Every group of trading companies has a version of the same weekly ritual. An enquiry lands with one brand, the rep reading it can see within twenty seconds that the work belongs to a sister company, and then the lead begins its journey through a Teams message, a forwarded email, a shared spreadsheet tab and a name half-remembered from a Christmas party. The commercial logic of owning several complementary businesses is that a customer of one should become a customer of two. The operational reality is that when groups try to share leads between group companies, the value evaporates somewhere between the forward button and Tuesday.
This piece is written for the sales director or CRO who lives with that friction weekly, and who usually ends up making the internal case to the FD that something has to change. The FD's question will be what the leak is worth. That number is more calculable than most groups assume.
Executive Summary
Cross-brand lead sharing fails at the handover rather than at the point of goodwill, and the cause is architectural: most CRMs are built around one company, one customer base and one permission boundary, so access between entities becomes all-or-nothing. Groups therefore route referrals through email and spreadsheets, where speed dies. Harvard Business Review's audit of more than 2,000 companies found the average response to a web-generated lead took 42 hours, with 23% never responding at all, and an internal referral inherits every one of those delays plus a re-keying step. McKinsey's cross-selling research shows the compounding effect on the other side: cross-sell lands fastest where the seller already has a relationship, and accounts requiring a cold introduction took roughly 18 months longer to reach the same result. The fix is record-level, entity-aware sharing that passes a single live record between brands with context and consent, which under the ICO's data sharing code is also the shape compliance wants.
Why do leads between group companies get stuck in email?
The honest answer is that the CRM will not let anyone do anything else. Mainstream platforms model a business as one organisation with one set of customers, one set of owners and one perimeter. Where a group needs a rep in Company A to hand a live record to a rep in Company B while keeping Company A's other 4,000 accounts invisible, the system typically offers two settings: see everything, or see nothing. In practice that leaves administrators choosing between a group where every brand can browse every rival brand's pipeline, and a group where nobody can see anything, which is why the referral leaves the system entirely and travels by email instead.
Once a lead leaves the system, it loses the three things that made it valuable: a timestamp, an owner and a next action. Salesforce's State of Sales research reports that sales reps spend around 60% of their time on non-selling tasks such as data entry, lead research and switching between tools. A cross-brand referral handled by hand is precisely that kind of task, performed twice, once by the rep giving the lead away and once by the rep re-typing it into a different instance.
Speed is what gets spent. Harvard Business Review's study The Short Life of Online Sales Leads audited how long more than 2,000 US companies took to reply to a web-generated enquiry, and the distribution, as summarised by Salesforce, was not flattering: 37% answered within an hour, 24% took longer than a day, and 23% never answered. The average among those who did respond within 30 days was 42 hours.
Figure
How long companies take to respond to an inbound lead
- Within one hour37%
- One to 24 hours16%
- More than 24 hours24%
- Never responded23%
Audit of more than 2,000 US companies responding to a web-generated test lead; distribution as summarised by Salesforce. A cross-entity referral adds a manual handover on top of these delays.
Source: Harvard Business Review, The Short Life of Online Sales Leads (2011)
Those figures describe a company answering its own enquiries. Add an entity boundary, a second CRM login and a colleague who has no target attached to the outcome, and 42 hours starts to look optimistic.
What is a stalled cross-brand referral actually worth?
Take an illustrative group of three trading companies: a commercial cleaning business, an M&E maintenance firm and a small security installer, all selling to facilities managers at the same kind of site. Say the cleaning company fields 60 inbound enquiries a month and one in six is a better fit for maintenance, so ten referrals a month should cross the boundary. If maintenance closes referrals at 20% and its average first-year contract is £18,000, those ten a month represent roughly £432,000 of annual order value in flight.
Now apply the handover. Half never get logged by the receiving brand because they arrive as an email with no owner, and of those that do, some are contacted a week late against a competitor who replied the same afternoon. A conservative reading puts well over £150,000 of that year's order value on the floor, and no report anywhere in the group will show it, because the leak happens in a medium the CRM cannot see. That invisibility is the reason this problem survives board scrutiny for years: nobody can produce the number, so the FD treats it as anecdote.
A referral that lives in an inbox has no owner, no timestamp and no place in anyone's forecast, which makes it indistinguishable from a lead you never received.
McKinsey's work on revenue synergies suggests the shortfall is systemic rather than local. In its M&A polling, fewer than one in five organisations achieved their cross-selling goals, largely because companies underestimate the operational complexity of taking one entity's products to another entity's customers. The corollary matters more for groups: McKinsey's six c's of cross-selling success found one technology services organisation hit an 80% cross-selling rate within a year at accounts where sellers already had strong relationships with the relevant decision-makers, while accounts where reps had to sell to unfamiliar people took about 18 months longer to reach comparable results.
Figure
Relationship strength decides how fast a cross-sell lands
80%
Cross-sell rate within a year
At accounts where sellers already had strong relationships with the relevant decision-makers
+18 months
Extra time to comparable results
At accounts where reps had to sell to people they did not know
Figures from one technology services organisation post-merger, cited by McKinsey. Separate McKinsey M&A polling found fewer than one in five organisations achieved their cross-selling goals.
Read that as an argument about handovers. The relationship is the asset being transferred, and a forwarded email transfers a name while leaving the relationship behind. A referral that arrives with the enquiry history, the site details, the original conversation and a warm introduction from the colleague the customer already trusts is a different commercial object entirely.
How do groups share leads between group companies today?
Four patterns cover almost every group, and they trade off the same two variables: how fast the lead moves, and how much of the group's data the receiving brand can see.
| Approach | How it works | What breaks | Who absorbs the cost |
|---|---|---|---|
| Email or chat referral | Rep forwards the enquiry to a named contact in the sibling company | No owner, no timestamp, no audit trail, invisible to reporting | The receiving brand's pipeline, silently |
| Shared spreadsheet | Referrals logged centrally, picked up periodically | Latency measured in days; duplicate and stale records | Sales ops, then the customer |
| Separate CRM per entity | Each brand runs its own instance, leads re-keyed on arrival | Double entry, no group view of the same customer | Reps, in admin time |
| One CRM, all-or-nothing access | Single instance, brands see everything or nothing | Protected accounts exposed, or sharing blocked outright | Whoever loses the internal argument |
| Record-level cross-entity sharing | A single live record shared to named users in another entity, with context and an expiry | Requires permissions designed for this in the foundations | The vendor, if it built for groups |
In practice the fifth row is where groups end up trying to get to by force. Elaborate automation is built to fake it: rules that clone records between business units, sync jobs that reconcile the clones, alerts that tell a manager the clone failed. Building those workflows is genuinely satisfying and they demo beautifully. They are also a half measure, and they get unpicked later, because the underlying platform still believes there is one company in the system and every rule is arguing with that assumption. Fine-grained visibility control, within a brand and between brands, has to sit in the foundations. Bolted on, it becomes maintenance work with a monthly bill attached.
Is sharing leads between group companies allowed under UK data protection law?
Yes, and the way the ICO frames it is instructive. Under the ICO's data sharing code of practice, disclosure of personal data between separate controllers counts as data sharing, and that includes organisations within the same group or under the same parent company. Two companies in one group are usually separate controllers with their own purposes, so passing a contact from one to the other is a sharing event, not an internal move.
The ICO's guidance on data sharing agreements asks controllers to be specific about which data is shared and for what purpose, and notes that it may be appropriate to attach permissions to particular data items so only staff in defined roles can access them. That is the same primitive the commercial argument demands. A system that can share one record, with named fields, to named people, for a named purpose satisfies the sales director and the DPO with a single mechanism. An all-or-nothing switch satisfies neither.
Why hasn't a mainstream CRM solved this?
Because the maths of a product roadmap is unsentimental. A platform sold to every business on earth optimises for the widest possible market, and steering towards a wide audience means fitting no specific audience precisely. Multi-entity groups are a smaller addressable market with a longer, more consultative buying process involving an FD, several managing directors and a group IT function. Features that only groups need lose the prioritisation argument to features every single-company customer will use, every quarter, forever. Nobody is being lazy. The incentives are simply pointing elsewhere, which is also why the gap is usually closed by implementation hours rather than by product.
That structural point is worth holding onto when a vendor demonstrates cross-entity sharing to you. Ask whether entities exist as first-class objects with their own permission boundaries, or whether an entity is a custom field that clever configuration treats as one. The first survives a reorganisation and an acquisition. The second survives until the consultant who built it moves on.
Groups thinking about this properly should also look at the two adjacent problems, since the same architecture decides all three: CRM permissions across group companies, which governs who can see a shared account once it has been shared, and group reporting, which is where the FD finds out whether cross-brand referrals are actually converting. The wider argument for building entity awareness into the foundations sits in our multi-entity CRM pillar, and the bill that arrives when reporting is commissioned instead of built is set out in the cost of custom CRM reports.
What does a good cross-entity handover look like?
The test is not whether the lead can be sent. It is whether anything meaningful survives the journey. A referral that works carries four things: the record itself rather than a copy, so both brands see one version of the truth; enough context that the receiving rep can open with the customer's own words rather than a cold call; a named owner in the receiving entity with an accountable response time; and a visibility boundary that exposes this account and nothing else. Add commission clarity and the behaviour changes on its own, because the rep giving the lead away stops treating generosity as a personal cost.
Most groups already know all of this. What they lack is a system where the good version is the easy version, which is why the spreadsheet keeps winning.
Frequently Asked Questions
Can companies in the same group share leads under UK GDPR?
Yes, provided the sharing has a lawful basis, is covered by the privacy information given to the individual, and is documented. The ICO's data sharing code treats disclosure between organisations in the same group or under the same parent company as data sharing between controllers, so a group cannot assume that common ownership makes it an internal transfer.
Should each company in a group run its own CRM?
Separate instances give each brand clean boundaries and no group view, which means every shared customer is reconciled by hand and every consolidated report is a data project. Groups that need to trade with each other's customers generally find that one system with entity-aware permissions costs less over time than several systems plus the manual work between them.
How do you stop reps in one brand poaching another brand's accounts?
With record-level visibility rather than blanket access. Where a platform only offers full access or none, groups tend to default to none and route sharing through email, which reintroduces the delay. Fine-grained control lets a specific account be shared with specific people while the rest of the entity's book stays invisible.
Who should get the commission when a lead crosses between group companies?
There is no single standard, and typical arrangements range from a fixed referral fee to a split of first-year margin between the originating and closing entity. McKinsey's cross-selling research found that almost three-quarters of the M&A executives it polled considered incentives important or critically important to cross-selling success, and that an incentive too small to matter signals that cross-selling is optional.
How quickly should a shared lead be contacted?
Faster than the group's current handover allows. Harvard Business Review's audit of more than 2,000 companies found an average response time of 42 hours to a web-generated lead and 23% of companies never responding at all, and internal referrals add a manual step on top of that baseline. Attaching a named owner and a response expectation at the moment of sharing is what removes the drift.