Where Stannary's next demand should come from.

Section 1The situation, and what we are assuming

Stannary Systems — Software for scheduling, evidencing and reporting regulated inspections. 31 people.

Stannary sells to companies that carry out statutory and contractual inspections on utility infrastructure — gas, water and electrical distribution assets. The product schedules the work, captures the evidence in the field, and produces the report the asset owner’s auditor will accept. Growth to date has come from two founders’ industry contacts and from a partner reseller. That source is now exhausted, and the one salesperson hired to replace it has spent five months sending the same email to anyone whose job title contains "operations".

Assumptions, stated so you can correct them

Everything below rests on these. If one is wrong, tell us which, because several conclusions move with it. We have marked them as assumptions rather than findings because that is what they are: no interviews were conducted, no survey was run, and no external research is cited anywhere in this document.

• Stannary has one full-time salesperson, no marketing function, and no analyst or researcher.
• The founders will approve messaging within two business days, and will not approve a message that names a customer.
• There is no CRM history worth mining: deals were tracked in a spreadsheet, and lost deals were not recorded at all.
• The product can already produce the evidence export that a large asset owner’s auditor accepts, because two existing customers pass their audits with it. That is a claim from the founders, not something we have verified.
• No budget exists for paid media, events or a second hire during the period this plan covers.

Section 3Why B is excluded

Segment B fails on one thing only, and it is the thing outbound cannot fix. The person who feels the pain does not hold the budget, and the person who holds the budget owns a system Stannary does not integrate with. A perfect email to the compliance manager produces a sympathetic reply and then a nine-month procurement process Stannary has no reference to survive. That is not a messaging problem and no amount of targeting work will make it one.

Segment A is the opposite shape. One person feels the pain, holds the budget, and can sign. The decision is measured in weeks. And Stannary already has two customers of exactly that shape who pass their audits, which means the claim we would make is one it can substantiate.

Excluding B also costs less than it looks. B is a small number of large organisations. They are reachable by name, at any time, by a founder. Nothing about excluding them from an outbound programme prevents the founders from working two of them by hand.

And two cuts inside A

Choosing segment A is not yet a list. Two more cuts, each with its reason:

Contractors under about 20 technicians. The pain is real and they will say so. The budget is not: at that size the work is absorbed by the owner personally and a subscription competes directly with their own time, which they price at zero. They also change size quickly, in both directions.

Contractors over about 200 technicians. They have usually already built something internal, and the decision becomes a committee with an incumbent to displace. That is a sale Stannary can win, but not one a cold email starts.

What would change this recommendation

An exclusion without a reversal condition is a guess wearing a suit. These are observable, and we will be watching for them:

1. Two or more of the next ten inbound conversations from an asset owner reach a named budget holder within 30 days without an IT integration gate appearing. That would mean the buying shape we assumed is wrong.
2. An asset owner buys without requiring integration with its asset-management platform. One instance is enough to reopen the question.
3. Fewer than roughly 40 qualifying framework or term-contract awards are published per quarter in the target geography. That does not rescue segment B, but it does mean the trigger below is too thin to build a programme on, and segment A has to be worked without it.

Section 4The buyer, and why now

Who. The operations director at an independent inspection contractor with roughly 20 to 200 field technicians.

Why now. They have just won or re-bid a framework or term contract with a utility asset owner, and have a fixed mobilisation window — commonly 60 to 90 days — before the first inspections are due under it.

Three things become true on the day of the award and were not true the week before. The evidencing and reporting standard is now the asset owner’s, not theirs. The volume they have to deliver is written into the contract. And there is a mobilisation budget with a date on it, which is the only moment in the year when this purchase is easy to approve.

Reason to talk. Naming the specific award, the date the first inspections fall due under it, and the evidence format that asset owner’s audit will ask for.

The three signals, and what each is for

• Published framework or term-contract award. The trigger itself. Dated, public, and specific enough to name in the first line.
• Field inspector job adverts posted within 30 days of the award. Corroboration that they are actually mobilising rather than holding a place.
• Technician headcount in the 20–200 band. The segment cut. Checked before the account enters the list, never after.

This hypothesis is wrong if: If messages carrying the named award do not out-reply the same messages without it, the trigger is not doing the work we think it is, and the expensive part of this programme — finding and dating the awards — should be stopped rather than scaled.

Section 5What is wrong with the message now

Four problems, in the order they cost the most. Each one names what we observed, why it costs something, and what goes there instead. The full mark-up of the current email is the second example.

M1The message leads with the product’s capabilities: scheduling, mobile forms, reporting dashboards.

Every competitor can write the same three words, so the sentence carries no information. Worse, it invites the reader to compare on a list of features, which is the only comparison Stannary loses — the generic field-service tools have longer lists.

Instead: Lead with the thing only Stannary can say: the evidence export that this asset owner’s audit accepts.

M2It is addressed to "operations managers" as a title, at any company with "field" or "facilities" in its description.

That definition includes companies with no statutory inspection obligation at all. Their non-reply is not a signal about the message; it is a signal that they were never in the market. It also makes every reply rate uninterpretable, because the denominator contains the wrong companies.

Instead: The segment cut and the trigger, applied before the account enters the list.

M3It claims a speed improvement as a percentage, in the words up to 50% faster

There is no denominator, no baseline and no source. A buyer who has evaluated three of these tools has read the same number three times. It costs credibility without buying attention, and it is the sentence a procurement reviewer will ask Stannary to evidence.

Instead: Either the real figure with its denominator and who it came from, or nothing. In this plan: nothing, because the figure does not exist.

M4The ask is a 30-minute demo, in the first cold email.

Thirty minutes is a real cost to an operations director in a mobilisation window, which is precisely when they have least of it. The ask is larger than the trust the message has earned.

Instead: A smaller ask with a clear end: send the evidence-format checklist for that asset owner, and offer fifteen minutes only if the checklist shows a gap. The tradeoff is real and worth naming — a smaller ask produces more replies and fewer of them are meetings. It is the right trade only because the checklist itself qualifies: someone who reads it and comes back has told us more than someone who accepted a demo out of politeness.

Section 6The test plan, in priority order

Three tests. Each one names the decision it can answer, and the order is not arbitrary: the most expensive assumption is tested first, so that the thing which would waste the most money is the thing we find out about soonest.

Test 1Does naming the specific framework award beat the same message without it?

How. Two variants of one message to the same segment, identical but for the trigger sentence. Equal allocation.

The decision it answers. Whether to build the programme on trigger detection at all.

Why in this position. Because it is the most expensive assumption. Finding, dating and verifying awards is most of the ongoing work. If the trigger does not move the reply rate, everything built on top of it is waste, and that is worth knowing in week two rather than month four.

Test 2Does the mobilisation-deadline frame beat the audit-evidence frame?

How. Same trigger, two openings: one about the date the first inspections fall due, one about what the auditor will ask for.

The decision it answers. Which pressure the buyer is actually acting on, which then sets the frame for everything downstream.

Why in this position. Second because it only matters if test 1 says the trigger works.

Test 3Does the checklist ask beat the fifteen-minute ask?

How. Winning message from tests 1–2, two closing asks.

The decision it answers. Whether the smaller ask trades enough extra replies for the meetings it gives up.

Why in this position. Last because it is the cheapest to change and the easiest to reverse. Nothing else depends on it.

Section 7What has to exist before any of this runs

Six inputs. Four of them are Stannary's, and the work does not start without them. Naming the blocker next to each is the point: an input with no consequence attached does not get produced.