Who should you target, and why now?
Outbound usually fails before the first email is written. Start with a narrow ideal customer profile: one type of company, one size band, one buyer. If you cannot describe the company in a sentence, your emails will not be specific either.
Then add timing. A static list tells you who could buy; a signal tells you who has a reason to look this month. Useful B2B signals are public and recent: a complaint about a competitor, a job post for the role your product supports, a funding round, a launch, a request for tool recommendations. Build the list from signals first and fill the gaps with firmographic search.
Size the list to what you can follow up on properly. Three hundred well-chosen companies a month, researched with care, will teach you more than three thousand you barely looked at. The guide to finding B2B leads shows how to work backwards from a meeting goal.
Which outbound motion fits your market?
Before you write a word, decide how you will choose who gets an email. Three motions cover most B2B outbound, and the right one depends mostly on how many companies could buy and how much each deal is worth. Many teams combine two, for example signal-based outreach for the broad market and account-based work on a short list of large targets.
| Motion | How it works | Fits when | Main risk |
|---|---|---|---|
| List-based volume | Everyone matching your filters gets the same short sequence | Very large market, simple product, small deals | Spam complaints, burned domains, little learned per email |
| Signal-based | Only companies with a recent, public reason to buy are contacted, and the email names that reason | A few thousand possible buyers, deals of several hundred dollars or more, signals you can see | Fewer leads in a very narrow niche |
| Account-based | A fixed list of named accounts, each researched by hand, several people contacted per account | A few hundred possible buyers, large deals, long cycles | Slow, and expensive per account |
What should you offer, and how should you write it?
Make a small, useful offer
The offer is what you ask for and what the prospect gets in return. Most outbound asks for 30 minutes to "learn about your goals", which gives the prospect nothing. A better offer is small, specific and useful even if they never buy. Match its size to the strength of the signal: someone comparing alternatives to a competitor can take a demo; someone who just raised a round probably wants one sharp idea first.
- A short teardown: "I found three places your onboarding emails lose people; want the notes?"
- A benchmark or sample output built from their own public data.
- A 15-minute call with a narrow agenda tied to the event that triggered the email.
- A time-boxed pilot with written success criteria, for later steps.
Write from the trigger, in their words
A good first email has three parts: why you are writing now (the trigger), why it matters to them (the problem) and one easy next step. Keep it under about 120 words, write it to one person, and cut every sentence that would be true of any company.
Personalization is a reason, not a compliment. "Saw your post asking how to cut churn reporting time" is personalization; "Love what you are doing at Acme" is noise. Use the vocabulary your customers use in interviews: if they say "double-booked rooms", do not write "resource allocation inefficiencies". The cold email templates guide has examples written this way, one per trigger.
How should you design the sequence, and which channels belong in it?
A sequence is the series of touches one contact receives. Three to five emails over two to three weeks is a sensible default, and each follow-up should add something new instead of "bumping this to the top of your inbox". Stop the sequence the moment someone replies; pause it on an out-of-office and resume after the return date.
Email is the backbone for most small teams because it scales and leaves a written record. LinkedIn works as a manual second touch on high-value accounts: a short connection note or a thoughtful comment on a recent post. Phone can work when you have a direct number and a specific reason to call. Add one channel at a time, and only once the first one works.
| Step | Day | Purpose | Example line |
|---|---|---|---|
| 1. First email | Day 1 | Name the trigger and offer a specific next step | "You mentioned that exporting reports from {competitor} takes a support ticket. Ours is one click; want a two-minute video of it on data like yours?" |
| 2. Proof | Day 4 | Add one new piece of evidence | "Here is our migration checklist for teams moving from {competitor}, including how saved reports come across: {link}." |
| 3. New angle | Day 9 | Answer the likely objection | "If being mid-contract is the blocker, we can import your history now and start billing when your renewal ends." |
| 4. Close the loop | Day 16 | Make it easy to say no | "I will stop here. If reporting becomes a priority, reply with later and I will check back next quarter." |
How do you protect deliverability and stay compliant?
None of this matters if your emails land in spam. Send from separate, authenticated domains, warm up new inboxes, cap daily volume per inbox, verify every address and keep spam reports far below the 0.3% ceiling in Google's email sender guidelines. The full setup is in the cold email deliverability guide.
Follow the law where your recipients are. In the US, the CAN-SPAM Act applies to business email: no misleading headers or subject lines, a valid postal address, and an opt-out you honor within 10 business days. In the EU, prospecting usually relies on legitimate interest under GDPR, which takes a documented balancing test, and recipients can object to direct marketing at any time. National e-marketing rules add conditions that differ by country.
Reach companies with a reason to buy this week
Startories finds the buying signal, verifies the decision-maker and runs the outreach until they book a call.
How should you handle replies?
Replies are where outbound turns into pipeline, and where many teams lose it by answering slowly. Answer positive replies within hours on business days, propose two specific times or a booking link, and confirm the agenda in one line. Startories sorts every reply into categories like the ones below and suggests an answer that you can approve before it goes out.
| Reply | What it means | What to do |
|---|---|---|
| Positive | Wants to talk | Book the call the same day with a calendar link and a one-line agenda |
| Interested, not now | Good fit, wrong timing | Ask when to follow up and set a dated task |
| Question or objection | Engaged but unsure | Answer in two or three lines, then offer the next step again |
| Referral | You reached the wrong person | Thank them and write to the person named, saying who sent you |
| Negative or unsubscribe | No | Stop at once and suppress the contact in every campaign |
| Out of office | Away | Pause and resume after the return date |
Answers to the four objections you will hear most
- "We already use {competitor}." Do not argue. Ask about the one gap your signal pointed to: "Makes sense. Is {the problem they posted about} solved there now? If so, I will leave it."
- "Send me some information." Often a polite exit. Send one short, specific item, never a full deck, and ask one question that shows whether there is a real need.
- "Not a priority right now." Ask when it might be, set a dated task, and stop the sequence. Do not move them into another campaign without asking.
- "How did you get my email?" Answer plainly: where you saw the signal, with the link, and that the address was found and verified for business outreach. Offer to remove them, and do it at once if they say yes.
Which metrics matter, and what does a weekly review look like?
Measure outcomes, not activity. Sends say little, and open rates are distorted by privacy features that load images automatically. Track these per campaign, per segment and per signal:
Check health first (10 minutes)
Bounces, spam reports and reply rate by inbox. Pause anything that looks wrong before you look at anything else.
Read every reply from the week (20 minutes)
The actual words, not a summary. Objections and "not relevant" replies point to the targeting rule or the sentence to change.
Compare segments and signals (15 minutes)
Which trigger, industry or size band produced meetings? Shift next week's volume toward it.
Change one thing
One new subject line, one new signal or one tighter filter per week. Change everything at once and you learn nothing.
Log it
Keep a running note of what changed and what happened. After two months it is your playbook.
- Positive reply rate: positive replies divided by contacts reached.
- Meetings booked and held: the number that pays for the program.
- Bounce rate and spam reports: the health of your sending domains.
- Pipeline and win rate: whether the meetings are with real buyers.
- Cost per meeting: total monthly cost divided by meetings held. The SDR cost guide shows how to compare it with a hire.
Worked example: a 90-day outbound plan for a five-person SaaS
The company is invented and every number is an assumption to replace with your own. Assume a five-person company sells uptime monitoring and public status pages to B2B SaaS companies. A typical contract is $3,600 a year. The founder runs sales and can give outbound about six hours a week. The goal: eight held meetings a month by day 90.
Days 1 to 14: foundations
Write the ICP and the one-page plan. Register the sending domains and start warm-up on day 1, because it takes the longest. Collect the first 150 companies from the three signals without contacting anyone, write the four-step sequence, and agree on what counts as a positive reply.
Days 15 to 30: soft launch
Send to 30 or 40 contacts a week. Approve every email by hand and read every reply. On day 30, compare the three signals: keep the ones that drew answers, and drop or rewrite the one that only drew "not relevant".
Days 31 to 60: full volume
Move to about 150 new contacts a month and change one thing a week. Add the LinkedIn note for larger accounts. On day 60, ask one question: does any signal produce positive replies? If none does, fix the offer before adding volume, since more sends of a weak offer only teach people to ignore you.
Days 61 to 90: double down or fix
Shift volume toward the best signal and test one new element, such as a second persona or a neighboring segment. On day 90, compute cost per meeting and decide: keep it founder-led, hand prospecting to an SDR, or automate it.
| Decision | What the team chose |
|---|---|
| ICP | US B2B SaaS with 10 to 100 employees and a customer-facing product; disqualify agencies and companies with a dedicated reliability team |
| Persona | CTO or head of engineering; the founder at companies under 30 people |
| Signals | Public complaints about a current monitoring tool; Product Hunt launches; a first DevOps or reliability job post |
| Offer | A free review of their public status page and alert setup, sent as five bullet points |
| Channels | Email first; a manual LinkedIn note only for accounts above 50 employees |
| Volume | About 150 new contacts a month from three warmed-up inboxes on two sending domains |
| Replies | The founder answers positive replies within four business hours |
| Review | Fridays, 45 minutes, following the weekly steps above |
Checking the plan against the founder's time
Six hours a week is about 26 hours a month (six hours × 52 weeks ÷ 12). At full volume, assume the weekly review takes 45 minutes (about 3 hours a month), replies take 15 minutes a working day (about 5 hours), and eight meetings take 45 minutes each with preparation (6 hours). That leaves about 12 hours a month for research, writing and approvals on 150 contacts, or under five minutes per contact. That is tight. Either the research gets faster, the signals do some of it, or volume drops. Run the same check on your plan before you commit to a number.
What are the most common outbound mistakes?
- A broad target. "B2B companies with 10 to 500 employees" is a database filter, not a strategy.
- Sending from the main domain. One bad campaign can push your invoices and support replies into spam.
- Fake personalization. A generated compliment about a company's "impressive growth" reads as automated and invites spam reports.
- Asking for too much, too soon. A 45-minute demo request in a first email to someone who has never heard of you.
- Follow-ups with nothing new. "Just checking in" three times teaches people to ignore your name.
- Judging by opens. Open data is unreliable; judge by replies and meetings.
- Stopping after two weeks. A new program needs several weeks of sending and weekly changes before the numbers mean much.
- Slow replies. A positive reply answered two days later may already have gone cold.
How to start your outbound program
Follow the first two weeks of the plan above: write the ICP, pick two or three signals, register sending domains so they can warm up, and fill in the one-page strategy table. Then send at low volume, read every reply, and run the weekly review from week four.
If you would rather not assemble the stack yourself, Startories runs the whole pipeline: signal detection, ICP checks, verified decision-makers, personalized sequences from warmed-up inboxes and reply classification, with reporting on meetings. Compare it with a hire in AI SDR vs human SDR, or see plans and pricing.
Frequently asked questions
What is an outbound sales strategy?
It is your plan for starting conversations with companies that have not contacted you: which companies to target and when, what to offer, how to write and send, how to handle replies, and which numbers you review each week to improve.
What is the difference between inbound and outbound sales?
In inbound, the buyer finds you first, through search, content or referrals, and you respond. In outbound, you choose the company and start the conversation. Outbound gives you control over who you talk to and when; inbound brings buyers who already know they have a need.
How many emails should an outbound sequence have?
Three to five emails over two to three weeks is a sensible default for cold B2B email. Each step should add something new, such as proof, a different angle or an easy way to say no. Stop as soon as the person replies.
Is outbound still worth it for small B2B companies?
It can be, when the target is narrow and the timing is right. Outbound lets you choose who you talk to and when, which inbound does not. It fits poorly when a customer is worth less than the cost of booking a meeting.
How long does outbound take to show results?
Allow a few weeks for domain setup and warm-up, then four to eight weeks of sending with weekly changes before you judge it. Replies tell you about targeting and copy long before you have enough meetings to measure.
Who should run outbound: a founder, an SDR or software?
The founder first, to learn what buyers say and which message works. After that, an SDR or an AI SDR can take over prospecting and follow-ups while a person handles the calls and closes the deals.