On this page
The cheapest pipeline you will ever find is already in your CRM. Someone at your company spent weeks on those deals. They had a champion, a quote and a reason to say no. Then the deal went to closed-lost and nobody looked at it again.
Most companies treat closed-lost as a graveyard. I think of it as a list of people who already know who you are.
Here is how we run it.
Why is closed-lost the first place to look for pipeline?
Because those buyers already know you, and because you do not need new spend to reach them. At InCheck, a funnel diagnostic showed a $3M goal needed about 350 deals, against about 120 the year before. That gap is hard to close with new leads alone. The company also had about 500 closed-lost deals sitting in its CRM from the last 5 years.
The reactivation run that followed produced 25 replies from the first 162 enrollments, with no new spend. That is a small batch and a modest number, and that is the point. You can start with a few hundred accounts and learn something real in a few weeks.
At AW Lake, a flow meter manufacturer, the opportunity was bigger and messier. Quotes moved into their ERP, Epicor, and never came back to the CRM. A closed-lost campaign was scoped on about 8,000 past quotes, and a follow-up showed about 5,000 Epicor quotes as the top priority. Years of work, invisible to sales.
Clean, check for a signal, reach out, hand off
Do you need to clean the list before you send anything?
Yes. Clean first. The fastest way to hurt your brand is to email the wrong person about the wrong deal.
At Aduro, the team was nervous about emailing former customers, and they were right to be. The fix was not courage. It was cleaning before importing and before any outreach.
Here is what to check before a single message goes out:
- Is it really lost? Remove accounts that are now customers, have an open deal, or are tied to a partner.
- Is the loss reason there? If reps typed "no budget" or left it blank, you will need to rebuild it from notes and calls. A message that fits why you lost will outperform a generic one.
- Is the contact still there? People change jobs. Validate emails and find who holds the role now.
- Is the quote history in the CRM? At AW Lake, quotes lived in the ERP. Bring them back, or you will not know what you offered, when or at what size.
- Is the account in your ICP today? Your ICP may have changed since the deal was lost. Some of those accounts no longer belong on the list.
None of this is exciting. It is the reason the campaign works.
How do you decide which lost deals are worth reopening?
Look for a fresh reason to talk. A lost deal with no new information is a cold lead with a long memory. A lost deal with a new signal is a warm one.
Signals we use:
- Website behavior. At AW Lake, datasheet and manual downloads on the site were treated as buying signals. A person who loses a deal and then downloads a manual is telling you something.
- People changes. A new owner of the decision, or the champion changing company, is a reason to reach out.
- Company changes. Hiring, funding, expansion or a new location can mean the problem is back.
- Fit. A lookalike of your recent wins. At InCheck, we looked at accounts that resembled the deals they had won.
You can run more than one at once. At Elastic Path, we re-ran 4,887 old ICP accounts through a two-signal check, and 543 were worth targeting. Fewer than one in eight. The rest would have been noise.
A lost deal with a new reason to talk is a warm lead
| Signal | What it can mean | What to do |
|---|---|---|
| Datasheet or manual download | Someone is researching again | Reach out with a relevant offer |
| New decision maker | The old no may not hold | Introduce yourself and the last conversation |
| Hiring, funding or expansion | The problem may be back | Reference the change |
| Looks like a recent win | Good fit on paper | Add to the next small batch |
| No new signal | A cold lead with a long memory | Wait and check again |
A note on tools. You can do the first pass in a spreadsheet. At volume, a data workflow that enriches each account and checks for signals will save weeks. Pick the tool after you know the rules.
What should the message say?
It should say why you are writing now, and it should fit why you lost. Not "just checking in."
A few patterns that hold up:
- Lost on price. Say what has changed on your side, or ask what budget looks like now.
- Lost to a competitor. Ask how it is going. Do not trash anyone.
- Lost to no decision. Offer a smaller first step.
- Champion left. Introduce yourself to the new person and say what the last team was trying to do.
- Lost on timing. Reference the date they gave you and ask if it is still right.
Keep it short and keep it honest. If you do not have a reason, do not send it.
Do not limit yourself to email. Surround the right accounts: email, LinkedIn and ads, all at once, when the signal says they may be ready. That is what we call allbound, and it works better than spraying a list.
Who handles the replies?
A person, quickly. Reactivation replies are warm, and warm replies go cold fast.
At TVEyes, we saw inbound leads waiting for hours, and sometimes days, because two people both assumed the other had them. The same thing happens with reactivation replies. Name one owner before you send. Then route replies to them the moment they arrive.
For the first touches, an AI agent can write the draft and a person can approve it. Let it send on its own only after you trust the output. For anything the buyer replies to, a human should take over.
How do you know whether it worked?
Count replies, meetings and reopened deals, not opens. And look at the first batch before you send the second.
Start with a small group, as InCheck did with its first 162 enrollments. Read the replies. Learn which loss reasons got answers and which got silence. Then rewrite the message and widen the list.
Track these:
- Replies per hundred sent. The first signal that the message and the list are right.
- Meetings booked. The first signal that the timing is right.
- Deals reopened, and their source. Mark each one so you can see what the campaign made, and keep the original loss reason on the record.
- Opt-outs. If these climb, slow down and fix the list.
Keep the numbers next to the original loss reason. After a few batches, you will know which kinds of lost deals are worth the work.
What to count after each batch
- Replies per hundred sentTells you the message and the list are right
- Meetings bookedTells you the timing is right
- Deals reopened, with sourceKeep the original loss reason on the record
- Opt-outsIf they climb, slow down and fix the list
What mistakes should you avoid?
- Blasting the whole list. Start small and learn.
- Skipping the clean-up. Old data wrecks the first impression.
- Ignoring why you lost. Generic messages get generic results.
- No owner for replies. The best leads will sit.
- Forgetting the lapsed trials and free users. At Chaos, we ran a pilot to re-engage lapsed trials, and used enrichment to find the ones with three or more seats of potential. Those were treated as land and expand opportunities.
- One and done. Reactivation is a rhythm. Come back every quarter with a new reason.
Where do you start this week?
- Export every closed-lost deal from the past few years.
- Remove customers, open deals and partners.
- Rebuild the loss reason for the top accounts.
- Pick one fresh signal and run the check.
- Send one small batch, name the owner for replies, and read what comes back.
Frequently asked questions
How far back should we go on closed-lost deals?
Go back as far as the data is good. InCheck had about 500 closed-lost deals over 5 years, and AW Lake had thousands of quotes. Older deals need more cleaning, since people change jobs, but they can still hold your best accounts.
Do we need new tools to run reactivation?
Not for the first batch. InCheck's first run, 25 replies from 162 enrollments, used no new spend. Once you want to enrich thousands of accounts and check signals at scale, a data workflow saves a lot of time, but choose it after you know the rules.
Should reps or marketing own the campaign?
Marketing can run the list and the sending, but a named sales owner should handle replies. Reactivation replies are warm and need a person quickly. The loss reason lives with sales, so ask reps to review the first batch of messages before they go out.
Is it okay to email people who said no?
Yes, when you have a fresh reason and a message that respects why they said no. The concern is real, as the team at Aduro showed. Clean the list first, remove customers and partners, cap daily volume, and stop at once for anyone who opts out.
Sources
- AW Lake discovery and follow-up, 2026: quotes that moved to the ERP and never returned, the closed-lost campaign on about 8,000 past quotes and the about 5,000 Epicor quotes follow-up.
- InCheck funnel diagnostic and reactivation run, 2026: the $3M goal, about 350 deals versus about 120, about 500 closed-lost deals over 5 years, and 25 replies from the first 162 enrollments with no new spend.
- Elastic Path, 2026: 4,887 old ICP accounts checked on two signals, with 543 worth targeting.
- Aduro and Chaos, 2025 to 2026: cleaning before outreach to former customers, and a pilot to re-engage lapsed trials.
- Letterdrop, When Should You Reopen a Closed-Lost Opportunity?: reopen only when buying activity has resumed, and log a task instead when it has not.