Building a Business Case for Dormant Account Revenue Recovery

The CFO doesn't buy "reactivation", the CFO buys a number with a source next to it. A business case for dormant account revenue recovery is that number, written down: how many accounts stopped trading, what it costs to call them, what comes back under three scenarios, and who owns the risk if the compliance desk says no. Most retention leads have the instinct and the MT4/MT5 export. What they don't have is the two-page document that turns both into an approved pilot. This guide builds it section by section, with Topcalls at $0.35 per minute on the cost side and every assumption flagged as measured or guessed.

Key Takeaways

• A dormant account revenue recovery business case has seven parts: problem statement, book sizing, assumptions, cost model, three recovery scenarios, risks, and one decision ask.
• Topcalls bills $0.35 per minute all-inclusive, so two 1.5-minute attempts across 5,000 dormant accounts cost $5,250 with telephony, recording and transcription included.
• ESMA found 74% to 89% of retail CFD accounts lose money, so the case should value a reactivated trader on the next deposit, not a multi-year lifetime.
• Harvard Business Review's Amy Gallo put the cost of acquiring a new customer at five to 25 times the cost of keeping an existing one.
• At $150 of value per reactivated trader, the 5,000-account pilot breaks even at 35 reactivations, a 0.7% rate; a 2% rate nets $9,750.
• Topcalls delivers a proposal within 48 hours of a strategy call and runs live campaigns within about two weeks, so the pilot fits inside one quarter.

1. What does a dormant account revenue recovery business case need?

A dormant account revenue recovery business case needs seven things on two pages: a problem statement in the broker's own numbers, a sized callable book, a list of assumptions with sources, a cost model, three recovery scenarios run against that cost, a risk table with an owner per line, and one clear decision ask. Miss any of them and the meeting turns into a debate about the missing part instead of a yes or no.

Brokerages get this wrong in a predictable way. The retention lead opens with the vendor pitch, and the CFO can't test it against the firm's own dormant book. Start with your book instead. The vendor comes in at the cost model, and the reactivation ROI model for dormant traders fills the scenarios.

The Dormant Account Revenue Recovery Business Case Template is the fill-in version of those seven sections: problem statement, book sizing waterfall, assumption register, cost model at $0.35 per minute, three scenarios, risk table and the decision ask, with blanks for your MT4/MT5 numbers.

2. How do you size the dormant book without overstating it?

Size the dormant book as a waterfall, not a single count. Start from every live account, remove accounts with a trade in the last 90 days, remove accounts with incomplete KYC, remove anyone without a valid marketing basis or on a do-not-call list, remove missing or malformed phone numbers, and what remains is the callable list. The callable list is the only number the business case is allowed to multiply.

The waterfall is where credibility is won or lost. A head of trading who sees "12,000 dormant accounts" in the first line and 5,000 in the cost model will assume the difference was hidden. Show each drop-off with the filter that caused it. The example below uses round numbers for a mid-sized CFD broker.

Two details make the waterfall defensible. Use the same dormancy definition your finance team already applies to inactive-account fees, so nobody argues about the 90 days. And run the consent and DNC filter before the phone-number filter, so the list you cost is the list the compliance desk has already seen. The reactivation list data checklist covers the export fields.

3. Which assumptions does the CFO check first?

The CFO checks three assumptions first: the value assigned to a reactivated trader, the reactivation rate you expect, and whether the revenue you claim would have arrived anyway. Put all three in an assumption register that states the number, where it came from, and whether it's measured from your own data or a stated guess. A guess that's labelled as one survives the meeting. One dressed up as a fact doesn't.

Value per reactivated trader is the one to be conservative on. ESMA's 2018 product intervention measures recorded that 74% to 89% of retail CFD accounts lose money, which means most reactivated traders won't fund the account for years. Value the reactivation on the next deposit and the spread or commission it generates over 90 days. Anything longer is a forecast, and finance will discount it to zero anyway.

The reactivation rate has the least data behind it before a pilot runs, so give it a range rather than a point. Use your best number from any manual win-back effort as the base case, halve it for the conservative case, and let the upside case sit where the vendor's claims would put it. The cost per reactivated trader post turns each rate into a unit cost.

The third assumption is the control group. Hold back a random 10% of the callable list, don't call them, and measure their redeposits over the same window. Revenue from the called group minus the control group's rate is what the campaign caused. Finance teams that have been burned by attribution before will look for this line specifically.

Retention economics back the whole exercise. Amy Gallo, writing in Harvard Business Review, states: "Acquiring a new customer is anywhere from five to 25 times more expensive than retaining an existing one." A dormant trader has already passed KYC and funded once. Nothing on the acquisition side of a broker's P&L is that cheap.

4. What does the cost model look like at $0.35 per minute?

The cost model at $0.35 per minute has one line: minutes dialled multiplied by $0.35. Topcalls includes the voice model, telephony, recording, transcription and analytics in that rate, with no per-seat, setup or bundle charges, so the business case doesn't need a second table of add-ons. Two attempts per account at an average of 1.5 minutes across 5,000 dormant accounts is 15,000 minutes, or $5,250.

Minutes per attempt is cheaper to measure than to argue about. Voicemails and no-answers end in seconds. A live conversation with a trader who wants to know why their MT4 login stopped working runs three or four minutes. Blended, 1.5 minutes is a fair planning number, and the pilot replaces it with a measured one in the first week.

Compare that to the internal alternative in hours rather than in wages you can't source. Ten thousand dials at 12 an hour is 833 agent hours before manager time, dialler licences or compliance review. Topcalls processes 63,000+ AI calls a day, so the same list is worked in days. Run your own figures through the dormant trader revenue calculator and paste the output into the cost model as a cross-check; the AI voice agents page lists what sits inside the $0.35.

5. How do you run three recovery scenarios?

Run three recovery scenarios by holding the cost fixed and moving only the reactivation rate: a conservative case at half your historical rate, a base case at that rate, and an upside case for what a live, multilingual voice agent does against a book that email has gone quiet on. Report net revenue and break-even for each. The CFO decides on the conservative case; the upside is there so nobody accuses you of hiding it.

The break-even rate falls out of the cost model in one line. Divide the campaign cost by the value per reactivated trader to get the number of reactivations needed, then divide by the callable list. At $5,250 and $150 per reactivated trader, that's 35 traders, or 0.7% of 5,000.

Write the sensitivity in words underneath the table. The base case survives if value per trader drops to $75, because break-even moves to 1.4% and a 2% rate still clears it. It doesn't survive if the callable list turns out to be 2,000 and the rate stays at 1%, because 20 traders at $150 is $3,000 against a cost that hasn't shrunk as much. Saying so is what separates a business case from a sales deck.

Tie the upside case to something measurable. Topcalls speaks 32 languages and answers within 500 milliseconds, which matters on a book where traders registered in Spanish, Arabic or Vietnamese and were last emailed in English. The customer reactivation solution page shows how a campaign picks the language per account.

6. Which risks belong in the business case?

Five risks belong in a dormant account revenue recovery business case: a missing marketing basis for part of the book, AI voice disclosure rules in the trader's jurisdiction, bad phone data, brand damage from a clumsy call, and counting redeposits that would have happened anyway. Each gets a mitigation and a named owner. The owner column is what tells finance the plan has been thought through past the revenue line.

Consent is the risk the compliance desk will raise first, so answer it before they ask. State the lawful basis per segment, how consent was captured at onboarding, and which accounts the waterfall dropped because the basis was missing. The marketing consent for dormant traders post covers the GDPR and PECR side, and Topcalls runs campaigns with TCPA, TSR, DNC and GDPR controls built in.

Regulatory context belongs here too. The FCA's PS19/18 restricted CFD sales to retail clients from 1 August 2019 and estimated consumers would save between £267 million and £451 million a year as a result. A reactivation call under those rules can't pitch trading incentives; it can check whether the trader still wants the account, fix the KYC or login problem that stalled them, and offer a conversation with a human. Write the call's permitted purpose into the risk table so the compliance desk signs off on the scope as well as the vendor.

Data and brand risks are cheaper to mitigate than they look. Recording and transcription on every Topcalls call are included in the rate, so QA can listen to a sample within the first hour instead of waiting for complaints. Topcalls runs at 99.9% uptime, and the secure infrastructure page lists the controls the compliance desk will ask about. Put the first 200 calls under manual review as a written step, with the retention lead as owner. The 10% control group covers the last risk: report its redeposit rate next to the called group's and the attribution argument ends.

7. What should the decision ask say?

The decision ask should be one paragraph naming the pilot size, budget, duration, owner and the number that counts as success. For example: approve a six-week pilot on the 5,000-account callable list at up to $6,000, owned by the head of retention, with a 10% control group, judged a success at 1.5% net reactivation or better. A single ask gets a single answer. Three asks get a deferred meeting.

Timeline evidence belongs in the ask, because "how long until we know" is the follow-up question. Topcalls sends a proposal within 48 hours of a strategy call, first setup takes about 15 minutes, and campaigns go live within roughly two weeks. Add a two-week calling window and a 90-day measurement tail, and the pilot reports inside the quarter it was approved. The reactivation campaign pilot post has the week-by-week plan.

Say what happens after the pilot in one line: scale to the full dormant book, or stop and write up why. Finance funds experiments with a stopping rule. For a second set of eyes on the numbers before the meeting, book a 30-minute call and bring the export.

Two pieces are worth attaching as appendices: AI voice agents for dormant trading accounts, which explains what a reactivation call sounds like, and win-back campaigns for inactive trading users, which covers the offer and follow-up sequence around it.

8. When does this business case not fit?

A dormant account revenue recovery business case doesn't fit when the callable list is under about 1,000 accounts, when consent records don't exist for most of the book, or when the compliance desk has ruled out outbound voice in the broker's main jurisdiction. Below 1,000 accounts the pilot can't produce a reactivation rate anyone trusts, and without a marketing basis the waterfall empties before the cost model.

It also fits badly when the dormant book is mostly traders who left with a complaint. Calling someone who churned after a stop-out dispute is a brand problem, whatever the numbers say. Filter those accounts out by support-ticket history and note it in the risk table.

And skip the exercise if the outcome is already decided. If the CEO wants the calls made and finance isn't gating the spend, run the pilot and write the case afterwards from real data. The document exists to win an approval.

The dormant book is an asset nobody prices. Two pages, seven sections and one honest scenario table put a number on it and get the pilot signed.