How to give an elevator pitch (in 30 seconds, without sounding rehearsed)

A short, structured pitch that earns the next conversation — and the five-line template you can rehearse out loud until it stops sounding like a script.

How to give an elevator pitch (in 30 seconds, without sounding rehearsed)

The phrase "elevator pitch" has been around since the 1980s and it has been mostly misunderstood ever since. Most people treat it as a sales close that has to be impressive — a tight, polished mini-monologue designed to wow somebody in under a minute. That framing is exactly why so many elevator pitches fail. The job of the pitch is much smaller, and the skill is correspondingly different.

This guide is the version I wish I had been handed the first time I had to introduce myself to someone whose attention I had not earned yet — at a conference, in a hallway, on a sales call, in an interview, at a networking event. The five-line template below works for all of those, and the practice routine is the part that actually makes it land.

What is an elevator pitch, really?

An elevator pitch is a short, structured introduction — usually 30 to 60 seconds — that gives a listener just enough information to decide whether they want a longer conversation with you. That is it. It is not a sale, not a resume, not a manifesto. It is a doorway.

Two things follow from that definition, and they govern everything else in this guide:

  1. The goal is the next conversation, not a yes. A pitch that ends with a clean handoff out-converts a pitch that ends with an ask, almost every time, because the listener is still in evaluation mode and a premature ask reads as pressure.
  2. You are budget-constrained. Thirty seconds is around 75 spoken words. You cannot afford a single sentence of throat-clearing, history, or jargon. Every line has to do real work.

Founders pitch a company, sales reps pitch a product, job seekers pitch themselves — but the shape is the same in all three cases. You are compressing identity, problem, and proof into a chunk small enough that a busy person can take it in without it costing them anything.

The five-part structure

There are a lot of elevator-pitch frameworks floating around, and most of them are some variation of the same five elements. Use this as the skeleton:

  1. Who you help. A specific audience, named in their own language. "Series A startups", "early-career data engineers", "criminal defence solicitors in regional courts." Not "businesses" or "people."
  2. The problem they hit. One concrete pain point that the audience would recognise instantly. Specific enough that the listener nods.
  3. What you do about it. A short description of your product, role, or service in plain words. No buzzwords, no category jargon. If your mum could not repeat it back, rewrite it.
  4. Why this beats the obvious alternative. A single differentiator from whatever the listener would otherwise reach for — a competitor, an in-house build, a different vendor, a different candidate. One reason, not five.
  5. One proof point or hook. A concrete number, customer name, outcome, or credential that makes the rest believable. Then end with a soft handoff — a question or a one-line offer — that invites a follow-up without demanding one.

Five lines. Roughly one breath each. Around 60–80 words total. Time it.

If you are pitching yourself rather than a company, swap "who you help" for "who I work with" and "what you do" for "what I do" — the five-element shape is identical. The mistake is to start with your CV or your degree; the listener does not have the context yet to care.

Three worked examples

These are written the way they would be spoken, not the way they would be written. Read them out loud and you will hear the difference.

Sales rep (B2B SaaS)

I work with mid-market RevOps teams who keep losing forecast accuracy whenever their sales org grows past about thirty reps. We build a pipeline-hygiene tool that sits on top of Salesforce and flags the deals that are quietly slipping — the ones reps stop updating two weeks before they die. Most teams in that band hire another ops analyst to do it manually; we do it automatically and catch about three weeks earlier on average. One customer recovered a million in slipped pipeline in their first quarter. Are you running RevOps at that scale yet?

Roughly 75 words, lands in about 30 seconds, ends with a question that opens the conversation instead of asking for a meeting.

Founder (early-stage)

We help solo founders running consultancy businesses who are tired of losing two hours every Friday on invoicing and admin. It is a small calendar-integrated tool that turns calendar events into draft invoices automatically — you confirm or edit, it sends. Most founders at that stage are stitching together Notion plus a spreadsheet plus a free invoice generator; we replace all three with one click on a Friday afternoon. We have got about four hundred consultants on it now and the typical user saves around an hour and a half a week. Happy to send you a one-minute video if it is useful.

Note the soft offer at the end — not a sign-up, not a meeting request, just a low-friction next step.

Job seeker (career switcher)

I work with product teams who are trying to ship faster but keep getting bottlenecked on data — the analyst is two weeks behind, every decision is stalled. I am a former backend engineer who moved into analytics about three years ago, so I write the SQL and the pipelines myself instead of waiting for them. At my last role I cut the data-request queue from eleven days to under two, which unblocked four product launches that quarter. I am looking for the next team that wants that kind of in-line analytics work — is that anything close to what you are hiring for?

Same shape, applied to a person rather than a company. The proof point is a real outcome, not a job title.

Common mistakes to cut

Once you have a draft, the way you make it better is almost entirely subtractive. The pitches that land are the ones with the least fluff. Cut these first:

  • The throat-clear. "So, basically, what we do is…" The first three words are dead air. Open with the audience.
  • The company history. Nobody in a 30-second window cares when you were founded or who used to work there. Save that for the meeting your pitch earned you.
  • Stacked buzzwords. "AI-native, end-to-end, vertically integrated revenue platform" tells the listener exactly nothing. Replace each one with a concrete noun.
  • The list of features. A list of three features is three pitches, badly compressed. Pick one differentiator and let everything else surface in the next conversation.
  • The hard ask at the end. "Can I have fifteen minutes next week?" lowers your conversion at this stage because the listener has not bought in yet. A question or a one-line offer outperforms a meeting request.
  • The rehearsed cadence. A pitch that sounds memorised reads as a sales script even when it is technically excellent. The fix is rehearsing it as speech, not as text — more on that below.

The 30-second budget

A natural conversational speaking rate is around 150 words per minute, so 30 seconds buys you roughly 75 words. Sixty seconds is around 150. That is not a lot of room. A useful drill: write the pitch in a doc, paste it into a word counter, and aim for 60–80 words for the 30-second version and 130–160 for the longer one. If you are over, you are cutting; you are not adding.

A second budget check: read the pitch out loud at the speed you would actually use, with a stopwatch. Reading it in your head is worthless — silent reading runs about 30% faster than speech, so a 40-second written pitch becomes 55 seconds said out loud, which is past the window.

The single most common reason a pitch goes over budget is that it tries to introduce two ideas in the "what you do" line. If you find yourself saying "and also" or "as well as", you have two pitches. Pick the one the listener cares about first and let the other one come up in the follow-up.

How to deliver it (so it does not sound rehearsed)

Content gets you to "technically correct." Delivery is what makes the listener actually want the next conversation. Three things matter, in this order:

Pace. A rehearsed pitch almost always speeds up — the speaker is uncomfortable and rushes. Force yourself to land each of the five beats with a half-breath in between. The pitch should feel slightly slower than your normal speech, not faster.

Eye contact and pause. End the pitch on a clean stop, not a trailing "...so yeah". A short silence after your last line is uncomfortable for you and informative for the listener — it is the cue that it is their turn. Hold it.

Conversational register. If you can hear yourself reciting, the listener can too. The fix is to draft the pitch as how you would actually say it — contractions, short sentences, the odd "basically" if that is how you talk — rather than how you would write it on a website. Then rehearse it out loud, not in your head, until the wording is yours.

How to practise it

This is the part most people skip, and it is the difference between a pitch that lands and a pitch that gets a polite nod. A short, repeatable rehearsal loop:

  1. Write the five lines. Time them silently first to confirm you are in the 60–80 word range.
  2. Record yourself saying it. Phone voice memo is fine. Listen back once. You will immediately hear the parts that sound rehearsed, the words you trip over, the place you ran out of breath.
  3. Rewrite the bits that sound off. Usually one or two phrases. Re-record.
  4. Run it under pressure. This is the move that matters. Pitch it to someone who can interrupt you, ask a sceptical question, or push back. The polished solo version always sounds different when somebody is actually listening — the test is whether you can still hit the five beats when the listener cuts in halfway through line three.
  5. Iterate one variable at a time. Cut a buzzword, swap a proof point, change the closing question. Run it again. Three to five honest reps tighten a pitch more than an hour of solo polishing.

If you do not have a willing human handy for step four, an AI conversation partner that can interrupt, push back, and give you live feedback on pace and clarity is the closest substitute — much closer than a mirror.

Key takeaways

  • An elevator pitch is a doorway, not a close — its job is to earn the next conversation.
  • Use the five lines: who you help, the problem, what you do, why it beats the obvious alternative, and one proof point with a soft handoff.
  • Aim for 60–80 words for 30 seconds, 130–160 for 60 seconds — measure it out loud, not in your head.
  • Cut the throat-clear, the company history, the stacked buzzwords, the feature list, and the hard ask.
  • Delivery beats content: slow the pace, end on a clean stop, draft it as speech rather than text.
  • Rehearse out loud under interruption — a few reps against real pushback beat an hour of solo polishing.

When to use which version

You will end up with two pitches, not one: a 30-second version for cold introductions and a 60-second version for warm ones.

  • 30 seconds — cold context. A stranger at a conference, a recruiter on a first call, a buyer who picked up because you got past the gatekeeper. The listener has given you no signal yet, so the budget is short. Use the five lines, soft handoff, stop.
  • 60 seconds — warm context. An intro from a mutual contact, a panel where you are introducing yourself, an interview opener. The listener has signalled some interest, so you have room for one extra sentence of context per beat. Same five-line skeleton, slightly more flesh on each line, same soft close.

If you only build one, build the 30. The 60 is easier to expand from the 30 than to compress to it.

Frequently asked questions

How long should an elevator pitch be?

Aim for 30 seconds for a cold introduction and 60 seconds for a warm one — roughly 60–80 spoken words and 130–160 respectively. Anything longer stops being an elevator pitch and starts being a presentation, which is a different conversation with different rules. Measure it out loud with a stopwatch, not in your head, because silent reading is about 30% faster than speech.

Should I memorize my elevator pitch word for word?

Memorize the structure, not the script. A pitch that comes out word-perfect almost always sounds rehearsed, which is exactly what you do not want — the listener should feel like they are being spoken to, not pitched at. Drill the five beats (who you help, the problem, what you do, why it beats the alternative, your proof point) until you can deliver them in any order if you have to, and let the exact wording flex with the conversation.

What's the difference between an elevator pitch and a sales pitch?

An elevator pitch is an opening — its job is to earn the next conversation. A sales pitch is a close — its job is to get a buying decision. Confusing the two is the most common reason elevator pitches fail: they try to do the sales pitch's work, end with a hard ask, and overwhelm a listener who is still in evaluation mode. Keep the elevator pitch short and end it on a question or a soft offer; the sales pitch comes later, with a buyer who has signalled they want it.

How do I start an elevator pitch without it sounding awkward?

Open with the audience, not with yourself. "I work with [specific audience] who..." beats "Hi, my name is..." or "So, basically..." because it gives the listener something to evaluate in the first three seconds. If they are in or near that audience, they lean in; if they are not, you have not wasted their time pretending the pitch was for them. Your name and title can come at the end if they are still needed.

Can I use the same elevator pitch in interviews and at networking events?

The skeleton is the same, the proof point changes. In an interview, your "why this beats the alternative" line compares you to other candidates, and your proof point is an outcome you delivered. At a networking event, the comparison is to the listener's current solution or vendor, and the proof point is a customer outcome or company milestone. Write one core pitch and swap the last two lines for the context — easier than building two pitches from scratch.