) move most: the balances you most want are the ones most in motion.
The rate backdrop makes this more expensive, not less. The FOMC held its target range at 3.50–3.75% on 29 July 2026, with three dissents in favor of a hike and inflation described as "elevated relative to the 2 percent goal" (Federal Reserve, July 2026). NerdWallet tracked about 70 CD rate increases across 50 institutions in June and July 2026, with top 6-month certificates at 4.20% APY and 1-year certificates near 4.00% (NerdWallet, July 2026). Whatever the committee does next, competitors' shelf rates have already moved.
Margin has less room to absorb it. FDIC's Q1 2026 Quarterly Banking Profile shows domestic deposits up 2.1%, the seventh straight quarterly increase, while industry net interest margin compressed 8 basis points to 3.31% (FDIC, May 2026). Deposits are growing; the spread that pays for them is shrinking.
So the board's request is reasonable and the usual tool is not. Jack Henry's 2026 Strategy Benchmark found deposit growth the number-one priority for 64% of bank CEOs (Jack Henry, April 2026), and most deposit teams have one lever to deliver it: a rate special.
Why rate is the expensive lever
The Curinos finding every ALCO should carry into its next meeting is this: broad-based rate promotions capture only about 25% new money. The other 75% is the institution's own balances, moved from a lower-rate account and repriced. Institutions that rely on broad rate promotions also see roughly two-thirds higher churn than those that do not (Curinos, March 2026).
That ratio turns a modest-looking special into an expensive one. The example below uses placeholder inputs; replace them with your own before anyone quotes it.
Illustration only. Assume a $500M book of rate-sensitive balances (money market plus certificates). You run a certificate special at 50 basis points over your shelf rate. It draws $100M into the special, and 25% of that is new money.
- New money: $100M × 25% = $25M
- Existing balances repriced: $100M × 75% = $75M
- Incremental interest expense on everything in the special: $100M × 0.50% = $500,000 a year
- Incremental cost per dollar of new money: $500,000 ÷ $25M = 2.00%, or 200 basis points over shelf
The shortcut is premium ÷ share of new money: 50 bp ÷ 25% = 200 bp. The special reads as a 50-point promotion on the rate sheet and a 200-point promotion on the dollars it actually brought in. If the $75M migrated from share savings rather than from a shelf certificate, the true premium is larger still.
Keep the two ledgers apart. The $25M is a balance-sheet effect: funding you did not have before. The $500,000 is an income-statement effect: interest expense you pay every year the special stays outstanding, three-quarters of it on balances you already held. Neither offsets the other, and no single "ROI" figure should add them together.
For scale, at the $43,863 average 5-month certificate opened through Cotribute, $25M of new money is roughly 570 funded certificates.
Where new money leaks
Deposit intent dies in three places before it reaches the general ledger, and each has a benchmark.
The form. Cornerstone Advisors and Alkami's 2026 Digital Banking Performance Metrics (May 2026) put industry digital checking at 23 completed and funded accounts per 100 applicants: 77% abandonment, or 3.36 abandoned applications for every completed account. The saver who compared your special against three others and clicked "open" is inside that 77%; you paid for the rate and booked nothing.
Funding. An opened account is not a booked deposit. MX reports drop-off as high as 49% with micro-deposits and as low as 1% with instant account verification. A two-day micro-deposit wait is a two-day window in which a competitor's certificate can be funded first.
Maturity. Callahan & Associates' Q2 2026 Trendwatch found money market balances up 8.9%, the fastest-growing share product, ahead of share drafts (+6.1%) and certificates (+5.9%) (Callahan, August 2026). The fastest-growing balance on your sheet is also the most rate-sensitive, and a certificate renewed only by matching the top of the rate table is the same money on a longer fuse.
Each leak belongs to a different department and none has a line on the income statement, which is why, after every campaign, nobody can say whether the institution bought deposits or paid more for the ones it had.
Step 1 — Win the deposit at intent
Intent peaks in the minutes after a saver has compared rates and decided to move. The flow that catches it has to be specific to the product they chose; a general form that asks about checking before it reaches the certificate term is one the rate-shopper leaves.
Cotribute runs share certificates, money market, high-yield savings and checking as product-specific open-and-fund flows through one entryway, configured from more than 90 flow templates. Term, rate, minimum balance and disclosures are parameters on the flow, not a build, and the application asks only what that product needs.
Existing members and customers are recognized at the door: pre-fill, skip logic and no re-verification. With single sign-on from Alkami or Banno, a member already inside digital banking opens the special without re-entering identity or address, and the new share posts to the same relationship. For a prospect, membership eligibility, identity verification and fraud rules run inside the same flow.
Across the Cotribute portfolio, the median institution takes 58 of 100 applicants to a funded account, 42% abandonment, against the industry's 23; the best-in-class client reaches 85%. Capitol Credit Union of Texas ($225M) saw digital become its top source of new accounts within 90 days. "Our initial results have been phenomenal — on average, $10,000 in new deposits per account opened and 70% reduction in processing time," said CEO Pierre Cardenas.
The product decides the balance. Cotribute's three-year client data puts the average money market at $6,620, share savings at $1,819 and high-yield savings at $941, against $43,863 for a 5-month certificate. A treasurer who wants balances should make the certificate and money market flows the shortest ones on the site.
Step 2 — Fund in the session
The difference between a booked deposit and a promise is whether the money moves before the applicant closes the tab. Every funding method Cotribute supports completes inside the application: instant verification of an external account, debit funding and direct-deposit switch. A one-time passcode gates money movement. Joint owners (with identity verification and watchlist screening), versioned disclosures and e-signature are handled in the same session, before funding.
Account verification and deposit switch run on providers your institution contracts with directly, at your pricing: Plaid, Atomic, ClickSWITCH, InstaSwitch, Knot, Pinwheel, Jack Henry SmartPay Vault and others. Cotribute orchestrates the step and charges no connector fees; it does not build or resell the switch.
The funded account writes to the core in real time. Cotribute is integrated with 13 core platforms and real-time on 11, including all six Fiserv cores, the Jack Henry cores and Corelation KeyStone. The core stays the system of record; no core conversion or digital banking replacement is required.
For the CFO, in-session funding does one more thing: it makes the funded rate a number. When funding is a follow-up email, the gap between "opened" and "funded" never appears in the campaign report. When it is a step inside the flow, it is a conversion rate for the scorecard.
Step 3 — Convert hot money before maturity
A certificate is hot money only if the only thing you offer at maturity is another rate. Sixty days before a term ends, the question is whether the balance becomes something less rate-sensitive than it was.
Cotribute's cross-sell agent, one of three AI Growth Agents in production since June 2025, reads origination and relationship data and recommends the next product for a maturing certificate: a laddered term, a money market for liquidity, or checking with direct-deposit switch for a member whose paycheck lands elsewhere. The agent recommends; your staff, or the rules your institution has configured, approve the outreach and act. Agents never open, decline or move anything on their own, and no member or customer data goes to public models.
The offer lands as a pre-filled flow. The member is already recognized, so there is no re-verification, and the new product is opened and funded in the same session. Deposit switch in that flow turns a rate relationship into a primary one, because a paycheck is harder to move than a certificate.
The prize is the churn differential Curinos measured: institutions dependent on broad rate promotions carry roughly two-thirds more churn than those that are not. A maturity book that rolls into relationships instead of renewals moves you from the first group to the second.
The CFO scorecard
Deposits per funded account is the number the ALCO should track, and it should go up without the rate sheet moving. The scorecard below keeps balance-sheet and income-statement effects in separate rows; nothing is added across the two.
| Metric | Ledger | How to compute | Compare with |
|---|---|---|---|
| Share of new money | Balance sheet | New-to-institution balances ÷ total balances into the promoted product | About 25% for broad promotions (Curinos, March 2026) |
| Deposits per funded account | Balance sheet | Funded balances ÷ funded accounts, by product | Cotribute three-year averages ($43,863 certificate, $6,620 money market, $1,819 share savings, $941 high-yield savings); Capitol CU's $10,000 per account |
| Balance at 90 days | Balance sheet | Balance on the books at day 90 ÷ balance at funding | Your own promotional cohorts; what left before the first statement was never new money |
| Funded rate | Funnel | Funded accounts ÷ started applications | Industry 23 of 100 (Cornerstone Advisors and Alkami, May 2026); Cotribute median 58, best 85 |
| Promotion cost | Income statement | Premium over shelf × all balances in the promoted product, annualized | $500,000 in the illustration, three-quarters of it on existing balances |
| Marginal cost of new money | Income statement | Promotion cost ÷ new money (or premium ÷ share of new money) | 200 bp over shelf in the illustration; your digital cohort carries no premium |
| Operating cost released | Income statement | Manual steps removed × loaded cost per step | Capitol CU's 70% reduction in processing time |
Two disciplines make the table useful. Tag every funded account with its source (campaign, organic search, digital-banking SSO, maturity outreach) at origination, so share of new money is computed rather than estimated. And report the balance-sheet rows to the ALCO and the income-statement rows to the budget review; a deposit number and a cost number side by side tell the truth, while a blended figure hides which one moved.
Frequently asked questions
Aren't certificates just hot money?
Only if you treat them that way. A certificate opened at the top of a rate table and renewed with another rate match is money on a timer. One opened in a flow that recognizes the member, funded in the session, and met before maturity with a recommended next product (money market, ladder, checking with direct deposit) becomes a relationship balance. At $43,863, the average 5-month certificate opened through Cotribute is a first deposit worth converting rather than renting.
Should stablecoins change our deposit strategy?
Not yet, but watch them. Curinos estimates more than $1 trillion of checking deposits is "at risk" from stablecoins and new payment rails (Curinos, March 2026). That points to the same conclusion as the rest of this guide: balances held for convenience move when a more convenient option appears, and the defense is to be the institution where the paycheck lands and the account opens in minutes. Treat stablecoins as a watch item on the ALCO agenda, not a product decision.
Does this work on our core?
Cotribute is integrated with 13 core platforms and real-time on 11, including all six Fiserv cores (DNA, XP2, Portico, Premier, Precision, ClearTouch), Jack Henry Symitar, SilverLake, CIF 20/20 and Core Director, and Corelation KeyStone; CUSA and Loan Director run near-real-time. Funded accounts post to the core as the system of record. No core conversion, LOS replacement or digital banking replacement is required.
How fast can we launch a new certificate special?
For an institution already live, a certificate special is a configuration change on an existing flow: term, rate, minimum and disclosures are parameters, so the constraint is your own product approval process rather than a development cycle. For a new implementation, account opening is typically in production about 30 days after discovery, with a sandbox around day 10, a named project manager and 60 days of hypercare.
See it working on your core
Bring your own numbers: promotional premium, share of new money and funded rate. We will run them through the Why Cotribute for Finance model and show a certificate opening and funding against your core in one session, balance-sheet and income-statement lines kept apart.
Book a live demo Talk with usSources
- Curinos, Curinos Review Q1 2026: the end of deposit inertia, 22 Mar 2026
- J.D. Power, 2026 U.S. Retail Banking Satisfaction Study, 26 Mar 2026
- Federal Reserve, FOMC statement, 29 Jul 2026
- NerdWallet, CD rates forecast, 29 Jul 2026
- FDIC, Quarterly Banking Profile Q1 2026, 27 May 2026
- Jack Henry, 2026 Strategy Benchmark survey of financial institutions, 28 Apr 2026
- Alkami / Cornerstone Advisors, 2026 Digital Banking Performance Metrics release (retail and business), 7 May 2026
- MX, account-opening statistics, updated 9 Apr 2026
- Callahan & Associates, "5 takeaways from Trendwatch" (Q2 2026), 19 Aug 2026
- Cotribute, AI benchmarks: three-year client data on funded balances by product — /resources/ai-benchmarks
