Grok Bot for presales: a battlecard the morning a competitor launches

You can give one Grok Bot a single beat: notice a competitor's launch and turn it into a battlecard before your next customer call. This playbook follows a solution architect whose Compete Bot builds battlecards and talk tracks, and whose noon report only speaks up when something matters. It fits solution architects, sales engineers, account executives and product marketers who get asked how they compare with a rival. It's a self-report in his own blog post, and we haven't run it.

Not tested: we haven't run a competitor-launch routine ourselves.

Who runs it#

Zach Woodward's X bio reads "Entrepreneurial Solution Architect @docusign solving real business problems." In a December 2025 post he describes himself "as a principal solution architect". In his Oct 2 post on Grok Bot, the work side sits under "Docusign day job."

Disclosure: He builds his own products. His X bio promotes TourneyHunter, a youth hockey tournament finder, and his Oct 16 post says "TourneyHunter and Compete.News came out of it last year", meaning Claude Code. He doesn't say whether Compete.News and his Compete Bot are related, so we don't link them. He writes as himself, not for Docusign, and we found no staff, partner or paid tie to Cursor, xAI or SpaceXAI on his blog or X profile.

The competitor loop#

He talks to one Bot: "I mostly talk to one bot. And that’s my chief of staff." It "hands work to about 15 specialist bots and brings the results back to me."

1. Four Bots cover the day job. "Compete tracks AI industry news and competitors, and builds battlecards and talk tracks. Cyber watches SaaS security vulnerabilities. AI Policy follows US AI regulation and legislation. CTO handles technical questions."
2. A launch turns into a battlecard. His example from recent weeks: "A competitor launched an AI product for legal teams. I had a battlecard the same morning." He doesn't name the competitor, show the battlecard or say what started the run.
3. A noon report filters it. "At noon on weekdays there’s a “water cooler” report. It pulls together what the research bots found and flags anything worth acting on. Most days it stays quiet, which is the point."
4. He decides. "It doesn’t coach my kids’ teams or run my deals. It gets me prepared faster and keeps the logistics from slipping."

A second work routine is plain logistics: "Every weekday morning a check copies my son’s school events from my personal calendar onto my work calendar. It was possible to get double booked before but now I don’t."

He doesn't list the connectors, sources or prompts his work Bots use. On X in September, describing Bots that watch Reddit, X and email for his hockey site and family, he wrote "I feel I have true virtual assistants" followed by "$200 a month" (post). He doesn't name the plan or say what the $200 covers.

His rules#

• "Nothing goes out under my name without my approval. The bots draft and I press send." His one exception is outside work: "the TourneyHunter brand X account, where I gave standing permission."
• "Each bot has a clear lane. The business email bot never touches my personal or hockey parent mail."
• "Quiet is a valid result. A bot with nothing new shouldn’t message me."

Why it matters at work#

Our read: in presales, a competitor launch matters most in the first calls after it, and that's when nobody has written it up yet. He names the gap himself: "A competitor launch I’d hear about a week late." One Bot with one beat, a fixed output (battlecard plus talk tracks) and a quiet rule closes that gap without adding another daily digest to read. He also ties the habit to customer work: "When I’m talking with our most sophisticated customers, the useful insights come from having tried it myself."

Set it up#

These steps are ours, built from his post and the docs. He hasn't shared his prompts.

1. One Bot for one beat. Cursor's docs say to create "a separate Bot when the work has a distinct goal, set of tools, working style, approval boundary, or recurring schedule" (docs). His Compete Bot is separate from his security and policy Bots.
2. Define the artifact. The docs say to "Specify the artifact and its acceptance criteria: a document with source links" (docs). Agree on the battlecard sections before the first launch.
3. Run it once by hand. "Start with a one-time task, make it reliable, save the method as a skill, and only then automate it" (docs). Try it on a launch you already know well.
4. Add the quiet report. His noon report is the only regular message. Cursor's chief of staff use case "Owns a source-linked digest of what changed and what needs your attention" (use cases).

Starter prompt (ours, untested)#

Your one job is competitive intel for [your product] against [competitor list]. Watch their newsrooms, blogs, release notes and pricing pages. When one ships or changes something, write a battlecard in [doc or folder]: what launched and when, who it's for, how it compares with us on [3 to 5 criteria], likely customer questions with a short talk track for each, and what we don't know yet. Link a source for every claim and mark anything you couldn't confirm. Never send, post or share it; tell me here with the link. At noon on weekdays, list anything new worth my time. If nothing changed, send nothing.

The battlecards, talk tracks, the noon report and the quiet rule are his. The source list, the battlecard sections, the source-link rule and the never-share rule are ours.

Gotchas#

• Drafts wait for you. "Email from a routine, an incoming message, or other work you didn't start with that request waits as a draft until you press Send" (Approvals help). That matches his rule, so plan for a click before anything reaches a rep or a customer.
• Test runs are real. "A test does real work" (Routines help). Point a test at a scratch doc, not the shared battlecard folder.
• Routines add up. "Each run spends usage, so a routine that runs often spends more than one that runs rarely" (Routines help). A daily noon report is cheaper than a watch that runs every hour.
• Lanes are not walls. On the Bots' computer, "browser sessions, files, and command-line credentials are shared" (docs), and "The screens are separate work surfaces, not separate security boundaries" (computer docs). He says as much himself: his Bots "share one computer, so files, logins, and playbooks live in one place" (his post). His lanes are a rule the Bots follow, not a lock.
• Your employer's rules come first. Our read: he runs this alongside a day job, and his post doesn't say what company data, if any, the work Bots see. Check your company's AI policy before connecting work accounts.

What to watch#

• Numbers. He gives no time saved, battlecard count or deal results. Watch his Substack for the first figure.
• The battlecard itself. He hasn't shown one, so we can't judge quality or sourcing. A shared example or template is the signal to look for.
• Changes. He writes: "I’m still adjusting the setup most weeks. More soon." A follow-up post would show what changed.

Related#

• Grok Bot for marketing: spend watches, reporting asks, and competitor briefs
• Grok Bot for founders and execs: an hourly attention list and a chief of staff that stays quiet
• Grok Bot for sales: a playbook for prospecting, sequences, and pipeline

Sources#

• Zach Woodward: How I Use Grok Bot: One Chief of Staff and a Team of Bots (Oct 2)
• Zach Woodward: I’m the human in the loop (Oct 16)
• Zach Woodward: Using Gong as a Solution Architect (Dec 2025)
• @zachwoodward on X: true virtual assistants, $200 a month (Sep 25)
• Cursor Docs: Work with Grok Bot
• Cursor Docs: Grok Bot use cases
• Cursor Help Center: Routines
• Cursor Help Center: Approvals
• Grok Bot docs: Computer and apps