Before you dial
On this page
The call starts long before the ringing does. Reps who "aren't good on the phone" are usually reps calling the wrong list at the wrong time with no reason to have rung.
Building the list
Work from a named account universe, not a keyword search. In UK life sciences the market is small enough to enumerate, and the sources are unusually good:
- NHS trust board papers and annual reports — published because board meetings are held in public. Risk registers, waiting-list figures, digital strategies and capital plans are all there.
- Integrated Care Board plans — where system-level priorities and money sit.
- Job adverts — the single best timing signal in the market. Two lab techs and a QA manager advertised at one site means expansion, and expansion means new systems.
- Grant and funding databases — UKRI, Innovate UK, published Series A/B announcements.
- Conference exhibitor and speaker lists — who is spending on visibility, and who is publicly identified with a problem.
- Publication affiliations — for research-facing products, the corresponding author's institution is a lead.
Then tier it. Tier 1 — strong fit and a live trigger: six call attempts each over three weeks, plus email and LinkedIn. Tier 2 — strong fit, no trigger: three attempts, lighter cadence, revisited quarterly. Tier 3 — fit unconfirmed: one attempt to test the hypothesis, then promote or drop.
The research that earns 30 seconds — and the research that doesn't
Earns it: a problem they are measured on, named in a public document ("your April board papers put specimen turnaround on the risk register"); a trigger event with a date — a new site, a funding round, a hiring spree, a contract expiry, a published inspection finding; a peer they would recognise; and their actual remit, so you don't pitch a lab system to someone who runs estates.
Wastes it: where they went to university, their marathon time, their five-year work anniversary — fake intimacy from a stranger is worse than none; company boilerplate read back to them; reading a 90-page annual report for one call; and personalisation for its own sake. A specific, correct observation about their problem beats a specific, correct observation about their hobbies every time.
The three-minute rule: three minutes of research for a tier-1 contact, thirty seconds for tier 3. If you can't find a personal hook in three minutes, dial anyway using the segment-level hook — "the thing pathology managers keep telling us" — which is true, useful and needs no research at all.
The volume maths, re-derived
Every rep is eventually asked "how many calls should you make?" The answer is arithmetic, so do the arithmetic. First define the stages precisely — most published statistics are useless because they blur these. A dial is one attempt to one number. A connect is any human answering, gatekeeper included. A target connect is the person you rang, on the line. A conversation is them engaging past your opener, roughly a call running beyond a minute. Then a meeting booked, and a meeting held — they turn up.
Here is a deliberately conservative planning model for a considered UK life-sciences sale. These are planning rates, not measured statistics — replace each with your own as soon as you have thirty days of data.
| Stage | Planning rate | Per 100 dials |
|---|---|---|
| Dials | — | 100 |
| Connects (any human) | 18% of dials | 18 |
| Target connects | 55% of connects | 10 |
| Conversations | 60% of target connects | 6 |
| Meetings booked | 25% of conversations | 1.5 |
| Meetings held | 75% of booked | 1.13 |
The derived numbers are the ones worth memorising: ≈17 dials per real conversation (100 ÷ 6), 4 conversations per booked meeting (6 ÷ 1.5), ≈67 dials per booked meeting (100 ÷ 1.5) and ≈89 dials per held meeting (100 ÷ 1.125).
Now scale it. A power hour of 45 dials run every working day, in a 20-day month, is 45 × 20 = 900 dials, which at 1.5% gives 13.5 meetings booked and, at 75% held, 10.1 held.
Reverse it for a target. To hold 8 meetings a month you need 8 ÷ 0.01125 = 711 dials, or 36 a day, which at 45 dials an hour is 48 minutes of dialling. That is the most encouraging calculation in this module: a self-sourced pipeline of eight held meetings a month costs under an hour a day — if you actually spend the hour.
Two sensitivities, because interviewers probe them:
- If you're half as good (0.75% booked per dial, 70% held), 900 dials yields 6.75 booked and 4.7 held. To still hold eight you'd need 8 ÷ 0.00525 = 1,524 dials a month — 76 a day, about an hour and forty minutes. Weak technique doesn't stop you hitting target; it doubles the price.
- If you improve one coachable variable — conversation-to-meeting from 25% to 35% — you get 2.1 booked per 100 dials instead of 1.5: a 40% lift from the same 900 dials. Technique compounds faster than volume, which is why section 7 exists.
Finally, remember RAIN's eight touches. At six call attempts per contact, 900 dials a month is only 150 contacts — so a tier-1 list of 100 accounts with two contacts each is 1,200 dials, roughly six weeks to work once. Plan coverage in weeks, not days.
Power-hour mechanics
- Same time, every day, protected in the calendar. 10–11am is the strongest window in the published data; 2–3pm the reliable second.
- Prepare the night before — names, numbers, one-line hooks, in one list, in order. Research during a power hour is procrastination wearing a lab coat.
- No CRM notes during the hour. Three words on paper; log everything in the fifteen minutes afterwards. 45 dialling, 15 logging, is a clean hour.
- Next number within ten seconds. The gap after a bad call is where quitting happens; removing the gap removes the decision.
- Dial through, review after. Never change your opener mid-block on the evidence of one hang-up.
- Call as a pair. Two people at one table, hearing each other, each make more dials and better ones. This is the origin of the "blitz".
- Stand up. It changes your breathing, which changes your voice.
Headset and environment
A wired USB headset with a boom mic beats earbuds and laptop microphones decisively — buyers hang up on bad audio without articulating why. Call from a room with soft furnishings: hard rooms sound like call centres, and call-centre acoustics prime call-centre expectations. Record yourself once and listen. And check your outbound number displays a real, reachable UK number — a regulation 24 requirement that also raises answer rates, since a withheld number is the easiest decline in the world.
Field noteParallel and AI-assisted diallers became common in 2026, and they change both the maths and the ethics. They lift dial volume several-fold, but the moment more calls are launched than there are humans to take them, you are in Ofcom's persistent-misuse territory — abandoned and silent calls, with no permitted threshold since the 3% safe harbour was withdrawn in 2017 and penalties up to £2 million. If you join a team running one, ask how abandoned calls are measured and who owns that number. If nobody knows, that is your answer.
3. The first 15 seconds
You have about fifteen seconds before the prospect decides whether you are a person with a reason or a nuisance with a script. Everything in this section is about that window.
Why tone beats script
Two reps reading identical words get different results, because the prospect is not parsing your sentence — they are classifying your sound. Four variables do almost all the work:
- Pace. Nervous reps speed up, which signals "I expect to be cut off" and invites being cut off. Speak slightly slower than feels natural.
- Inflection. End statements with a downward inflection. A rising one turns "I'm calling about specimen turnaround" into a request for approval. Only genuine questions should rise.
- The pause. After "Hi Ruth, it's Alex from Corvus Labs" — stop for a full beat. Reps who charge through the pause sound like a recording, because recordings don't pause.
- Register. Calm, unhurried, mildly amused — a peer ringing a colleague, not a supplicant asking for time. The hardest thing to fake and the most valuable thing to build.
The script's job is to free your attention so you can listen. Know the beats; never read the words.
Five openers that work, and when each fits
1. The permission-based opener — the default for senior, clinical and public-sector buyers.
"Hi Ruth, it's Alex from Corvus Labs. [beat] I know I'm interrupting your morning — can I have thirty seconds to tell you why I rang, and you can tell me if it's nonsense?"
Fits almost everywhere, especially cautious, formal or clinical buyers. Why: it hands control to the prospect, so they don't have to seize control by hanging up. Risk: said fast or apologetically it becomes a plea. Say it slowly.
2. The "I know I'm interrupting" honest opener — a blunter cousin.
"Hi Ruth — Alex from Corvus Labs. This is a cold call. You can hang up, or you can give me twenty seconds and decide after."
Fits operational and commercial roles, people who value directness, anyone who has already brushed you off politely once. Why: naming the thing everyone is thinking discharges the tension, and offering the hang-up openly makes taking it feel rude. Risk: reads as a gimmick with very senior clinical buyers.
3. The context-led opener — Gong's top performer.
"Hi Ruth — Alex from Corvus Labs. We work with the pathology teams at [two named comparable trusts]. Not sure whether our name's crossed your desk yet?"
Fits markets where you have genuine, recognisable peer customers. Why: it borrows credibility before asking for anything, and gives them a question they can answer in one word. Risk: dishonest if the customers aren't real or aren't comparable. Never inflate this.
4. The trigger opener — strongest in the NHS, where the documents are public.
"Hi Ruth — Alex from Corvus Labs. I was reading your trust's April board papers, the section on specimen turnaround. That's the reason I rang. Have you got a minute, or should I call back?"
Fits any buyer whose problems are documented publicly. Why: it proves in one sentence that you did work almost no caller does. Risk: only works if you actually read it — prospects test this.
5. The internal-signpost opener — only ever if literally true.
"Hi Ruth — Alex from Corvus Labs. I spoke to Jamie in procurement earlier and he said specimen tracking sits with you rather than with him. Is that right?"
Fits accounts where you've already had a real conversation with someone else. Why: it converts a cold call into a warm-ish one and gives them an easy factual question. Risk: catastrophic if exaggerated. Never imply a referral you did not receive.
The openers that get you hung up on
- "How are you today?" — the recognised opening line of every telemarketer in Britain. Gong measures it at 7.6%, well below the honest openers, and here it reads as insincere before you finish the sentence.
- "Did I catch you at a bad time?" — Gong's worst at 2.15%. It offers a scripted, socially costless exit, and most take it.
- "Am I speaking with Ruth Vance?" — sounds like identity verification, which sounds like a scam or a debt collector.
- Company boilerplate. "We're a leading provider of AI-enabled laboratory workflow solutions" is not a sentence any human has said aloud to another human.
- False familiarity. "Ruth! It's Alex!" implying prior acquaintance buys four seconds and costs the relationship the moment it's clocked.
- The fake pretext. "Just following up on my email" when the email was a template, or "this isn't a sales call" when it is — precisely the behaviour the ICO singled out in the case in section 1.
- Anything over fifteen seconds before you stop for breath. If they haven't spoken by second twenty, you're broadcasting.