Share article
A few years ago, I sent our CFO an email asking for budget for a new feedback system. I wrote that customers “feel like we’re listening to them” and that it would “strengthen the brand relationship”. The reply came back in twenty minutes, one line: “Michael, and how much money will this make us?”
I had no answer. And ever since, I’ve known that this is the moment that decides whether CX gets a budget or ends up as a nice presentation gathering dust in a drawer.
CX managers are almost always right that a better customer experience helps the business. The problem is they say it in a language the CFO doesn’t understand, or doesn’t trust. “Customers will be happier” isn’t a financial argument. It’s a wish. And CFOs don’t approve wishes, they approve numbers with a return on them.
As it happens, my colleague Dan here at X Pulse recently wrote about exactly this disconnect, just from the other end. He wrote that the CX team actually has a metric for this problem sitting right under its nose, it just doesn’t use it: CLV, the value of a customer over the whole relationship. When I read it, I nearly cheered, because that’s exactly what I’d been missing in meetings for years. I agree with him completely, and it’s a great starting point. But one metric isn’t a business case yet. This piece takes it a step further: it’s a guide to switching fully from the language of CX into the language of finance, including how to get the return calculated and how to avoid the mistakes that get a CFO to say no. Not so CX managers stop talking about customers, but so their arguments survive the CFO’s first question.
Forget satisfaction, talk money
The first thing you need to do is stop selling a feeling and start selling four concrete financial levers. Every CX initiative, whether it’s a new onboarding flow, a faster complaints process or an AI assistant in customer support, can be tied to at least one of them.
Churn, meaning customers leaving. This is the strongest lever, because the impact shows up in revenue immediately. Work out how many customers leave each month and what that costs you in lost turnover, and you get a number the CFO grasps instantly. There’s a well-known finding from Frederick Reichheld’s research at Bain & Company that increasing customer retention by five percentage points can lift profit by 25 to 95 percent, depending on the industry. It’s not a universal constant you can slap onto any business, but it’s proof that even a small change in retention has a disproportionately large effect on profit. That’s exactly the kind of line a CFO wants to hear.
Retention and account expansion. A customer who stays usually also buys more. Track how customer lifetime value (CLV, roughly how much money a business earns from a customer on average over the whole relationship) changes depending on their experience. Segment customers by satisfaction score or complaint history and check whether the ones with a worse experience expand less or buy fewer add-on services. Almost every time, they do. And this is exactly the number Dan was describing, and I have to say I’m pleased we landed on it independently: CLV is one of the few metrics CX and the CFO can look at together and reach the same conclusion. The trouble is that in a lot of companies it sits in a spreadsheet on the finance team’s drive, and the CX person never gets near it. If you don’t have it, that’s the first thing to ask for, before you build any business case at all.
Upsell and cross-sell. A satisfied customer is more willing to buy a higher tier or an add-on service. This can be measured fairly precisely if your CRM data is linked to your feedback data.
Cost to serve. This is the lever CX people underrate most often, and yet it’s extremely easy for a CFO to read. Every support call, every escalated ticket, every repeat call about the same issue costs the company real money. When you cut the number of support contacts through a better product or clearer communication, you save money that can be calculated down to the penny. This is also why numbers like FCR (first contact resolution, the share of cases resolved on the first contact) translate so easily into financial terms. A low FCR means repeat contacts, and repeat contacts mean unnecessary costs.
If you can place your initiative in at least one of these four categories, you have the basis for a business case. If you can’t, it might not be an argument for the CFO at all, but for marketing or brand instead.
A framework for calculating ROI: three steps, no magic
Return on investment (ROI) in CX isn’t some kind of dark art. It’s the same process as for any other business investment, you just need to feed in the right numbers. Break it into three steps.
Step 1: Costs. List everything it’s going to cost you. Not just the software licence or the price of training, but the time of the people who’ll work on it, implementation costs, and any dip in productivity in the first weeks while the team gets used to something new. Almost everyone makes this mistake: only the “visible” cost gets counted, i.e. the supplier’s invoice, while internal people’s time gets forgotten. A CFO spots that gap immediately, and it instantly undermines your credibility.
Step 2: Benefits. Take the four levers from the previous section and work out a specific, conservative scenario for each one. If you think a new initiative will cut churn by two percentage points, calculate what that means in money given your current customer base size and average customer value. If you expect support contacts to drop by 15 percent, multiply that by the average cost of one contact (worked out from support team costs divided by cases resolved per year).
Step 3: Payback period. Calculate how long it takes for the investment to pay for itself. This number is often more important to a CFO than the overall ROI percentage, because it shows risk. An investment that pays back in six months is a different proposition to one that pays back in three years, even if their total ROI is the same.
Sounds simple. It isn’t. The hardest part is step 2, estimating the benefits, because that’s where there’s the most room to fool yourself. So always calculate conservatively, and give a range for your numbers rather than one magic figure. Better to present a range like “savings of 800,000 to 1.4 million crowns a year” than a single “savings of 1.1 million crowns”. A range shows you’ve thought about it realistically, rather than picking whichever number happened to fit the slide.
The mistakes CX people keep making in front of leadership
I spoke recently with a colleague who used to run CX at a fairly large retail company, and we agreed the same three mistakes keep coming up.
The first is arguing from satisfaction without converting it into money. NPS (Net Promoter Score, a measure of how willing customers are to recommend the company) or CSAT (customer satisfaction with a specific interaction) don’t interest a CFO on their own. What interests them is what happens to revenue or costs when those numbers improve. If you can’t make that link, leave the metrics off the first slide and open straight with the financial impact instead.
The second mistake is overestimating benefits and underestimating costs. I see this in almost every business case that’s crossed my desk. CX people tend to reach for the most optimistic scenario, because they believe in their project, which is human and understandable, but financially risky. A CFO sees through this, because spotting exactly that is their job, on dozens of other projects every day. The moment they catch one inflated number, they stop trusting the whole document, even the parts that were accurate.
The third mistake is skipping the comparison with doing nothing. The question isn’t just “what does this cost and what will it bring in”, but also “what will it cost us if we do nothing”. If you have data showing churn is rising or cost to serve is climbing, this is exactly where it belongs. Showing the CFO that the status quo also has a cost, it just never appears as its own line item, is often a stronger argument than any estimate of benefit.
And one more thing, not strictly a mistake, but almost part of the business case itself: never turn up with just a number. Turn up with how you’ll measure success once the project is live. A CFO wants to know that in three, six and twelve months you’ll sit down again and compare the estimate against reality. That alone can decide whether next time your budget gets approved faster, or whether you get grilled with detailed questions all over again.
A one-page business case
Here’s the template I use when I need to justify a CX investment. Feel free to copy it and adapt it to your own business, just make sure you plug in real numbers from your own company, not the ones from this example.
Initiative: Introducing proactive order status notifications (SMS and email) for an online shop with 40,000 active customers a year.
Problem: 22 percent of customer support contacts are “where is my order” queries. The average cost of one support contact is 180 crowns (wage costs, systems, team lead’s time). Annually, that’s roughly 26,400 unnecessary contacts costing 4.75 million crowns.
Investment cost: Notification tool licence 380,000 crowns a year, implementation and integration with the online shop 250,000 crowns one-off, an estimated 60 hours of internal time for setting up templates and testing. Around 730,000 crowns in total for the first year.
Expected benefit: A conservative estimate of a 60 percent drop in “where is my order” contacts (based on similar e-commerce implementations) means saving around 15,800 contacts a year, or 2.84 million crowns in support costs. On top of that, some modest improvement in retention can be expected, since customers who get active order updates tend to complain less and trust the brand more, but I’ve deliberately left this effect out of the ROI calculation, because I can’t back it up reliably with my own data. It’s a bonus, not part of the claim.
Payback period: The 730,000 crown investment pays for itself out of the 2.84 million crown saving in under 4 months.
How we’ll measure success: Tracking the share of “where is my order” contacts in total support volume, measured monthly for the first year. Review at 3 and 6 months, comparing real numbers against the estimate.
Notice that nowhere in this example does the word “satisfaction” or “experience” appear. There’s a problem, a cost, a concrete benefit, a payback period, and a way to verify it all. That’s exactly what a CFO needs to see to say yes.
To close
CX pays off, that’s not in question. The question is whether you can prove it with numbers that someone whose job is checking numbers will actually believe. Once you learn to speak the language of finance, you stop asking for budget and start presenting an investment decision. And that’s exactly the moment a CX manager becomes a partner to leadership, not just the voice of the customer that gets a polite hearing before the decision gets made on someone else’s numbers anyway.
