Signal based outbound for SaaS
The buying signals worth acting on, how to score and deduplicate them, and the five day sequence each signal type should trigger, with routing between marketing and SDRs.
On this page 7 sections
- What counts as a signal, ranked by strength
- Why champion job changes beat everything else
- Scoring and deduplication, so accounts get one play
- The five day play per signal type
- Routing and ownership between marketing and SDRs
- Measuring by signal source, then cutting what does not pay
- What to build first
- Frequently asked questions
The short answer
Signal based outbound replaces static prospect lists with a queue of accounts showing a recent, observable change: a champion changing jobs, relevant hiring, a tech install or removal, funding, product usage, or review site activity. Each signal triggers a short sequence whose first line names the signal. Champion job change plays convert to meetings at roughly three to six times the rate of a cold ICP list because the recipient already knows the product works.
Key points before you start
The difference between outbound that works and outbound that gets reported as spam is whether the recipient can tell why they were contacted today rather than any other day. That is the entire argument for signals. You are not finding better people. You are finding the same people at the moment something changed.
What counts as a signal, ranked by strength
Not all signals are equal, and the gap between the top and the bottom of this list is enormous. Rank yours before you build anything.
| Signal | Typical weekly volume | Strength | Decay window | Where to get it |
|---|---|---|---|---|
| Champion changes jobs | 2 to 10 | Very high | 30 to 90 days | CRM contact monitoring, Sales Navigator |
| Free plan usage spike | 5 to 25 | Very high | 3 to 7 days | Your own product analytics |
| Hiring for a role implying the problem | 10 to 40 | High | About 3 weeks | Job boards, Clay, company careers pages |
| Competitor tool removed | 1 to 8 | High | 2 to 4 weeks | Tech install data, BuiltWith style sources |
| Funding round announced | 3 to 15 | Medium | One quarter | Crunchbase, press |
| Review site activity in category | 2 to 12 | Medium | 1 to 2 weeks | G2 buyer intent |
| Competitor mentioned publicly | 5 to 20 | Medium | 1 week | Social listening, Common Room |
| Repeat content engagement | 15 to 60 | Low to medium | 10 days | Marketing automation, website analytics |
| Third party topic surge | 20 to 80 | Low | 2 to 3 weeks | 6sense, Bombora style providers |
Two things stand out. The strongest signals are also the rarest, which is fine because you want a small queue. And the weakest signal, third party topic surge, produces the most volume, which is precisely why teams over rely on it and then conclude signal based outbound does not work.
3x to 6x
Meeting rate on champion job change plays versus a cold ICP list
Aggregated outbound program reviews
Why champion job changes beat everything else
Someone who used your product at their last company and moved to a new one is the closest thing to a pre sold buyer that outbound produces. They do not need the category explained, they know whether the product worked, and in a new role they usually have 90 days of licence to change things.
Set this up by monitoring your CRM contacts for employer changes, which most enrichment tools handle automatically. Include churned accounts, lost deals where the champion liked you, and current customers’ users. That last group produces the most volume and is the one teams forget.
The first line that works
‘Saw you’ve landed at Ramp as Head of RevOps. You ran the rollout at your last company and I remember you pushed hard for the Salesforce sync to be built properly. Curious whether the same problem exists over there, or whether you inherited something cleaner.’ Nothing about your product. It references shared history, names a specific thing they cared about, and asks a question they can answer in one line.
Timing matters more than for any other signal. Week one they are drowning. Month six they have already picked tools. The 30 to 90 day window is where this converts, so build the delay into the queue rather than firing on the day the title changes.
Scoring and deduplication, so accounts get one play
Three rules keep a signal program from becoming the thing it replaced.
Score every signal on three dimensions: base strength from the table above, recency against that signal’s decay window, and fit against your ICP. Multiply rather than add, so a strong signal at a badly fitting account still scores low. An account with a fresh champion job change and perfect fit scores near the top; a two week old topic surge at a marginal account falls off the queue automatically.
Deduplicate at account level, always. One account, one active play, chosen by highest score. Then suppress that account from every other queue for 21 days. Without this, a company that raised a round, posted two relevant jobs and had someone browse G2 gets three emails from three people in one week, and you have taught them your company is disorganised.
And cap the queue. If more than 120 signals fire in a week for two SDRs, raise the score threshold rather than working the list faster. A signal queue worked badly converts no better than a cold list, and it burns the signals.
The dedup failure I see most
Marketing runs a nurture, SDRs run a signal play, and the ABM program runs ads, all against the same account, with no shared suppression. From the buyer’s side this reads as four unconnected approaches from one vendor in ten days. Build the suppression table before you build the second play, not after the complaint.
Editable working copy
Download this template
Save an editable working copy of the framework on this page. Add your own owners, evidence and decisions.
The five day play per signal type
Every signal gets a short, specific sequence. Five working days, three to four touches, then the account exits the queue and returns to nurture. Short sequences protect the signal’s freshness and stop you sending a fifth email about something that happened a month ago.
The standard five day shape
- Day 1, email naming the signal
One line on the observed change, one line on the problem it implies, one specific question. Under 80 words. No attachment, no link unless it is the proof.
- Day 2, LinkedIn view and a comment
Not a connection request with a pitch. View the profile and engage with something they actually posted. This makes the day 3 call a warm name rather than a cold one.
- Day 3, call with a voicemail
The voicemail repeats the signal in one sentence and says an email is waiting. Call connect rates roughly double when a same week email referenced something real.
- Day 4, the proof asset
Send the one piece of content that answers the implied problem. A teardown, a benchmark for their segment, a short Loom. Marketing should have this built per signal type in advance.
- Day 5, close the loop
A two line note that assumes no. 'Sounds like this is not the quarter for it. I will stop here, and if the on call problem resurfaces you know where I am.' This produces more replies than any other step.
- Exit and suppress
Account leaves the queue, gets suppressed for 21 days, returns to the nurture programme. Log the signal type against the outcome.
The day four asset is where marketing earns its keep. Each signal type needs its own piece: a hiring signal asset about what changes when the team crosses a size threshold, a funding signal asset about what breaks at the next stage, a competitor removal asset comparing migration paths. Building eight of these takes a quarter and then the SDR team stops improvising.
Routing and ownership between marketing and SDRs
Signals arrive from systems marketing owns and get worked by people sales owns, which is exactly the seam where programs fail.
| Responsibility | Owner | Success measure |
|---|---|---|
| Signal sources and integrations | Marketing ops | Signal freshness, false positive rate |
| Scoring model and thresholds | Marketing | Queue size versus capacity |
| Deduplication and suppression | Marketing ops | Accounts receiving more than one play |
| Per signal assets and first lines | Marketing | Reply rate by signal type |
| Working the queue | SDR | Touches within SLA, meetings held |
| Disqualifying a signal | SDR, with a reason code | Reason codes feeding the scoring model |
The SLA that matters: a fresh high scoring signal must be touched within one working day. A product usage spike touched on day six is not a signal, it is a list entry. Write the timing into your sales and marketing SLA with the decay window per signal type, because a generic ‘follow up within 24 hours’ clause does not capture that a job change can wait a week and a usage spike cannot wait an afternoon.
Where multiple people at one account start firing signals, you are no longer running outbound, you are running a group motion, and the coordination advice in buying group marketing applies. Treat that as a promotion into the ABM tier rather than more sequences.
Consultation request
Talk to a SaaS marketing strategist
Tell us about your marketing bottleneck and request a working session. We will confirm availability before scheduling.
Measuring by signal source, then cutting what does not pay
Report reply rate, meeting rate and opportunity rate separately for each signal type every month. This is the whole point of the structure, and most teams skip it because the CRM does not make it easy.
Typical pattern after two quarters of measurement: champion job change and product usage carry the program, hiring and competitor removal pay for themselves, funding is mediocre, and third party topic surge produces the most volume and the fewest opportunities. That last finding surprises people who bought an intent platform, which is why the honest assessment in intent data providers compared is worth reading before you renew.
Then act on it. Retire signals that produce under half the program average opportunity rate for two consecutive quarters. The temptation is to keep them because they fill the queue. A full queue of weak signals is worse than a half empty queue of strong ones, because your SDRs stop believing the queue means anything.
The honest cost of this approach
Signal based outbound needs roughly a quarter of setup before it produces anything: integrations, scoring, suppression logic, eight asset builds, and retraining a team that is used to working lists. It also produces less volume, which looks like a step backwards on every activity dashboard for the first two months. Teams that cannot tolerate that dip revert to lists by week six and conclude signals do not work.
What to build first
Do not build nine signals. Build two.
Start with champion job change, because it is free, strongest, and almost nobody runs it properly. Set up CRM contact monitoring, a 30 day delay, and one first line template. Then add whichever of product usage or hiring is easier given your data. Run both for six weeks, measure meeting rate against your existing list based outbound as documented in outbound demand generation for SaaS, and only then add a third source.
The examples in B2B SaaS demand generation examples show what the finished version looks like across several companies, and the account selection discipline overlaps heavily with account based marketing for SaaS. If you are still deciding whether signals belong in an inbound heavy or account heavy motion, the tradeoffs in inbound vs ABM will settle it, and the whole thing should be recorded in your demand generation plan template so the scoring logic outlives the person who built it. For where this sits in the overall mix, see SaaS demand generation.
One rule to leave with. Before any message goes out, the sender should be able to finish this sentence in one line: I am contacting this person today because. If they cannot, the message does not send.
Editable CSV worksheet
SaaS Demand Generation planning worksheet
A practical demand gen planning worksheet: decisions, owners, evidence and next actions.
Frequently asked questions
What are buying signals in B2B SaaS?
Observable changes at an account that make your problem more likely or more urgent right now. The common ones are a former user changing jobs, hiring for a role that implies the problem, a funding round, installing or removing a competing tool, unusual product usage on a free plan, activity on review sites, and repeated engagement with specific content. Each carries a different strength and decays at a different rate.
What is the best buying signal for outbound?
A champion job change. Someone who used and liked your product joining a new company arrives with the evaluation already done in their head, budget authority in many cases, and a desire to bring familiar tools with them. Meeting rates on this play routinely run several times higher than a cold ICP list. Monitor it through CRM contact changes and enrichment tools.
How fresh does a buying signal need to be?
It depends on the type. Job change plays work best in the 30 to 90 day window after someone starts, because day one is chaos and month six means they have already chosen tools. Hiring signals decay in about three weeks. Product usage signals decay in days. Funding signals hold value for a quarter. Set a decay rule per signal type and drop anything past it.
How do you avoid spamming one account with multiple signal sequences?
Deduplicate at the account level before anything sends. Give each account one active play at a time, chosen by the highest scoring signal, and suppress the account from other queues for 21 days. Without this rule, a company that raised a round, hired two engineers and installed a competitor gets three emails from three people in one week and reports you.
Do I need intent data to run signal based outbound?
No. The strongest signals are free: job changes visible on LinkedIn, job postings, funding announcements, and your own product usage data. Third party intent data adds account level topic surges, which are useful but weaker and noisier than the observable signals. Start with the free ones and add paid intent once the motion works.
How much volume does signal based outbound produce?
Far less than list based outbound, which is the point. A mid market SaaS company monitoring six signal types across a 2,000 account universe typically surfaces 40 to 120 qualified signals a week. That is a full queue for two SDRs working properly, and it converts at multiples of a cold list.
The saas-marketing.net editorial team Research and editorial
We research, write and maintain every page on this site. The library explains marketing decisions through practical frameworks, explicit assumptions and references. Corrections can be requested through the contact page.
Published September 11, 2026. Last updated .