In short
Source: Sales Surge, outbound implementations on PipedriveAs of: 16 November 2026
In this article
What does an outbound stack on Pipedrive look like?
An outbound stack on Pipedrive has five steps, and each tool does one job. Pipedrive is the record: every other tool writes back to it. What works is a system where data, the sequence and the CRM line up, so every reply reaches a person with context.
- Source. LinkedIn Sales Navigator, Apollo search, website visitors, event lists. Output: a raw list with name, company and role.
- Enrich. Apollo adds a verified email address, phone number and company data. Output: complete records you are allowed to mail.
- Reach out. Instantly sends the emails, Dux-Soup handles the LinkedIn steps. Output: replies, accepted invites, booked calls.
- Sync. Make.com writes every event back to Pipedrive. Output: the full history of a prospect in one place.
- Hand off. Pipedrive automations and a Slack alert put warm leads in front of sales. Output: qualified meetings in the pipeline.
The rule underneath: every step has to be measurable. If you cannot see how many contacts go from source to meeting, adjusting becomes guessing. What this guide does not give you is a list. Outbound starts with knowing who you want to talk to, and you know that better than we do.
What do you need to set up before your first cold email?
Before the first cold email goes out you need separate sending domains, SPF, DKIM and DMARC on each of them, two to three weeks of warm-up, and a one-click unsubscribe. Skip one of these and your mail lands in spam, however good the copy is.
Domains and mailboxes
- Buy two or three domains that resemble your main domain. Never send cold email from the domain your invoices and support mail come from.
- Create two or three mailboxes per domain, with a real name and a profile photo.
- Point each domain's website to your main site, so a curious recipient lands somewhere real.
Authentication
- SPF: says which servers may send on behalf of your domain.
- DKIM: signs every message, so the receiver can see it was not changed.
- DMARC: tells the receiver what to do when SPF or DKIM fails. Start with a monitoring policy, then tighten it.
Warm-up and volume
Warm up every new mailbox for two to three weeks before the first campaign, and keep the warm-up running afterwards. Start around 20 to 30 emails per mailbox per day and build up slowly. Most teams stay well under 50 cold emails per mailbox per day. More volume means more mailboxes, not a harder push on one.
The Google and Yahoo rules for bulk senders
Since February 2024 Google and Yahoo enforce rules for anyone who sends a lot of mail to their users. Google's threshold is 5,000 messages a day to Gmail addresses, but the rules are worth following at any volume, because they describe what inbox providers reward.
- Authenticate with SPF and DKIM, and publish a DMARC record.
- Offer a one-click unsubscribe in the message header and honour it within two days.
- Keep the spam complaint rate below 0.3 percent. Aim for under 0.1 percent.
- Keep your bounce rate under 2 percent by verifying every address before it enters a sequence.
Is cold email allowed under GDPR?
B2B cold email can be lawful under GDPR on the basis of legitimate interest, as long as you can show the balancing test, the message fits the recipient's role and every mail has a working unsubscribe. GDPR is not the only rule. National ePrivacy laws decide when unsolicited electronic messages are allowed, and they differ per country. In the Netherlands, for example, the rules are stricter for sole traders and personal addresses than for company addresses.
- Write down your legitimate interest and the balancing test once, per campaign type.
- Mail people in their professional role, about something that fits that role.
- Say in the first mail where you got their details.
- Put a working unsubscribe in every mail, and sync it to Pipedrive so no other tool mails that person again.
- Keep a record of where each contact came from in the Lead Source field.
Which fields do you need in Pipedrive?
You need four groups of fields before a single email goes out: where a lead came from, how good the contact data is, where the person is in your outreach, and how well they fit. Start with fewer fields than you think you need. An empty field is worse than no field, because nobody fills it any more.
| Field | Type | Values | Why |
|---|---|---|---|
| Lead Source | Single option | Apollo, LinkedIn, Inbound, Event, Referral | So you can tell later what each channel produced |
| Source Campaign | Text | Campaign ID or name | So you compare campaigns, not only channels |
| First Touch Date | Date | Set automatically | To measure how fast you respond |
| Email Verified | Single option | Yes, No, Unknown | Decides who may enter a sequence |
| LinkedIn URL | Text | Profile address | What Dux-Soup and Make.com match on |
| Outreach Status | Single option | Not contacted, In sequence, Replied, Bounced, Unsubscribed | So you know who is still in a sequence |
| LinkedIn Status | Single option | Not connected, Requested, Connected, Messaged | So you can follow Dux-Soup |
| Last Outbound Touch | Date | Updated automatically | So nobody gets contacted three times at once |
| ICP Fit | Single option | Strong, Good, Fair, Poor | So you start with the best fit |
How do you enrich leads with Apollo?
Enrich every lead in Apollo before it enters a sequence, and only mark an address as verified when Apollo is confident. Half-filled records mean bounces, and bounces hurt the domain for everyone.
When to enrich
- When new leads come in, before they go into a sequence.
- Once a month for existing contacts without an email address.
- Before you approach a large account, to see who works there.
Deduplication rules
- Match on email address first. That is your key.
- No match? Match on LinkedIn URL.
- When they disagree, Apollo wins on contact details and Pipedrive wins on history.
- Never merge automatically. Someone reviews the pair first, which is how Surge Dedupe works too.
Apollo does not scrape. It combines data that is publicly available, and you remain responsible for using it within GDPR. More on the Apollo side is on our Apollo page.
How do you build a sequence in Instantly?
Build a sequence of three to five touches over two to three weeks, with a first line that is about the recipient and a last email that says you stop. Whether it works depends on two things: does it arrive, and is it about something.
- Write the first line about the company or the person, based on something real.
- Keep each email short enough to read on a phone.
- Ask one small question, not for an hour of their time.
- Spread sending over the day instead of sending everything at 9:00.
- Test subject lines against each other with at least 100 sends per variant before you decide.
- Turn on stop-on-reply in Instantly itself, and do not rely only on the webhook.
Setting up Instantly with Pipedrive step by step is on our Instantly page.
How do you add LinkedIn with Dux-Soup?
Use Dux-Soup for the steps you would do by hand anyway, within LinkedIn's limits. It is an assistant, not a spam cannon. The line we hold: if you would not do it manually, do not automate it.
| Rule | What it means | Why |
|---|---|---|
| Weekly limit | Stay around 100 connection requests per week, so roughly 15 to 20 per working day | LinkedIn limits invites per week |
| Warm up | Start at 5 to 10 a day and build up over two weeks | A sudden jump looks like a bot |
| Own note | Always a personal line with the request, never the default text | More people accept, and it stays decent |
| Random pauses | 30 to 90 seconds between actions, never the same | Fixed intervals give automation away |
| View first | Visit the profile a day or two before you send the request | That is how a person works too |
| Narrow targeting | Use Sales Navigator filters so you only reach relevant people | Few good contacts beat many random ones |
Typical workflows
- Profile view, wait a day, request with a personal note, wait for the accept, first message.
- React to something the person posted, wait two days, view the profile, send the request.
- After an email reply, find the person on LinkedIn and connect with a reference to that email.
What Dux-Soup can and cannot do with Pipedrive is on our Dux-Soup page.
How do replies land in Pipedrive automatically?
Replies land in Pipedrive through a webhook from Instantly to Make.com, which finds the person, updates the status, stops the sequence and alerts the owner. Written out, the flow looks like this:
- Instantly detects a reply and sends a webhook to Make.com.
- Make.com searches Pipedrive for the person by email address, then by LinkedIn URL.
- Found: Outreach Status becomes Replied. Not found: Make.com creates the person and the organization.
- Make.com pauses every other sequence for this contact in Instantly.
- Make.com creates an activity on the person with the reply text, assigned to the owner.
- A positive reply creates a deal or lead in the outbound pipeline.
- Slack or email alerts the owner, with a link to the record.
Next to that flow, five more scenarios cover most setups. Start with three or four, and only build further when those run without trouble. Our Make.com page has more on how we build scenarios.
| Scenario | Trigger | What it does | Timing |
|---|---|---|---|
| Apollo to Pipedrive | New enriched contact | Search by email, create or update person and organization, log the enrichment | Real time or daily |
| Dux-Soup to Pipedrive | New LinkedIn connection | Match on LinkedIn URL, set LinkedIn Status to Connected | Daily |
| Deal moves, outreach stops | Deal reaches a qualified stage | Pause every sequence for the contact in Instantly | Real time |
| Bounces | Bounce in Instantly | Set Email Verified to No and Outreach Status to Bounced, flag for re-enrichment | Real time |
| Unsubscribes | Unsubscribe in any tool | Mark the person in Pipedrive and remove them from every sequence | Real time |
Which numbers do you measure, and what is normal?
Measure reply rate, positive reply rate, meeting rate, time to first follow-up and cost per meeting. What is normal differs too much per sector, list and offer to give one number. Set your own baseline in the first four weeks and steer on the trend, not on what a single Tuesday does.
| Metric | Formula | What it tells you |
|---|---|---|
| Reply rate | Replies divided by delivered emails | Whether the message lands and the list is right |
| Positive reply rate | Interested replies divided by all replies | Real interest, with the noise taken out |
| Meeting rate | Meetings booked divided by contacts in a sequence | What outbound adds to the top of the funnel |
| Time to follow-up | Hours from reply to first contact by sales | Whether sales picks up what outbound produces |
| Lead to deal time | Days from first touch to won deal | Whether the whole funnel moves |
| Cost per meeting | Total outbound spend divided by meetings booked | Where your next euro should go |
| Invite accept rate | Accepted divided by sent connection requests | Whether you target the right people on LinkedIn |
Instantly and Apollo publish benchmark reports with averages every year. Read them as context with the year attached, not as a target. Put these metrics on a Pipedrive dashboard and look at it once a week. For reporting across objects that Insights cannot do, there is Surge Reports.
Where does outbound usually go wrong?
| What you see | Cause | Fix |
|---|---|---|
| Bounces above 2 percent | Addresses not verified | Enrich and verify before sending |
| Hardly any replies | Targeting too broad, copy fits anyone | Narrow the ICP, rewrite the first line, test subject lines |
| LinkedIn restricts the account | Too much automation | Lower the volume, pause for two or three days, vary the timing |
| Spam complaints | People it does not concern, or no unsubscribe | Always an unsubscribe, sharper targeting, stop after the last touch |
| Sequence continues after a reply | Reply detection fails | Stop-on-reply in Instantly plus a weekly test of the webhook |
| Pipedrive fills up with junk | No filter at import | Score on ICP fit at import and keep poor fits out of the pipeline |
| Sales ignores the leads | Leads do not reach their day | A deal and activity on reply, plus a Slack alert |
| Duplicate contacts | No matching at sync | Match on email and LinkedIn URL, review duplicates weekly |
| Nobody knows what it produced | Lead Source not filled | No import without Lead Source and Source Campaign |
| Domain on a blacklist | Too much mail too fast from a new domain | Respect the warm-up, use separate outbound domains |
How are the tools in an outbound stack priced?
Each tool in an outbound stack has its own pricing model: per user, by credits, by emails sent or by operations. That gives five cost lines, the CRM, data, sending, LinkedIn and the glue between them, plus domains and mailboxes. How each line is priced matters more than today's list price, because your volume drives the bill.
| Tool | Role | How it is priced | Check at |
|---|---|---|---|
| Pipedrive | CRM and record | €14 to €79 per user per month, billed annually | pipedrive.com |
| Apollo | Data and enrichment | Free plan, paid plans per user with credits | apollo.io |
| Instantly | Sending and warm-up | Plans by emails and contacts per month | instantly.ai |
| Dux-Soup | LinkedIn steps | Per user per month | dux-soup.com |
| Make.com | Sync between tools | By number of operations per month | make.com |
| Domains and mailboxes | Sending infrastructure | Per domain per year, per mailbox per month | your registrar and mail provider |
Prices change, so check them on the day you decide. For high volume with your own deliverability and a direct sync to the CRM, there is also Surge Reach, our bulk mail tool with domain warm-up and CRM sync. It replaces the sending part of the stack, not your targeting.
Should you build outbound yourself or outsource it?
Build it yourself if you have someone with time for domains, Make.com and Pipedrive, and the patience for a warm-up. Outsource it if you want the whole funnel standing in one go, with the integrations tested. We build outbound funnels from €9,500.
How we work
- Audit: what is there now, how clean the data is, and what the copy says.
- Blueprint: the flow from source to enrichment, sequence and CRM, on one board.
- Build: domains, Pipedrive fields, Apollo, Instantly, Dux-Soup and the Make.com scenarios.
- Tune: we watch the numbers with you and adjust where it gets stuck.
Sales Surge has delivered more than 150 CRM implementations. Sander Vergouwen (co-founder) and Bart-Jan Hin (co-founder, more than 20 years in sales, CRM and RevOps) work on the projects themselves. We are a Pipedrive Platinum Partner and among the top 8 Pipedrive partners worldwide since July 2026, Apollo Fully Certified for 2025 and 2026, an Instantly Certified Expert, a Make.com Certified Partner and a Dux-Soup partner.
What a full funnel includes is on outbound sales funnel. The bigger picture of scaling outbound is in scale outbound sales, and the basics are in our outbound knowledge hub.
Glossary
- Cold email: an email to someone you do not know yet, meant to start a conversation.
- Deliverability: how much of your mail actually lands in the inbox instead of spam.
- Warm-up: building up volume slowly so a new domain earns a reputation.
- Sequence: a series of messages that go out automatically at set moments.
- Bounce rate: the share of your mail that does not arrive because the address does not exist.
- ICP: Ideal Customer Profile, the description of the customer you fit best.
- SPF, DKIM and DMARC: settings that prove to the receiving server that you are the sender.
- Break-up email: the last email in a sequence, saying you stop.
Frequently asked questions
Is cold email legal under GDPR?
B2B cold email can be lawful under GDPR on the basis of legitimate interest, if you have done the balancing test, your message is relevant to the role, you say where you got the data and every mail has a working unsubscribe. National ePrivacy rules sit on top of GDPR and differ per country. This is not legal advice, so have a lawyer check your setup and your list.
How many emails can I send per day?
Start low on a new mailbox, around 20 to 30 per day during warm-up, and build up over two to three weeks. Most teams stay well under 50 cold emails per mailbox per day. If you need more volume, add mailboxes and domains instead of pushing one mailbox harder. Watch bounces and spam complaints before you raise the number.
What if my domain gets blacklisted?
Stop all sending from that domain right away. Find the cause first: unverified addresses, a list nobody asked for, or too much volume too fast. Request delisting only after you have fixed it, otherwise you end up on the list again within a week. This is why outbound runs on separate domains and never on the domain your invoices come from.
Should I start with email or LinkedIn?
It depends on who you want to reach. Senior decision makers often respond better on LinkedIn, operational roles more often on email. Most sequences we build combine both: a profile view and connection request alongside the first emails. Test both channels on the same segment and compare the reply rate per channel in Pipedrive.
How long should an outbound sequence be?
Three to five touches over two to three weeks is what we see work most often. More touches rarely add replies and mostly add irritation and spam complaints. The last email says plainly that you stop. After that, a contact only comes back into a sequence when something real changes, such as a new role or a new trigger.
Is Dux-Soup safe for my LinkedIn account?
It is safe when you use it the way a person would work. Stay under LinkedIn's weekly invite limit, which in practice is around 100 connection requests per week, so roughly 15 to 20 per working day. Warm up slowly, add random pauses and always write your own note. Automation that you would never do by hand is where accounts get restricted.
Can I do this without Make.com?
Partly. Apollo and Instantly have native Pipedrive integrations that cover the basic sync. Make.com gives you control over the flow: matching on email first and LinkedIn URL second, stopping a sequence when a deal moves, and logging every step in one place. As soon as replies have to land in Pipedrive with context, we use Make.com.
How long does the setup take?
The basics are in place in 2 to 4 weeks: domains, warm-up, fields in Pipedrive and a first sequence. With every integration in place and tuned, it takes 6 to 8 weeks. The warm-up of new domains is the part you cannot rush. Tuning the messaging never fully stops, and that is fine.
What if sales does not follow up the leads?
Make the follow-up part of their day, not an extra task. A reply creates a deal or lead in Pipedrive at once, with an activity for the owner and a Slack alert. Then measure the time from reply to first contact and put that number in the team meeting. Without that number it stays a good intention.
What do I do with a not interested reply?
Respect it straight away. Remove the contact from every sequence, log the reason in Pipedrive and mark the person so no other campaign picks them up. No second attempt with a new angle. The reason field is useful later: if many replies say the same thing, your targeting or your offer needs work.





