Outbound for portfolio companies, run per company
GrowthPal is a B2B outbound and GTM engineering agency operated by Inferno Media, LLC, working from Kathmandu, Nepal.
Last reviewed 29 October 2026
We run outbound for portfolio companies: campaigns that help a PE-backed company sell its own product to its own customers. We do not source deal flow, find acquisition targets or contact company owners on behalf of a fund. This page covers the four questions a portfolio program raises that a single-company program does not, and links out for everything else.
What is in scope, and what is firmly not
Who this is for
- You are an operating partner with several holdings. You want one supplier, comparable reporting and less setup cost per company, and you want to know honestly how much of that is real.
- You run a PE-backed company and the plan says grow pipeline. The questions that slow this down are usually who signs, who pays and whose name is on the sending domain, not whether the channel works.
- You want the comparison problem solved before it starts. If eight companies report a reply rate against eight different denominators, the portfolio view is noise. Fixing definitions costs nothing at the start and is close to impossible later.
- You want to be told when not to do this. Several named cases below are ones where outbound is the wrong instrument for a PE-backed business, including the two that cost us the sale.
Who this is not for
- Deal flow and origination. Not at any price. We do not run outbound to source acquisition targets, reach business owners or intermediaries, or build proprietary deal pipeline for a fund. That is a different trade and we are not in it.
- Fundraising and investor relations. No LP outreach, no placement support, no investor communications. Same answer, same reason.
- You are looking for the other company with this name. A separate and unrelated business also called GrowthPal works in M&A deal sourcing. This site belongs to Inferno Media, LLC, working from Kathmandu, Nepal, and sells B2B outbound. If you came here for deal origination, we are not it.
- You want one message rolled out across the portfolio. That is the thing most suppliers sell here and it is mostly false. The message, the buyer, the list and the offer do not transfer, and the section below explains exactly why.
What a portfolio engagement adds over a single-company one
| Component | What is done | What the client receives | Who owns it at the end |
|---|---|---|---|
| Contracting structure | Decide which of three shapes applies before anything is signed, since the choice decides who is accountable for the data and whose name a recipient sees. | A written recommendation with the data protection consequence of each shape spelled out. | Client |
| Shared definitions | Impose one set of counting rules across every holding: what a contacted account is, what a reply is, what a positive reply is, what a meeting held is, what a qualified opportunity is. | A definitions document each company signs up to before its first send. | Client |
| Per company market sizing | Count the addressable accounts for each holding separately, with the filter logic recorded, because this is the denominator every other number should be read against. | An account count and filter logic per company. | Client |
| Separate sending identity per company | Each holding gets its own domains, its own authentication records and its own reputation. Nothing is consolidated across the portfolio. | A sending estate per company, registered in that company's name. | Client |
| Portfolio reporting format | One report shape, one order of numbers, one definitions footnote and one known limitations footnote, so eight reports can be read in an hour instead of a day. | A standard report, per company, never blended. | Client |
| Market consumption tracking | Report what share of each company's addressable market has been contacted, and how many months of market remain at the current rate. | The number that predicts which holdings stall and when. | Client |
| Cross company suppression | Where two holdings work the same accounts, suppress across them so one buyer does not receive sequences from two of your companies in the same month. | A suppression rule that spans the portfolio, with the list contents staying per company. | Client |
| Per company message work | Build the message, the buyer definition, the list and the offer from scratch at every holding, because none of those transfer. | A separate program per company, sharing only the method. | Client |
How a portfolio engagement starts
Settle the contracting shape first
Who signs, who pays and who decides the targeting are three separate questions and they do not have to have the same answer. Getting this wrong at the start is expensive to unwind, because it is settled by conduct rather than by what the paperwork says.
What you provide. A view from the fund's counsel on which shape fits your structure.
Agree the definitions once, for everyone
The definitions are arbitrary, which is exactly why they can be imposed identically. Imposing them identically is the only thing that makes cross-company comparison possible at all.
What you provide. Sign off from each management team, not just from the fund.
Size each market separately
Addressable accounts per holding, with the filter logic written down. A company with 2,000 addressable accounts and one with 400,000 are running different programs and should never be judged on the same number.
What you provide. Access to each company's existing customer data and its own disqualification criteria.
Stand up a separate sending estate per company
Separate domains, separate authentication, separate reputation. The published spam rate ceilings apply per sender, so a consolidated identity means one holding's bad list damages every other holding that shares it.
What you provide. A registrar account and a DNS owner at each company.
Start with one company, not all of them
The first holding is where the method gets calibrated to your reporting and your sponsor's expectations. Running eight launches at once means eight versions of the same avoidable mistake.
What you provide. A decision on which holding goes first, which should be the one with the clearest buyer rather than the biggest number.
Rebuild the message at every company
Process, definitions, infrastructure and suppression carry over. The message, the buyer, the list and the offer do not. The per company work is most of the work and it does not shrink with the second holding.
What you provide. Time from each management team. This is the part the fund cannot do on their behalf.
Report per company, never blended
Each holding gets the same report shape with its own numbers. The portfolio view is a set of comparable per company numbers read side by side, plus the handful of facts that genuinely compare.
What you provide. A standing review with the operating partner and separate ones with each management team.
Watch consumption, not just output
The number that matters most over a hold period is how much of each market has been used up. It is the one that says which company stalls next quarter, and almost nobody tracks it.
What you provide. Agreement that market consumption appears in the board pack.
The four questions a portfolio program raises
Everything else about running outbound is covered elsewhere on this site and linked at the end. Outbound for portfolio companies raises exactly four questions a single-company program does not, and none of the pages competing for this work answers any of them. Those four are the whole of this section.
One. Who contracts, who pays, and who is the controller
The contracting question is not administrative. It decides who can be fined, whose name appears in the privacy notice the recipient reads, who can lawfully instruct us to change the targeting, and what happens to the data when a company is sold. Two rules set the frame. Where two or more parties jointly determine the purposes and means of processing, they are joint controllers, they have to set out their respective responsibilities in a transparent arrangement, the essence of that arrangement has to be made available to the people whose data it is, and a recipient can exercise their rights against either party whatever the arrangement says between them. Separately, a processor that starts determining purposes and means is treated as a controller for that processing. The label in the contract does not decide the role. The conduct does.
| Fund contracts centrally | Portfolio company contracts directly | Framework plus per company statements of work | |
|---|---|---|---|
| Who signs | The fund or management company signs one master agreement covering several holdings | Each portfolio company signs its own agreement | The fund signs a framework for commercial, service and data protection terms; each company signs a short statement of work under it |
| Who decides who gets emailed | Risk: the fund does, or appears to. Setting the target list, the segments or the offer is determining purposes and means | The portfolio company does, unambiguously | The portfolio company does, by design. The framework deliberately contains no targeting decisions |
| Likely data protection role | The fund is at real risk of being a controller or joint controller alongside the company, with the supplier as processor to both. If the fund only pays and never directs it may be neither, but that is decided by conduct | Clean. The company is controller, the supplier is processor, the fund sits outside the processing entirely | Clean if the framework is genuinely commercial and the statement of work genuinely carries the targeting. Contaminated the moment the framework specifies buyers, lists or messaging |
| What the recipient sees | If joint controllership attaches, the essence of the arrangement has to be made available, so the fund's involvement becomes visible to every recipient | Only the portfolio company | Only the portfolio company, provided the framework stays out of targeting |
| Advantage | Fastest to stand up. One negotiation, one security review, one assessment, consistent terms | Cleanest. No fund-level exposure, survives the sale of any one holding without unpicking anything, and the management team owns what it is accountable for | Most of the speed of the first with most of the cleanliness of the second. Any company can start or stop without affecting the others |
| Disadvantage | Highest exposure. A recipient can exercise rights against the fund directly, and a failure at one holding has a path to the fund's name. It also creates the appearance that the fund operates the business | Slowest. Every company runs its own negotiation and its own agreement, terms drift, definitions drift, and a management team can quietly decline | Needs discipline. The framework will be tempting to fill with targeting guidance, and the moment it is, it collapses into the first column |
| When it fits | Rarely. Only where the holdings are a genuine single operating group already sharing data flows | The default for a fund with few holdings, holdings in different sectors, or any holding with its own compliance function | The default for a fund running one supplier across four or more holdings and wanting comparable reporting |
- The sending identity is always the portfolio company's, never the fund's. The published spam rate ceiling is a per sender ceiling, so a consolidated identity means consolidated fate, and one holding's bad list damages every other company sharing it.
- A fund signing centrally is authorizing our sub-processors on behalf of every holding at once. In outbound that means the sending infrastructure, the data providers and the enrichment providers. Most operating partners have not considered that.
- If the fund wants comparable assurance across holdings, the audit right has to sit in each company's contract and has to be exercisable by someone the fund can rely on.
- Settle the exit question at signature, not at the exit. Deletion or return at the controller's choice applies to the suppression list, the reply corpus, the sending domains and the mailbox estate. Decide which of those follow the company out the door.
- Who pays and who signs can differ, and often should. A fund can subsidize the first engagement while the company signs. What a fund should not do is sign, pay and direct, because that combination is what makes it a controller in fact.
Two. What transfers between holdings, and what does not
Every supplier in this market sells transferability, because transferability is what makes a portfolio-wide engagement look efficient. The claim is mostly false, and being specific about which parts are false is the most useful thing we can tell you.
| Transfers across the portfolio | Does not transfer, and has to be rebuilt at every company |
|---|---|
| Process. The order of operations: deliverability before volume, market sizing before sequence design, a written test log before any claim about what works. This is a property of the channel, not of the business. | The message. A claim about a specific problem a specific buyer has and what it costs them. Two holdings in the same sector are selling different things to people with different problems. |
| Measurement definitions. What counts as a contacted account, a reply, a positive reply, a meeting held, a qualified opportunity. Arbitrary, therefore impose them identically. | The buyer. Job title overlap is not buyer overlap. The same title at two companies is not persuaded by the same evidence and does not read the same way. |
| Infrastructure patterns. Domains per company, how they relate to the primary domain, what authentication policy to publish and when to tighten it, how it is monitored. | The list. A function of the message and the buyer, so it inherits their non-transferability. Even against nominally the same accounts, the right contact inside them differs. |
| Suppression discipline. The rules for who is excluded and why. The mechanism is portable; the contents are not. | The offer. What the recipient is asked to do and what they get for it, which is set by the sales cycle, the price point, the proof assets and the capacity to deliver. |
| The reporting format. One shape, one order, one definitions footnote. This is what lets an operating partner read eight reports in an hour. | Trigger and timing logic. The events that make an account worth contacting are industry specific and often company specific. |
| Negative knowledge. The failure modes already seen: claims that get a domain filtered, lists that look big and are not, replies that look positive and are not. The most underrated transferable asset. | Capacity to receive. A program that books meetings a company cannot staff destroys value. Capacity is a per company fact and a precondition rather than an output. |
The honest summary: a supplier carries the method, the measurement, the infrastructure and the scar tissue across a portfolio. It rebuilds the message, the buyer definition, the list and the offer at every single company. The efficiency gain is real and it is entirely in the first column. It is a reduction in setup time and coordination cost, not a reduction in the per company work of figuring out what to say and to whom. A corollary worth stating against our own interest: we already ran this for another company in your portfolio is a weak proof point. The useful version of that sentence is that we already know how your sponsor wants this reported and we already passed your sister company's review, which is about friction, not about results.
Three. Why a blended cross-portfolio reply rate is not a metric
An operating partner wants one view across eight companies, and the instinct is to average. Averaging is where the number stops meaning anything. A reply rate is replies divided by something, and that something is almost never the same thing twice.
- Different denominators. One company reports replies against emails sent, counting every message in a five-step sequence. Another reports replies against contacts entered into sequence, counting each person once. For identical underlying performance the second number is roughly five times the first, and the average describes neither.
- Different market sizes. A company selling into 2,000 addressable accounts should post a far higher rate than one selling into 400,000, because it sends fewer messages at higher precision. Weight the blend by volume and the large market dominates it completely. Weight it equally and you have treated the two markets as equally informative, which they are not.
- Different reply definitions. Does an out of office count? Does a routing reply naming someone else count, or is that a different event? Does a negative reply count? Three companies will answer three ways and none will have written the answer down.
- Different program stages. A company eight weeks into warm-up is sending a fraction of steady-state volume to its best known accounts. One fourteen months in is working colder segments. Neither number is comparable to the other, or to its own number from six months ago.
- Different cycle lengths. A reply in a three-week cycle and a reply in a nine-month cycle are not the same asset. A portfolio number that treats them as interchangeable will systematically favor the short-cycle business and starve the long-cycle one of attention.
- Survivorship inside the denominator. If a company pauses its worst segment, its reply rate rises without anything improving, and the blended number rises with it. That will be read as progress.
A blended cross-portfolio reply rate is not a weak metric. It is not a metric. It is an average of quantities with different denominators, different definitions and different market sizes, and the thing it most reliably measures is how many of the holdings happen to be in warm-up this quarter. What works instead is a small set of per company numbers against shared definitions, read side by side: addressable market size with the filter logic that produced it, accounts contacted as a share of that market, positive replies per hundred accounts contacted, meetings held rather than booked, pipeline in the company's own definition, time to meeting held as a distribution, and suppression volume. Only three facts genuinely compare at portfolio level: authentication and deliverability health per sending domain, because the mailbox provider defines those the same way everywhere; program stage as a label rather than a number, since most apparent performance differences are stage differences; and months of addressable market remaining at the current rate.
Four. When in the hold to start, and when not to start at all
The consensus answer is the first 100 days, which is directionally right for the wrong reason. Outbound does not produce results in 100 days. What happens in the first 100 days is that the things outbound depends on get decided: domain and subdomain strategy, CRM structure, how an opportunity is defined, who owns routing, what the sales team is paid on. Set those without outbound in mind and the program that starts in month nine is building on foundations laid against it. The correct action in the first 100 days is usually not to start sending. It is to make the decisions that make sending possible later, which costs very little.
| Stage of the hold | What is true | What belongs here |
|---|---|---|
| Diligence | No access, no authority, no data | Nothing operational. At most, a view on whether the market is large enough to support outbound at all, because that affects the commercial plan |
| First 100 days | Management is newly accountable, systems are being touched, budget exists | Decide the domain and sending architecture. Fix the CRM definitions. Establish what currently produces pipeline. Do not launch a program into an unstable organization |
| Months 4 to 12 | The operating plan is real, the team is known, the offer is understood | The honest start window. Deliverability, then one segment, then a written test log, then volume. Expect the first meaningful pipeline read at the far end of this window, not inside it |
| Years 2 to 3 | The program either has a working segment or it does not | Expand into adjacent segments. This is where market consumption starts to bind. If no segment has worked by now, stop and diagnose rather than adding volume |
| Last 12 to 18 months | Every number will be read by a buyer and their diligence provider | A program started here will not produce a credible track record in time. What is worth doing is making the existing one legible: documented process, clean definitions, a provable attribution story. A buyer pays for a repeatable motion, not a recent spike |
- The addressable market is too small. If the realistic count of accounts that could buy is in the low thousands or fewer, a program will consume the market before the message is right. Named-account selling by a human is the correct instrument.
- There is nobody to take the meetings. Capacity is a precondition, not an output, and booking into an organization that cannot run them converts interest into damage.
- The company does not know who buys. If the last twenty wins came from twenty directions with no pattern, outbound amplifies the confusion. Fix that against the existing customer base first.
- The product cannot be described in two sentences to a stranger. That is usually positioning and sometimes product, and outbound solves neither.
- Retention is the actual constraint. If the company loses customers faster than it wins them, acquisition spend makes the numbers worse. This is the most common misdiagnosis in a PE-backed business under revenue pressure, because acquisition is visible and churn is not.
- The sponsor wants a number by a specific board date. Outbound does not respond to deadlines. Compressed to hit a date, it will be run at volumes its deliverability cannot support, and the channel will get the blame.
- The company is mid-integration after a merger or carve-out. Domains, brand, CRM and territory are all in flux, and anything built now will be rebuilt.
Stating the retention case and the board date case on a commercial page costs us business. They are on the page because an operating partner who has seen either one happen will recognize that we have too.
What to ask before buying outbound for portfolio companies
A portfolio appointment multiplies whatever the supplier gets wrong. These six questions surface it before four management teams have signed.
| What to check | The question to ask | What a good answer looks like |
|---|---|---|
| Transferability | What exactly carries over from one of our companies to the next? | Method, definitions, infrastructure and known failure modes. If the answer includes the message, the buyer or the list, they are describing something they cannot deliver. |
| Contracting | Who should sign, and what does that do to our data protection position? | A recommendation with the controller consequence attached. A supplier indifferent to who signs has not thought about who is accountable when a recipient complains. |
| Portfolio reporting | How will you report across eight companies? | Per company, against shared definitions, never blended. If they offer you a single portfolio reply rate, they have told you they do not understand the denominator problem. |
| Sending identity | Whose domains do these emails come from? | Each company's own, registered in that company's name. Anything consolidated across the portfolio means one holding can damage the others. |
| Market consumption | How will we know when a company is running out of market? | A reported share of addressable market contacted and months remaining at the current rate. Almost nobody tracks this and it is the number that predicts a stall. |
| Declining work | Which of our companies would you tell us not to do this for? | At least one, with a reason. A supplier that wants all eight has not looked at all eight. |
What we can show you, and what we cannot
| What GrowthPal has published | What it counts | |
|---|---|---|
| 4M+ cold emails sent for 40+ B2B companies, across Prabesh Khanal's five years of outbound work, including campaigns led before GrowthPal was founded in 2025 | Emails sent, cumulative across every engagement in the founder's career. Not one campaign, not an average, and not confined to GrowthPal. | Source |
| The interested-reply definition and counting rules | Not a figure. A link to the published methodology. | Source |
What stands behind this
- We have not run a portfolio-wide engagement yet. There is no private equity case study on this site, because there is no private equity engagement to write one about. The figures above cover outbound work generally and are defined on the glossary page. If you need a supplier with a named sponsor reference, we are not that supplier today and would rather you heard it here.
- The scope limit is real, not positioning. We do not do deal origination. Not as an add-on, not as a favor, not under a different name. A separate and unrelated company called GrowthPal does work in M&A deal sourcing, which is exactly why this page says so plainly.
- The data protection material is cited to the regulation. The joint controller rule, the processor obligations and the rule that conduct rather than labels decides the role are all sourced at the foot of this page, read in the text as originally adopted. This describes the documents and is not legal advice.
- The authentication standards cited here are current. The policy specification most guidance in this market still points at was replaced in May 2026 by three standards track documents. Those are the ones listed below.
Questions operating partners ask
Do you do deal sourcing or proprietary deal origination?
No. We run outbound for portfolio companies, meaning campaigns that help a PE-backed business sell its product to its own customers. We do not contact company owners or intermediaries on behalf of a fund, do not build acquisition pipeline and do not do fundraising or investor outreach. If that is what you need, a separate and unrelated company also called GrowthPal works in M&A deal sourcing, and this is not that company. This site belongs to Inferno Media, LLC, working from Kathmandu, Nepal.
Should the fund contract centrally or should each company sign?
In most cases each company should sign, because that keeps the controller position clean, survives the sale of any one holding without unpicking anything, and leaves accountability with the management team that owns the result. Central contracting is faster to stand up and gives consistent terms, but it puts the fund at real risk of being treated as a controller alongside the company, which means a recipient can exercise their rights against the fund directly. A framework with per company statements of work gets most of both, as long as the framework stays out of targeting decisions.
Can the fund pay even if the portfolio company signs?
Yes, and that is often the right structure. A fund can subsidize the first engagement to get a company started while the company signs and directs the work. The combination to avoid is signing, paying and directing all three, because that is what makes the fund a controller in fact regardless of what the paperwork says. How that cost is allocated between the adviser, the fund and investors is a question for your finance and compliance function, and it should be settled before the engagement starts rather than after.
How much cheaper is the second company than the first?
Less than you have been told. What gets cheaper is setup: the method is already agreed, the definitions already exist, the reporting format is built, the infrastructure pattern is known, and we already know how your sponsor wants this presented. What does not get cheaper is the message, the buyer definition, the list and the offer, and that is most of the work. Expect a real reduction in coordination cost and a small one in per company effort.
Can we see one reply rate across the portfolio?
You can see one, but it will not mean anything, so we do not produce it. It averages quantities with different denominators, different reply definitions and markets that differ by two orders of magnitude, and it moves mostly with how many of your companies happen to be in warm-up. What you get instead is the same small set of numbers per company against shared definitions, plus the three facts that genuinely compare across holdings: deliverability health per sending domain, program stage, and months of addressable market remaining.
Which of our companies should go first?
The one with the clearest buyer, not the one with the biggest revenue target. The first holding is where the method gets calibrated to your reporting and your expectations, so it should be the one most likely to produce a clean read. Launching four at once means four versions of the same avoidable mistake, discovered at the same time, with four management teams forming the same opinion about the channel.
When in the hold should this start?
Months four to twelve is the honest start window. The first 100 days matter, but for a different reason: that is when the domain architecture, the CRM definitions and the compensation design get set, and those are very hard to change later. The useful first 100 days action is making the decisions that make sending possible, which costs almost nothing, rather than launching a program into an organization that is still moving.
We exit in eighteen months. Is it too late?
For starting a program and having it produce a credible track record, usually yes. A buyer and their diligence provider pay for a repeatable motion rather than a recent spike, and a program started now will be read as the latter. If something already exists, the valuable work is making it legible: documented process, clean definitions and a provable attribution story. We would rather sell you that than a program that will not have time to prove itself.
Will one portfolio company's problems affect the others?
Not if the sending estates are separate, which is why they always are. Each holding has its own domains, its own authentication and its own reputation, because the mailbox providers' published complaint ceilings apply per sender. Consolidate the sending identity across a portfolio and one company's bad list degrades delivery for every other company sharing it. The one thing we do join up is suppression, so a buyer at a shared account does not receive sequences from two of your companies in the same month.
Would you ever tell us not to run this at a company?
Yes, and the two cases worth naming here cost us the work. The first is when retention is the real constraint: if a company loses customers faster than it wins them, acquisition spend makes the numbers worse, and acquisition is visible while churn is not, which is why this gets misdiagnosed. The second is when the sponsor wants a number by a board date. Outbound does not respond to deadlines, and compressed to hit one it gets run at volumes its deliverability cannot carry.
Sources
- Regulation (EU) 2016/679, Article 26, joint controllers, as adopted by the EU, legislation.gov.uk, The National Archives, as adopted, 27 April 2016. https://www.legislation.gov.uk/eur/2016/679/article/26/adopted. Accessed 29 October 2026.
- Regulation (EU) 2016/679, Article 28, processor, as adopted by the EU, legislation.gov.uk, The National Archives, as adopted, 27 April 2016. https://www.legislation.gov.uk/eur/2016/679/article/28/adopted. Accessed 29 October 2026.
- RFC 7489, now obsoleted, status page, IETF, RFC Editor, March 2015, obsoleted 2026. https://www.rfc-editor.org/info/rfc7489. Accessed 29 October 2026.
- RFC 9989, Domain-Based Message Authentication, Reporting and Conformance, IETF, RFC Editor, May 2026. https://www.rfc-editor.org/info/rfc9989. Accessed 29 October 2026.
- RFC 9990, DMARC aggregate reporting, IETF, RFC Editor, May 2026. https://www.rfc-editor.org/info/rfc9990. Accessed 29 October 2026.
- RFC 9991, DMARC failure reporting, IETF, RFC Editor, May 2026. https://www.rfc-editor.org/info/rfc9991. Accessed 29 October 2026.
- Email sender guidelines, Google, Workspace Admin Help, requirements effective 1 February 2024. https://support.google.com/a/answer/81126. Accessed 29 October 2026.
- External research hub, value creation, UK Private Capital, formerly the BVCA. https://www.bvca.co.uk/Research/External-Research-Hub. Accessed 29 October 2026.
- Private equity explained, Invest Europe. https://www.investeurope.eu/about-private-capital/private-equity-explained/. Accessed 29 October 2026.
- GrowthPal reporting methodology and counting rules, GrowthPal. https://growthpal.io/glossary/#methodology. Accessed 29 October 2026.
Related
- Outbound sales diagnosis the single-company version of the measurement questions above.
- Enterprise outbound security review, processing agreements and multi-stakeholder approval in full.
- GTM engineering the infrastructure patterns that do transfer between holdings.
- Cold email agency how a single program is run, which is the same at every company.
- Glossary and reporting methodology the shared definitions a portfolio program depends on.
Send us the list of holdings and what each one sells. We will tell you which are worth running outbound for, which are not, and why, before anyone signs anything.
A plain text version of this page is at /private-equity-cold-email-agency.md, and the site index for language models is at /llms.txt.