Proposals that used to take 3 hours now take 40 minutes. Account research that was a half-day job is now done in 20 minutes. That's what AI-capable sales teams look like.
Train Your Sales Team →Most sales training focuses on skills: objection handling, discovery questions, closing. That's still important. But there's a growing gap between the sales team that can do the skills and execute the admin fast — and the team that can only do one.
The admin isn't small. Proposal writing, account research, follow-up drafting, call summarisation — these aren't edge cases. They're the majority of a salesperson's non-meeting hours. When you cut the time on those tasks by 60–70%, you don't just give people their time back. You change what they can actually pursue in a week.
Here's what the before/after looks like across the most common sales tasks:
Before: 2–4 hours per proposal, starting from a previous template, manually customising each section. After: 35–45 minutes. Call notes go in, a structured draft comes out, the rep edits and personalises. The thinking is still theirs. The blank page is gone.
Before: 45–90 minutes on LinkedIn, company website, Google News, and maybe a competitor comparison. After: 15–20 minutes. AI synthesises company background, recent news, product positioning, likely pain points — structured for a meeting prep document, not a raw search dump.
Before: Each follow-up written manually, often similar but not consistent, delayed by the time it takes to write. After: Draft from meeting notes in under 5 minutes. The rep decides what to adjust. Every follow-up goes out faster and with more specificity to what was actually discussed.
Before: CRM notes incomplete or written hours after the call when memory has faded. After: Call notes pasted in, AI produces structured CRM update — next steps, key objections raised, decision-maker status, follow-up date. 5 minutes, not skipped.
Before: Informal, inconsistent, often based on outdated knowledge. After: Structured competitive comparison produced against specific competitor, including positioning gaps and the 3–5 likely objections the prospect will raise based on that competitor's messaging.
Before: Long call notes that nobody re-reads, or no notes at all. After: Structured summary: what was agreed, what was raised as a concern, open questions, recommended next steps. Shareable with the manager, usable in the next meeting, actioned the same day.
This isn't an AI literacy seminar. It's not a showcase of tools. The training is structured around the actual tasks your sales team does every week — and we build working AI workflows for each one during the session.
The primary tool is Claude Cowork — Anthropic's Claude in a shared, multi-agent workspace that allows sales reps to build reusable prompt templates, iterate on outputs together, and deploy workflows their whole team can access. No API keys. No coding. No IT dependency.
Here's what we cover in a standard sales team programme:
We are not promising CRM replacement or automation of the entire sales process. What we deliver is real workflow augmentation — the tasks that take the most time and produce the most inconsistency, made faster and more consistent without losing the human judgment that makes sales work.
Every ANCHR AI Labs programme ends with a Build Challenge: a structured, timed exercise where participants build something they can actually use in their work the following week. Not a case study. Not a simulation. A real artefact.
For sales teams, the standard Build Challenge is:
Take your actual CRM notes from a recent qualified call. Build a Claude Cowork prompt template that takes those notes as input and outputs a first-draft proposal in your company's tone — with the correct structure, the right level of personalisation, and the prospect's specific business priorities surfaced in the opening section. Target: first draft ready in under 5 minutes from notes to output.
We also run shorter sprint challenges depending on the team's priorities:
The format is individual work with group review. Participants see what their colleagues built, we discuss what worked, what didn't, and how to make each workflow more robust. The output is not only the template — it's the judgment about when and how to use it.
The programme is designed for non-technical sales professionals. You don't need to know anything about AI before you walk in. You do need to know what your actual sales workflow looks like — because that's what we're going to augment.
This works best for:
Works equally well for B2B product sales, enterprise software, professional services, financial services, and any industry where written materials and preparation are a significant part of the sales motion. We've run this with Singapore-based teams selling regionally across SEA, and with teams selling into global accounts from Singapore.
Minimum recommended group size: 6 participants. Maximum for full Build Challenge format: 24 participants. Larger groups available with adjusted structure.
We'll discuss your team's current workflow, where the biggest time losses are, and what a programme looks like for your context. No obligation. Usually a 20-minute call.
Train Your Sales Team →AI produces a first draft. The salesperson edits, adjusts, and approves it before it leaves their hands. The personal touch isn't in typing — it's in what you decide to include, what you emphasise, and how you position your solution. AI removes the blank-page problem and the repetitive reformatting. It doesn't make decisions about the deal. That stays with the rep. The proposals that come out of AI-trained sales teams are often more personalised, not less — because the rep now has time to think about customisation rather than time to just hit send on a template.
This is the most common concern and it's a legitimate one. We cover data hygiene explicitly in the training — what to put in prompts, what to anonymise, and what should never go into an AI tool regardless of vendor. Claude Cowork uses Anthropic's API under the hood, and Anthropic has clear policies on data handling. That said, we train teams to use anonymous case frameworks, generic account descriptors, and redacted versions of documents when building prompt templates — so the template gets trained on the logic of your workflow without exposing live client data. Your legal or compliance team will need to sign off on the specific approach, and we're happy to speak with them directly if needed.
That's exactly who this is designed for. Non-technical is the baseline assumption. We don't cover APIs, model architecture, or anything that requires prior AI knowledge. The skills we teach are writing skills — specifically, how to brief an AI tool in the same way you'd brief a capable junior colleague. If someone can write a clear email, they can learn to write a clear prompt. The learning curve is real but short, and most participants are building working workflows within the first two hours of a full-day session.
For a standard sales team programme covering the full range of use cases — proposals, account research, follow-up, CRM — we recommend a full-day session (7 hours including breaks). For teams that want to focus on one or two specific workflows, a half-day (3.5 hours) works well. We don't run 1-hour AI taster sessions for sales teams — the Build Challenge requires enough time to actually build something useful, and that takes at least 3 hours to get right. Enquire about timing and we'll recommend the right format for your team size and scope.
Yes, and we'd strongly recommend it. The pre-programme briefing process includes understanding your sales cycle, your proposal format, your typical account profiles, and your competitive landscape. The Build Challenges are run using your actual materials wherever possible — your proposal templates, your CRM field structures, your follow-up sequences. The more we understand your context before the session, the more directly applicable the outputs are. Generic AI training produces generic results. ANCHR programmes are built around your workflow.