138. A list can’t learn. Your acquisition pipeline should.

I started with a spreadsheet of cleaning companies from a 2023 campaign.

It was fragmented, out of date, and missing a lot of the information I needed. Some companies had no useful contact details. Others were in the wrong geography or did the wrong type of work.

I used it anyway.

My first campaign this year started with that list and a version of the email sequence I’d used in the 2023 campaign. The aim was to build momentum, learn from the response, and improve from there.

It did get me moving.

It also showed me the difference between having a list and having a pipeline.

A list tells you who exists.

A pipeline tells you who fits, who is ready, and what happens next.

The first time I bought a cleaning business, I didn’t have a clear acquisition thesis.

In 2015, I bought ProClean. The deal closed. The hard part started afterwards.

I hadn’t understood the business well enough before buying it. Integration was messy. Costs appeared late. The acquisition dragged down the other businesses I had built.

The lesson was not “don’t buy another business”.

It was to set the rules before looking for one.

My current search starts with cleaning businesses doing £1–5M in turnover. I'll consider businesses between £600,000 and £1M, but the recurring revenue and operating structure need to justify it.

The target isn’t residential cleaning. It’s contracted work across commercial buildings and large residential estates, including:

Office, industrial, and retail cleaning can fit. Large-scale deep cleaning, voids, and specialist work can fit too.

The question isn’t which sector label sits on the website.

It’s how the business operates.

I want recurring revenue, a dense geographic footprint, and a business that can survive a handover. The owner’s age matters less than the situation. Someone approaching retirement with no succession plan is an obvious fit, but a younger owner might have other reasons to sell.

I also want second-tier management and enough operational stability to build from.

Not perfect. But stable enough to improve.

My first list couldn’t answer the questions that mattered.

That’s when I started rebuilding the list as a market map.

Plimsoll gave me better sector and financial information. Companies House helped me verify the company, its directors, and its filing history. I added geography, revenue, service, and ownership filters.

That changed the work.

Instead of starting with “How do I find the email address?”, I could start with “Does this company deserve further research?”.

If the starting data is poor, every stage after that slows down. Crap in, crap out.

Better filtering improves the rest of the system.

I tried to automate the enrichment process.

Take company data → find the website → identify the owner → locate and verify an email address → create the record in HubSpot → prepare it for an Instantly campaign.

The market made that harder than it sounded. The UK cleaning market is fragmented, with 90% of UK cleaning businesses employing fewer than 10 people.

Most companies have outdated websites, unclear email addresses and no LinkedIn profile. Revenue and service information is hard to verify.

The numbers from my first full run showed the extent of the problem:

The verified email hit rate was roughly 57%.

Useful, but nowhere near complete.

My original plan was to build one system that managed the end-to-end process. In practice, that made it harder to see where the process was breaking.

I’m currently rebuilding the process as a modular system. Company screening, contact enrichment, email verification, CRM management, campaign delivery, and reply handling. Each is tested separately.

Less elegant than the original idea. But more useful.

Manual enrichment is the slowest part of the process.

I use a combination of Hunter, Apollo, Google, LinkedIn, company websites, and Companies House. Sometimes the information is available in seconds. Other times it takes a few searches to confirm that I have the right business and the right owner.

The downside is time. The upside is that I learn more about each company.

I can eye-test each website, LinkedIn profile and company record. I can see the type of work a company delivers and the clients it serves. Not something that automation can reproduce.

One business I spoke to described itself as residential cleaning. On that label alone, I'd have rejected it.

The operating profile told a different story. It provides communal cleaning and waste management across build-to-rent estates. The sites are large, the work is recurring, and the client relationships can support more than one service.

That’s a Tier 1 target.

Tiering prospects tells me where to spend my attention. It doesn’t tell me whether to buy the business.

That's why the tiering system matters. It helps me decide which companies deserve deeper research, personal outreach, and longer follow-up.

Email is my primary route. I use Instantly to manage campaigns and HubSpot to log company records and contact history.

LinkedIn is my secondary route. Once an email campaign ends, non-responders move into a LinkedIn cycle.

I'll test direct mail in future campaigns.

The first full outbound run produced one interested reply in its first week. Not enough data to declare the system a success, but enough to show it can create a conversation.

The next step is keeping the conversation moving.

HubSpot is my source of truth. Every contact, campaign, reply, status, and next action sits here.

My rule is simple: A reply isn't a lead until it has a status and a next action.

“Not right now, but keep in touch” isn’t a dead end.

It becomes:

The same applies to a positive reply, an unsubscribe, or no response at all. Each outcome changes what happens next.

Without that structure, every new campaign starts from zero.

The system isn't finished, but it's come a long way since my first campaign back in April.

I'm still improving the enrichment stage. I need to add reporting and better campaign analysis. My outreach content (emails, LinkedIn, letters) is at different stages of development.

The metrics I care about are:

This feedback shapes the next campaign.

A high bounce rate points back to the data. Replies from poor-fit companies point back to the filters. Strong responses from one tier should change where I spend research time. Repeated objections should change the message or the timing.

That's the loop I am trying to build.

I started by thinking the problem was sending more emails.

It wasn't.

The problem was getting better at deciding who to contact, what to say, and when to try again.

A list can't do that.

A pipeline can.

That's all for this week.

Next, I’ll explain how I sort companies into my tiering system — and decide which ones deserve attention first.

Matt Harris

The Growth Lab

Most acquisition emails never earn a reply.

If you run a cleaning or Soft FM business, I’d like to know what would make you answer one — even if the answer were “not now”.

Reply and tell me. I read every response.

Growth Lab Capital is actively looking at Soft FM businesses across London, the Home Counties, the Midlands, and the South East. If you are an operator thinking about what comes next — even if it is a few years away — I am happy to have a conversation. No broker. No pressure. [BOOK A TIME HERE]

The Growth Lab Intelligence newsletter is a playbook for buy-and-build by Soft FM operators. Deals, growth, operations — no theory, no fluff. Subscribe for the next one.

📌 Daily on LinkedIn and X — Soft FM. Buy & Build. No fluff.