Illingworth says parallel enrichment cut a 12-hour Clay job to 30 minutes
The number is his own, on his own list, and the tool he is replacing is the one he sells an alternative to.
Richard Illingworth says he ran enrichment on 250 contacts in parallel inside Claude Code and finished in 30 minutes, against what he describes as a 12-hour Clay job on the same list. He posted the claim while giving away what he calls the full Claude Code stack he built to replace Clay, which he labels the most overrated tool in outbound. It is the same build he has been pushing all week, earlier framed as a swap for the Apollo, Clay and Instantly chain.
What the build actually is
He describes seven steps and five prompts in one loop. Define the ICP, build a reference-company list, pull Apollo filters, enrich in parallel, write the sequence, QC it, then feed post-send data back in. The guide he is offering includes the five prompts, a tier 1 and tier 2 bucketing logic that decides which contacts get hit first, split between sequencer-confirmed and broader signals, and an Instantly settings reference he gives as daily limit 15, open tracking off, Monday to Friday US East.
In a separate post he extends the same argument to the whole workflow. His framing is that most senders are stacking three or four SaaS tools that each cover about a third of the job, with Apollo on the list, Clay on enrichment, Instantly on sending and Lemlist sitting on top, held together by manual copy-paste between every step. His video version covers parallel Apollo scraping across hundreds of accounts at once, parallel enrichment without Clay or Zaps, sequence writing that pulls voice and offer from source material, and a post-send analysis loop he says tells you which variable moved the reply rate.
What is evidence and what is not
The 30 minutes versus 12 hours is one operator's number on one 250-contact list, self-reported, with no detail on what fields were enriched or how accuracy compared. He says he has sent millions of cold emails over the last couple of years and that enrichment was always the bottleneck, with credits ticking while the job ran. He also takes a shot at the output quality on the other side, saying his version avoids the personalization line that gets deleted on sight.
He is distributing the guide in exchange for a repost, and the thing being replaced is the thing he is selling an alternative to. For people sending, the useful part is not the time saving but the structure. Parallel enrichment plus a post-send loop is a workflow you can test against your current stack on one list before you cancel a seat.
same 250 contacts, done in 30 minutes.
none of the AI slop "i saw you went to [uni]" personalization that gets binned on reflex anyway
Know what is landing before you burn another domain
One email a day. Under three minutes. What the operators found out yesterday.