Your CX budget may not be ineffective at all. It is simply unprovable. For the next budget round that makes little difference, because cuts land where nobody can defend a figure.
The figure behind the headline
In the survey "The state of customer journey success 2024-2025" by CX Network, a specialist platform for CX leaders, 23% of respondents said they struggle to demonstrate any financial return from their journey management at all. The question was how they prove this return to their senior management (news item on the survey). Journey management means the ongoing management of the customer journey: designing and improving touchpoints, from first interest to contract renewal.
Before you use the figure, you should know three pieces of context.
First, its age. The survey ran from 15 July to 26 August 2024, and the report appeared on 25 September 2024. So the figure is about two years old.
Second, the sample. The respondents were 107 members of CX Network from the Americas, Europe, the UK and Ireland. That is a small, self-selected group of professionals and does not represent the economy as a whole.
Third, the sender. The report was produced together with a software vendor and sponsored by it. That does not make the results wrong. It is still part of the context.
What the survey shows
The remaining 77% named a way to prove the return: 27% through customer retention and its revenue impact, 22% through positive feedback and satisfaction scores, 19% through cost savings from more efficient processes, 9% through more closed deals.
The same survey asks about obstacles. In 2024 the most frequently named were isolated data silos and a lack of integrated systems, cited by 76%. In the previous year's survey the figure was 64%. The explanation in the news item: silos make it impossible to follow a single customer's experience across the entire journey. The increase is an indication, but not yet a clean comparison over time.
What it does not prove
It says nothing about mid-sized companies in the DACH region. Whether the share would be higher or lower at a company with 20 million euros in revenue cannot be derived from it. So do not extrapolate the 23% to your market.
Nor does it prove that the remaining 77% have solid evidence. Citing satisfaction scores as proof shows a mood, and that is still some way from a return in euros. How large the share without genuine financial proof really is cannot be read from the survey.
And it proves no cause. That silos are the most frequently named obstacle does not mean they are the reason for the missing proof. The survey shows that the two occur together. Why we still start with the silos follows below from the logic of the data chain.
As a warning signal, it is enough. Even among people who deal with customer journeys every day, a significant share struggle to put a figure on the benefit. Can you?
Why the burden of proof sits with management
Proof fails exactly where budgets are decided. If you are the managing director or head of operations who releases money for journey and CX work, you have to be able to explain to shareholders, the advisory board or the bank what the investment has achieved. "Customers are more satisfied" is rarely enough when an effect on revenue or costs was expected.
Without proof, a pattern repeats itself. In the budget round, a spend with no proven return sits next to items with a clear calculation and is cut first. Journey maps get drawn, initiatives start, and after a few months nobody follows up, because no measuring point gives them a reason to. In the end, CX is seen as a cost centre: necessary, but to be kept as small as possible.
An analyst blog post by Forrester from January 2026 describes the same mechanism. Journey work gains credibility with leadership when initiatives are tied to outcomes. Where data and metrics are not connected, ROI stays anecdotal and slow, and journey management stays vulnerable in budget cycles. The post is an analyst's assessment without figures of its own.
Which systems get connected, who is responsible for customer data and which business metric is fixed in advance are decisions the CX team or marketing cannot make alone. Senior management makes them.
The fallacy: more dashboards do not produce proof
When the benefit cannot be proven, the same reflex often follows: a new reporting tool, better visualisation, ten additional metrics. Behind it is the assumption that the proof already exists and only needs to be made visible. That is rarely true.
A dashboard does not calculate anything new. It shows what the connected data can deliver. If the journey initiative sits in the marketing tool and the business result in the ERP, the screen shows two truths side by side: here, 4,000 customers were contacted; there, renewals rose in the quarter. (The figures are made up.) No interface can tell you whether one has anything to do with the other. For that, you would need to know which customers received the initiative and whether exactly those customers renewed. That connection is made in the data.
More metrics don't help either. Open rates, satisfaction scores and ticket volumes document activity. Management's question remains unanswered: what would we have lost without this initiative?
Whatever sits in separate systems without a shared key cannot be linked, however good the chart on top of it looks.
So for every new reporting project, ask first which data it brings together. How it looks comes afterwards.
Where the chain breaks: data silos between initiative and revenue
At its core, proof of ROI for journey work is a simple calculation. You need three pieces of information: which customer an initiative reached, what that customer did afterwards and what that means in economic terms. In many mid-sized companies, these three pieces of information sit in three systems.
- Newsletters, campaigns and onboarding emails run in the marketing tool. There, the customer is an email address.
- Enquiries and complaints sit in the support system. There, the customer is a ticket number or a sender.
- Contracts, renewals and revenue sit in the ERP or CRM. There, the customer is a customer number.
Each system works on its own. None of them knows that the email address, the ticket sender and the customer number are the same company. What is missing is a shared customer ID, meaning an attribute that lets you find a customer unambiguously in every system. Without it, the question "Did the onboarding flow improve the renewal rate?" goes unanswered. The answer is not bad in that case. There simply is no answer, because the data never comes together.
The typical break points
In mid-sized companies, the chain usually breaks in the same places.
At the handover between tools. A lead is qualified in the marketing tool and handed over to sales. A new record is created in the CRM, often with no reference to the campaign it came from. From then on, the customer's origin can no longer be traced.
At manual exports. Anyone who wants to bring campaign and revenue data together pulls lists from two systems and matches them in a spreadsheet. That works once, for a presentation. It cannot be repeated, depends on one person and breaks at every variant spelling of a company name.
At the missing BI team. If you go through the organisation chart of a company with 5 to 100 million euros in revenue, you will rarely find anyone whose job is to bring data together across system boundaries. Marketing looks after the marketing tool, service the ticketing system, accounting the ERP. The connection between them belongs to nobody.
This leads to the most important break point: there is no owner of the customer data. Nobody is responsible for making sure that "customer" means the same thing in every system. As long as that role is missing, every link remains a one-off project. How to settle these responsibilities in a workshop is shown in data ownership mapping.
Why every new tool makes the problem bigger
A gap is usually followed by another tool: one for surveys, one for chat, one for automation. Each brings its own customer record and its own logic of who a customer is.
As a result, the number of connections needed grows faster than the number of tools. With three systems, you need three connections so that each is matched with every other. With five systems, it is ten. Every connection that does not get built is a place where the individual customer drops out of sight. More tools therefore do not automatically mean more knowledge about your customers.
Self-test: can you answer these five questions today?
Before you decide on new tools, find out whether proof is possible in your company at all. The five questions need neither consultants nor a new system. Sit down with the people who look after the CRM, the support system and accounting.
1. Can a single customer be traced unambiguously across all systems? Pick any existing customer and look them up in the marketing tool, the support system, the CRM and the ERP. Is it clearly the same customer every time, or does someone have to match names and email addresses by hand? If you get stuck here, you are at the most important point.
2. Is it documented which journey initiative started when, and for which customers? An initiative might be a new onboarding flow, a changed callback process in service or a reminder before a contract ends. If the start date and recipient group exist only in a presentation or in the project lead's head, the initiative cannot be tied to any result later.
3. Is there a comparison? Impact only shows in comparison: a measurement from before the start, or a group of customers who did not receive the initiative. Without either, you cannot separate what the initiative achieved from what was down to the season, a price change or chance.
4. Who is responsible for the customer data? This means a specific role: who decides which customer number applies, who cleans up duplicates, who approves a new connection between two systems? If the answer is "everyone, a bit", it is nobody.
5. Are cancellations, repeat purchases and service costs attached to the same customer ID? Check whether, for a group of customers, you can see in a single report who cancelled, who bought again and what service cost for them. These are exactly the figures that translate journey work into euros.
How to read the result
| Result | Meaning |
|---|---|
| Five times yes | Proof is possible. If it is still missing, the cause lies in the analysis and the architecture is fine. |
| No on 1 or 5 | Proof is currently impossible, however good the reporting is. |
| No on 2 or 3 | The data might be enough, but initiatives are not planned to be measurable. That can change from the next initiative onwards. |
| No on 4 | Any fix for the other questions only lasts until the next tool. |
Many companies we talk to answer no to at least one of questions 1, 4 or 5. That is no reason to stop journey work. It shows you where to start.
Becoming able to prove value without a complete rebuild
Proof is created before the initiative. If you only ask how to measure the benefit after the start, you will usually find neither a baseline nor a connection between the systems. You don't need a new platform for this. You need five decisions before the next initiative starts.
1. One journey and one business metric. One section of the customer journey and one figure that appears in the accounts. For example, onboarding of new contract customers and cancellations in the first year. Or the service process after a complaint and the repeat purchase rate. If you try to prove three goals at once, you end up proving none.
2. A consistent customer ID across the two or three relevant systems. For example, the customer number from the ERP, which also appears in the marketing tool and the support system. Only in the systems that matter for the chosen journey. Often it is enough to add the number as a mandatory field and match existing records once. That is data work, and you don't need to buy software for it.
3. A baseline measurement. Record where the chosen metric stands today: which period, which customer group, from which system. Without this value, all you can do later is claim that things have improved.
4. A comparison group or a before/after comparison. A comparison group is the cleanest option: some customers go through the new journey, a comparable group does not. If a change affects all customers, you are left with the before/after comparison. It is weaker, because the season, prices and market conditions also move the figure. So fix the period in advance and note other influences as you go. How a comparison group is kept for the duration of a measure and evaluated quarter by quarter is covered in proving the value of CX with a control group.
5. Three roles with names in the organisation chart. Who is accountable for the journey and the initiative? Who keeps the customer ID clean in all the systems involved? Who reports the result to management, and how often? If these roles stay open, the connection falls apart again as soon as the next tool arrives or an export stops running.
You can build this while the business keeps running, one journey at a time. A dashboard can display the result afterwards.
What you decide differently on Monday
The consequence is a budget rule: only release new journey or CX budgets if it is clear before the start how the benefit will be measured and which data needs to be connected for that.
An approval request then answers the five decisions from the previous section before any money flows. If it cannot, that does not automatically mean no. It means measurability gets established first and the initiative is funded afterwards. That is often the smaller step.
Operating system instead of advisory
This is where UNIT ONE comes in. We don't produce recommendation slides that end up filed away after the project. We support you in building the architecture that makes proof possible: the connection between initiative and business result across your existing systems, the shared customer ID, the baseline measurement, the responsibilities and the reporting cycle. The goal is an operating system for your customer work that your team carries on by itself.
Nobody can seriously promise you in advance whether a particular journey initiative will work. The question is whether you will know afterwards. Would your next budget approval pass this test? If you would like to check that together: enquire about the Reactivation Audit.
Frequently asked questions
Why is the ROI of journey management so hard to prove? In a CX Network survey from 2024, 23% of respondents said they struggle to demonstrate any financial return at all; the most frequently named obstacle was isolated data silos. Without a connection between customer data, operational metrics and financial figures, a customer's path from initiative to revenue cannot be traced.
Is a better CX dashboard enough to prove the benefit? As a rule, no. If the initiative and the revenue sit in different systems with no shared reference to the same customer, no interface can create the connection.
What are data silos, and how do I recognise them? A data silo is a system whose data is not connected to the others. You can recognise silos when reports are built from manual exports, when figures differ from department to department and when nobody can trace a single customer without gaps from the first campaign to contract renewal.
Do I need a new platform or a BI team? Not to get started. It is enough to connect the two or three systems of one specific journey through a consistent customer number, plus a baseline measurement and a comparison.
Which metric is suitable as a first proof of ROI? One that translates directly into euros: cancellations, repeat purchases or service costs per customer. Satisfaction scores alone are rarely enough for management, because the step to revenue is missing.
Who should be responsible for proving ROI? Management, because whoever releases budget has to be able to justify the benefit. Operationally, each journey needs a named person for measurement and results, and clear responsibility for the customer data.