A group of trading entities eventually discovers that it has two account managers calling the same procurement director. One sells cleaning, the other sells mechanical and electrical maintenance, and the customer thinks of both of them as the same supplier because the invoices arrive under the same parent brand. When the same customer buys from multiple subsidiaries, the CRM is usually the last place anyone finds out. It has two records, two owners, two pipelines, and no concept that the relationship behind them is one thing.
Executive Summary
When the same customer is served by multiple subsidiaries, a CRM built for a single trading company will model the relationship as two unrelated accounts, because its data model assumes one account, one owner, one legal entity. The commercial damage is not the duplicate record itself; it shows up as uncoordinated pricing at renewal, cross-sell that never happens, and a group that cannot answer what a customer is worth in total. Gartner research found that only 28% of sales leaders say their account management channels regularly hit cross-selling objectives, and that 49% of account managers are measured on a single revenue goal that makes no distinction between retention and growth. Fixing this needs three things in the foundations: a group-level account identity with entity-level children, visibility that is partial rather than all-or-nothing, and a relationship owner who is distinct from the revenue owner. The system can hold the structure; only the group's incentive design makes anyone use it.
Why does the same customer show up twice when multiple subsidiaries use one CRM?
Almost every mainstream CRM treats an account as a single commercial object with a single owner. That is a reasonable default for the overwhelming majority of buyers, who run one trading company and want to know which rep is responsible for which logo. Duplicate detection is built on the same assumption: two records with matching company details are an error to be merged, not a legitimate statement that two of your businesses hold two separate contracts with the same customer under different terms, different payment cycles and different account managers.
The mechanism behind this is economic rather than technical. Platforms are designed to be generic and broadly applicable, because the widest possible market is where the licence revenue sits. Steering for a wide audience means fitting no audience exactly, and groups of companies lose out more than most: they are a smaller addressable market with a longer, more drawn-out buying process, so the primitives they need never justify the roadmap space. What arrives instead is a bolt-on, usually a custom object and a pile of automation rules written by an implementation partner, sitting on top of a model that still believes an account belongs to one company. We have written separately about why generic CRMs fail multi-company groups and about how one CRM or one per company changes the decision.
The customer, meanwhile, has already consolidated. Forrester's 2026 State of Business Buying research found that the typical buying decision now involves 13 internal stakeholders and nine external influencers, with procurement acting as a decision-maker in 53% of cycles. A procurement function that sophisticated will notice that three of its suppliers share a registered address long before your group's CRM does.
Figure
Account growth is where sales organisations consistently underperform
28%
of sales leaders say account management regularly hits cross-selling objectives
Even inside a single company, cross-sell is the weakest motion in the sales organisation.
49%
of account managers are measured on one undifferentiated revenue goal
No separation between retaining the account and growing it.
2.2
alternative suppliers considered at every renewal, on top of the incumbent
The renewal conversation is competitive by default, not a formality.
Survey of nearly 700 account managers and sales leaders.
Source: Gartner, Position Sales Teams to Grow Accounts, Not Just Retain Them
What does a shared account across two trading entities actually cost?
Take an illustrative group: a commercial cleaning business turning over £6m, an M&E maintenance business on £4m, and a small security installer on £2m, all three sold into the same regional supermarket chain. Say cleaning bills £340,000 a year, maintenance £120,000, and security £45,000. Nobody in the group has ever seen £505,000 on one line, because the reporting rolls up by entity and the customer name is spelled three different ways.
At renewal, the chain's procurement team runs a consolidated tender covering all three services and asks for a group rate. Each entity prices defensively, with no sight of what the others are charging or how much total margin is on the table. The group ends up discounting three times to defend three contracts it could have defended once, and it does so without knowing whether the customer sits in its top ten or its top fifty. McKinsey's work on large accounts found that key accounts can represent 30 to 50 percent of revenue and margin for many B2B companies, which is exactly the concentration a group cannot see when its largest relationship is split across three sets of books.
The cross-sell side is worse, because it is invisible. The cleaning account manager hears that the chain is refitting twelve stores. Passing that to the maintenance entity costs him a week of relationship capital, introduces a sister company he cannot control into his own renewal, and earns him nothing, since his number is cleaning revenue. He keeps quiet, rationally. KPMG, writing in Harvard Business Review, noted that acquiring a new customer can cost as much as 25 times more than growing revenue with an existing one, which makes a suppressed introduction between two of your own brands one of the most expensive silences in the group.
A group that cannot total up one customer across its entities is negotiating against a buyer who already has.
Should a shared customer be one CRM record or one per subsidiary?
Groups generally try one of three structures. Each has a failure mode, and the choice determines what reporting and permissions can do later.
| Approach | What it looks like | Where it breaks |
|---|---|---|
| Separate record per entity | Each trading company keeps its own account, contacts and pipeline for the customer | No total customer value, no coordinated renewal, duplicate contacts receiving contradictory outreach |
| One merged record | A single account owned by whichever entity sells the most | The smaller entities lose their own history and terms; one entity's team gains visibility of contracts it has no right to see |
| Parent-child with scoped visibility | A group-level customer identity with an entity-level child record under each trading company | Needs record-level permissions and group roll-up in the foundations; bolted on, it degrades into either of the two above |
The third model is the one groups actually want, and it is the one most systems cannot hold without heavy customisation. It requires that visibility be configurable at a fine-grained level both within a brand and between brands. Most products manage this within a single brand and do it poorly; between brands it is close to impossible, so what the group is offered is an all-or-nothing setting where a user can either see everything across the group or nothing outside their own entity. Neither answer is the one a shared account needs, which is that the maintenance account manager should see that cleaning holds a contract, see who owns it, see the renewal date, and see nothing of the commercial terms. We go further into that calibration in our piece on CRM permissions across group companies.
Figure
What changes when the account spans more than one trading entity
| Dimension | Single trading company | Group of trading entities |
|---|---|---|
| Account owner | One rep, unambiguous | A revenue owner per entity, plus a relationship owner for the group |
| A duplicate record means | A data quality error to merge | Two legitimate contracts under different terms |
| Shared account plan | Optional, usually one document | Mandatory, and must be visible to users in different entities |
| Visibility needed | Team or territory level | Record level, partial, different per field |
| Incentive design | One quota, one P&L | Credit must land in two P&Ls without double-counting at group level |
| Customer's view | One supplier | One supplier, whatever the group's structure says |
McKinsey's prescription for large accounts is a single account leader, a shared account plan, transparency across business units, and incentives aligned to the account rather than the unit.
Source: McKinsey, How to Unlock Growth in the Largest Accounts
Who should own a customer that two group companies sell to?
Separating two roles solves most of the argument. The revenue owner is the account manager inside each trading entity, accountable for that entity's contract, margin and renewal. The relationship owner is one named person accountable for the group's total position with that customer, including what the group should stop selling them. McKinsey's prescription for large accounts is close to this: a clear account leader running a cross-functional team with representatives in each of the customer's business units, a shared account plan with named owners, transparency on performance across units, and incentives designed for the good of the account rather than the unit.
Groups usually get the first three and skip the fourth, which is why the structure collapses. If the relationship owner is also the managing director of the largest entity, every judgement call about which brand leads a tender will be read, often correctly, as self-interested. Harvard Business Review reported in December 2025 that research from the Strategic Account Management Association and ZS found the share of firms treating strategic account management as a clear strategic priority fell from 65% in 2020 to 33% in 2025, with lack of executive support the leading reason programmes stall. A group-level account owner without a group-level sponsor is a job title, not a mechanism. We have treated the attribution question itself at more length in who owns the customer when two group companies sell to the same buyer.
What can a CRM genuinely fix here, and what can it not?
What the system can do is hold the truth: one customer identity at group level, child records per trading entity, a relationship owner who is not the same field as the record owner, visibility set field by field rather than entity by entity, and revenue that rolls up to a group total without being counted twice. Atrium is built around those primitives rather than offering them as configuration, which is the difference between a group seeing its true top ten customers on a Monday morning and commissioning a report about it. The related problem of what custom CRM reports cost is the same architecture failing one step downstream.
What no system fixes is the comp plan. If the cleaning account manager's bonus is cleaning revenue and nothing else, a perfectly modelled shared account will sit in the CRM unread. McKinsey's work on digital key account management describes the gain as institutionalising knowledge that previously lived only in reps' heads, which only happens when the reps have a reason to put it there. Equally, a group should be honest that some accounts should not be shared at all. A customer buying a commodity service from one entity on thin margin is not a strategic relationship because three brands happen to touch it, and forcing a group account structure onto it adds coordination cost for no return. The test is whether the customer's own procurement treats you as one supplier. If they do, the group has to be able to answer as one.
Frequently Asked Questions
Should each subsidiary keep its own CRM record for a shared customer?
Yes, with a parent record above them. Each trading entity needs its own account record to hold its contract, terms, contacts and pipeline, because those genuinely differ between entities. A group-level customer identity sitting above those records is what makes total customer value, coordinated renewals and cross-entity reporting possible without merging commercially sensitive data into one view.
Who should own an account that spans two group companies?
Typical practice in group structures is to separate the revenue owner, who is the account manager inside each trading entity, from a single relationship owner accountable for the group's whole position with that customer. McKinsey's research on large accounts recommends a clear account leader supported by a cross-functional team, a shared account plan with named owners, and incentives aligned to the account rather than the business unit.
How is revenue credit split between two subsidiaries selling to the same customer?
Credit normally has to land twice at entity level, once in each subsidiary's P&L against its own contract, while being counted only once at group level to avoid inflating consolidated revenue. Gartner research found that 49% of account managers are measured on a single revenue goal with no distinction between retention and growth, which is the underlying reason cross-entity introductions are under-rewarded.
Can standard CRM permissions restrict access between group companies?
Most mainstream CRMs manage record-level visibility inside a single brand and offer coarse, often all-or-nothing controls between brands, so a user either sees the whole group or only their own entity. Granting a user in one subsidiary read access to the existence, owner and renewal date of a sister company's contract while hiding its commercial terms usually requires either custom development or a platform with cross-entity permissions in its foundations.
Does a shared group account improve the chance of winning the renewal?
Gartner research found that customers consider an average of 2.2 alternative suppliers in addition to the incumbent at every renewal or repurchase decision, so coordination across entities mainly reduces the risk of being outflanked on price by a buyer who already sees the group as one supplier. McKinsey reports that managing the full customer experience across the account can improve key-account revenue by 5 to 10 percent or more while holding margin.