The stack
The problem
Competitive intelligence in most companies lives in one person's head and goes stale the moment they get busy. Someone finally notices that a competitor launched a feature or changed pricing weeks after it happened, usually because a deal was just lost over it and the rep got blindsided on a call. Marketing reacts late, sales has no current talking points, and the 'battlecard' was last updated two quarters ago. The intel exists; it just never reaches the people who need it in time to use it.
The information is public and continuous. Competitors post on their blogs, update changelogs and pricing pages, announce funding, hire visibly on LinkedIn, and get written about in the press. The problem has never been access; it is that the collect-filter-summarize loop is a tedious recurring chore that competes with everything else on a marketer's calendar, so it reliably slips. And a chore that slips is a chore that does not happen. This is exactly the kind of well-defined, repeatable research task that automation handles better than a human ever will, because the automation never gets busy.
An automated pipeline runs on a weekly schedule: Perplexity gathers fresh, cited, time-bounded developments for each competitor; Claude filters out the noise and summarizes only what matters to GTM with a concrete 'so what' for sales and marketing; and the digest posts to a Slack channel. The team gets consistent, current intel without anyone owning the chore, and reps walk into calls actually knowing what changed this week.
The opinion that determines whether anyone reads it: ruthlessly cap the length and force a 'so what' on every item. An exhaustive two-page digest is an unread digest. The job is not to report everything that happened; it is to surface the three things that change a sales conversation this week and say exactly why. A digest that respects the reader's time gets read; one that dumps every press mention gets muted in a week.
How it works
- Maintain a list of real competitors, tiered, plus the specific questions you care about each week
- On a weekly schedule, query Perplexity per competitor with a tight, time-bounded prompt for the last 7 days
- Pass all the raw findings to Claude to filter noise, rank by GTM impact, and add a 'so what' per item
- Post the formatted, length-capped digest to a dedicated Slack channel automatically
- Cross-post the 'what to tell reps' section to the sales channel where reps actually are
- Tune the prompts for signal-to-noise over the first few weeks, then let it run hands-off
See it run
The playbook
Define competitors and the questions that matter
List your real competitors and write the specific questions you want answered every week: pricing or packaging changes, new features or product launches, funding and headcount moves, notable customer wins or losses, messaging or positioning shifts, and leadership changes. Being specific is what keeps the digest focused on GTM-relevant moves instead of drowning in every press mention and SEO blog post a competitor publishes.
Tier your competitors. Group them by how often you actually compete with and lose to them, so the digest can lead with the ones that matter most and not give a fringe competitor equal weight to your primary rival. The tiering also lets you query primary competitors more thoroughly than the long tail.
Write down your key battlegrounds, the specific dimensions you win and lose on, because Claude will use these to judge what counts as material. 'We compete on multi-region planning depth and lose on entry price' tells the model exactly which competitor moves to elevate and which to drop.
TipExplicitly add 'changes to their pricing page or free tier' to the watch list. Pricing moves are the intel reps need most and the easiest to miss, because competitors rarely announce a quiet pricing change but it shows up immediately in your deals as new objections.
Build the scheduled research step
In n8n (or Make, or Zapier), create a workflow on a weekly schedule trigger, for example every Monday at 7am. For each competitor, call the Perplexity API with a tightly scoped, explicitly time-bounded query so you get only the last week's developments, each with a citation. Perplexity's citations are what make the digest trustworthy and let any reader click through to verify rather than taking the summary on faith.
Loop over your competitor list so each competitor gets its own focused query rather than one vague catch-all that returns mush. In n8n, a 'Split In Batches' or loop node over the competitor list, calling Perplexity once per competitor, collects clean per-competitor findings you then pass on together.
The explicit 'last 7 days only' framing is load-bearing. Without it, Perplexity happily returns the competitor's biggest news from two years ago, which floods the digest with stale items and trains the team to ignore it. Bound the time window hard in the prompt.
What has {{COMPETITOR_NAME}} ({{COMPETITOR_URL}}) done in the LAST 7 DAYS only? Cover only items from the past week, each with a source link and a date:
- Product launches, feature releases, or changelog updates
- Pricing or packaging changes (including any free-tier change)
- Funding, acquisitions, or major hires / leadership changes
- Notable customer announcements, partnerships, or marketing campaigns
- Public positioning or messaging shifts
If nothing material happened this week, respond with exactly: 'No material developments this week.' Do NOT include older news to fill space. Do not speculate.Filter and summarize for GTM impact
Pass all the per-competitor Perplexity outputs to Claude in a single step with a prompt that ranks items by how much they matter to YOUR GTM team and writes a concrete 'so what' for each: what it means for sales conversations and what marketing should consider doing. This is the step that turns a news dump into actual intelligence. Tell Claude to drop anything immaterial outright and to omit competitors that had a quiet week entirely, so the digest stays short enough that people actually read it.
Have it preserve the source links from Perplexity all the way through to the digest, so every claim stays clickable and verifiable. Intel nobody can check gets either ignored or, worse, repeated wrong in a deal.
Cap the total length hard in the prompt, around 350 words. The length cap is not a nice-to-have; it is the single thing that most determines whether the digest survives past month one. A tight digest gets read every week; a comprehensive one gets muted.
You are a competitive intelligence analyst for our GTM team. Below are this week's raw findings per competitor, with sources.
{{ALL_PERPLEXITY_OUTPUTS}}
Our product (one line): {{ONE_LINE}}.
Our key battlegrounds (where we win/lose): {{KEY_BATTLEGROUNDS}}.
Produce a weekly digest:
- Include ONLY material items. Drop minor PR, routine blog posts, and anything immaterial. If a competitor had nothing material, OMIT them entirely.
- Rank items by impact on OUR deals: HIGH or MED.
- For each item, one line: Competitor | What happened (1 line) | Why it matters to us (1 line 'so what') | Source link.
- End with a section 'What to tell reps this week': 1-3 punchy, specific talking points reps can use on calls.
Keep the WHOLE digest under 350 words. Be specific. No fluff, no hedging. Banned words: unlock, leverage, supercharge, game-changer.TipCap the length hard and mean it. A 350-word digest gets read every week and shapes how reps talk; a two-page one gets skimmed once and muted. If you must choose, drop the medium-impact items, not the word cap.
Format and post to Slack
Send the formatted digest to a dedicated Slack channel like #competitive-intel using n8n's Slack node or an incoming webhook. Use Slack formatting so it is scannable: bold competitor names, the 'so what' line clearly distinct, and clickable source links. Post at the same time every week so the team learns to expect it and builds the habit of reading it, which is half the battle for any internal digest.
Cross-post the 'What to tell reps this week' section to the actual sales channel, not just the intel channel. Reps live in their team channel, not in #competitive-intel, so the talking points have to land where they already are or they will not be seen before the next call.
Format the Perplexity source links as real Slack hyperlinks rather than raw URLs so the digest stays clean and the verify-this-claim path is one click. A wall of bare URLs makes even a tight digest look like spam.
Add a battlecard update loop (optional)
Extend the workflow so that a material competitor move triggers a suggestion to update the relevant battlecard. When a competitor changes pricing or ships a feature you compete on directly, have Claude draft the revised battlecard section and post it for the product-marketing owner to review and approve. This keeps competitive collateral current continuously instead of letting it rot between quarterly reviews, which is when battlecards usually go stale and reps quietly stop trusting them.
Keep a human approval step on battlecard edits. The digest can flag and draft, but a person who owns positioning should approve the change before it becomes the official rep-facing guidance, because a competitive response is a positioning decision, not just a fact update.
Store the battlecards somewhere reps actually use, and link to them from the weekly digest so the update and the source intel travel together.
Tune the signal-to-noise, then run it hands-off
For the first few weeks, read the digest critically and tighten the prompts. If too much noise gets through, sharpen the 'material only' instruction and add explicit examples of what to drop. If it is missing things, broaden the Perplexity query or add specific sources to watch. This calibration period is short but matters, because a digest that cries wolf in week one never earns the standing read.
Sanity-check the high-impact items by clicking the source yourself during the tuning period. AI summaries occasionally misread a headline, and you want to catch that pattern before a rep repeats a wrong claim in a deal.
Once tuned, it runs hands-off and needs only an occasional refresh of the competitor list as the market shifts. Revisit the list quarterly so you are not still tracking a competitor that pivoted away or missing one that just emerged.
What you get
A scannable weekly Slack digest of only the material competitor moves, each with a 'so what' line and a source, plus rep talking points.
*Weekly Competitive Intel, Jun 10*
*HIGH IMPACT*
• *Northstar* | Launched a usage-based pricing tier starting at $0/mo | *So what:* undercuts us on entry price, expect price objections on SMB deals this week; lead with total cost at scale, not sticker price. | <link|source>
• *Northstar* | Hired a VP Sales from a logistics incumbent | *So what:* signals a deliberate push into our core vertical; flag named accounts where we're competing. | <link|source>
*MED IMPACT*
• *Apex* | Shipped an API for bulk data exports | *So what:* closes a gap we used to win on; stop leading with exports, reframe to our multi-region planning depth. | <link|source>
*What to tell reps this week:*
1. Northstar's $0 tier is bait, it caps usage hard. On any Northstar deal, get to total cost at their actual volume fast.
2. We still win vs Apex on multi-region planning depth; their new export API does not touch that.
3. If you're in a logistics deal against Northstar, expect a more aggressive sales motion now that they've hired into the vertical.
_(Competitors with no material news this week omitted.)_Pitfalls to avoid
Including everythingAn exhaustive digest is an unread digest. Force 'material items only,' omit quiet competitors entirely, and hold a hard word cap so the thing stays genuinely useful and survives past the first month.
No sources carried throughIntel nobody can verify gets ignored or, worse, repeated wrong in a live deal. Always carry the Perplexity citations through to the digest as clickable links so any claim can be checked in one click.
Not bounding the time windowWithout an explicit 'last 7 days only' instruction, Perplexity floods the digest with stale, big-but-old news to fill space, which trains the team to ignore it. Bound the window hard in every query.
Stale competitor listNew competitors emerge and old ones pivot away or fade. Revisit the list quarterly, or the digest faithfully tracks the wrong companies while missing the rival you actually started losing to.
Treating the summary as gospelAI summaries can misread a headline or overstate a move. For high-impact items, a human should click the source before a rep acts on it in a deal, treating the digest as a fast pointer rather than verified truth.