research

PRAPI Research

Original research on PR, AI in journalism, AEO citation patterns, and the economics of source requests. Drawn from PRAPI's corpus of journalist queries and named practitioner contributions. Contributors are cited in full with a link back to their work.

Published reports

• 2026-07-03 · 12 contributors cited

  Running earned media across a portfolio: the system operators use to not drop the ball

  Managing earned media across multiple brands or clients demands a robust system to prevent missed pitches, follow-ups, and coverage tracking. Twelve practitioners shared their workflows, revealing common pitfalls like reliance on memory, scattered data, and dropped follow-ups. Successful systems combine a centralized tracker—often a shared spreadsheet or CRM pipeline—with disciplined weekly reviews and clear next-step actions. The biggest impact comes from treating pitches like sales leads, automating reminders, and enforcing accountability for follow-ups, transforming chaotic outreach into a scalable, reliable process.

  Read the report →
• 2026-07-03 · 8 contributors cited

  Keeping your outreach human when everyone is using AI

  Across eight practitioner responses, a clear consensus emerges: AI can efficiently draft outreach structures, but human input is crucial for personalization and credibility. Operators emphasize manually crafting the opener and referencing specific, recent details about the recipient to avoid generic, templated impressions. Common pitfalls include AI-generated generic openers, legal misstatements in regulated industries, and culturally off translations. Successful workflows combine AI's speed with human research and rewriting, preserving authenticity and driving significantly higher response rates.

  Read the report →
• 2026-06-26 · 7 contributors cited

  What operators do to get cited by ChatGPT, Perplexity, and AI Overviews (2026)

  We asked operators what actually got them cited by ChatGPT, Perplexity, and Google's AI Overviews, and what turned out to be snake oil. Seven who are showing up in AI answers responded on the record. The consensus: there is no GEO trick. You get cited when you are the clearest, most credible, most machine-legible source on a specific question, and ignored the moment you try to shortcut that.

  Read the report →
• 2026-06-23 · 8 contributors cited

  The B2B tools operators actually swear by (2026)

  We asked founders and operators a plain question: which B2B tools have become part of how you actually run your company, and which did you try and then quietly drop. Six operators answered. The pattern is consistent. The tools that earn a permanent seat remove a recurring operational task and get used every day, usually by collapsing a fragmented, manual workflow into one place. The tools quietly abandoned are the bloated all-in-one suites and clunky CRMs that looked complete in a demo but never earned adoption. For these operators, integration and daily use beat best-in-class features every time.

  Read the report →
• 2026-06-19 · 12 contributors cited

  How founders actually run their companies on AI (and what they dropped)

  We asked founders and operators a plain question: which AI has actually become part of how you run your company, what you use each one for, and which ones you tried and then dropped, and why. Eighty-nine people answered. The pattern was close to unanimous, and it cuts against the marketing. AI earns a permanent place in a real company when it removes routine load and sits close to the operator's own data and workflow. It gets cut, fast, when it tries to replace judgment or puts distance between the founder and the customer. Pavankumar Kamat of Panto AI said it cleanest: the AI that survives "shortens decision loops and preserves trust; anything that promised to replace judgment got cut fast." David Hunt of Versys Media drew the same line from the other side. The systems that stayed "augment judgment," he said, and the ones they dropped "tried to replace judgment outright." The kept stack is unglamorous: first-draft engines for internal text, orchestration that routes leads and removes human lag, meeting capture that turns talk into tasks, research that replaces fifteen open browser tabs. The discarded pile is remarkably consistent across very different businesses: generic copy generators that cost more to edit than to write from scratch, autonomous customer-facing bots that erode trust on contact, and bulk content engines that AI search quietly ignores. This report is the field guide founders wrote for each other: what to keep, what to kill, and the one filter that tells them apart.

  Read the report →
• 2026-06-19 · 7 contributors cited

  AI Assistants Are Already Sending Founders Real Customers (2026)

  Founders running real businesses can now point to specific customers who arrived because an AI assistant recommended them by name. This report collects first-hand accounts from operators who traced revenue back to citations in ChatGPT, Perplexity, and Gemini: the deals they closed, how fast those buyers converted, and what earned the citation in the first place.

  Read the report →
• 2026-06-19 · 4 contributors cited

  The AI Attribution Gap: Founders Cant Yet Measure What AI Sends Them (2026)

  The same founders who see their brands cited by ChatGPT, Perplexity, and Google AI Overviews mostly cannot prove those citations produced revenue. This report collects the honest null results and half-traced anecdotes from operators wrestling with stripped referrers, self-reported surveys, and the question of whether an AI citation is a sales channel or something harder to measure.

  Read the report →
• 2026-06-17 · 8 contributors cited

  AI-written pitches in 2026: the operator cut

  Every inbox in PR now carries AI-drafted pitches, and every serious recipient runs some kind of AI screen. We asked operators and founders who pitch (and who get pitched) what actually clears those screens. The pattern is consistent: AI that does the prep wins, AI that does the voice loses. This is the operator cut of our AI-written-pitches study, featuring practitioners who answered on the Connectively source-request board.

  Read the report →
• 2026-06-16 · 24 contributors cited

  AI can draft your pitch but it can't send it: what gets a pitch through, what gets it deleted in three seconds, and why the polish is now the tell

  We asked operators who pitch for a living, and who get pitched all day, a blunt question: when you use AI to write a pitch, cold email, or source-request answer, what actually earns a reply, and what gets ignored, filtered, or flagged? The answer was close to unanimous, and it inverts the promise of the tools. AI is a first-draft engine, not a send button. Raw AI output gets reply rates near zero, and the operators who win rewrite 60 to 70 percent of every draft before it goes out. Several tracked the gap precisely: one logged 58 percent reply rates on fully manual pitches against 22 percent on AI-first ones; another watched response rates fall from 40 percent to 5. The thing AI cannot supply is the only thing that earns a reply: a specific proof the recipient cannot get anywhere else, tied to their actual work. The sharper finding is on the receiving side. The people being pitched now spot AI in three to ten seconds, and the polish itself is the giveaway: flawless grammar with no voice, generic buzzwords, "catalog blindness" that proves the sender never read you. In an inbox full of machine-clean drafts, sounding like AI is now a negative signal. This report is the field guide both sides wrote: the tells that get you deleted, the specifics that get you through, and the hidden cost of letting the tool hit send.

  Read the report →
• 2026-06-15 · 35 contributors cited

  What AI engines actually cite for marketing and PR: why extractability and proof beat authority, and where earned coverage still wins

  We asked founders, PR pros, and operators a question every marketer is now quietly anxious about: when ChatGPT, Claude, Perplexity, or Google AI Overviews answer a marketing or PR question, which sources do they cite, and which do they ignore? More than sixty practitioners answered, and the result is not a single rule but a real argument with a clear center of gravity. The loudest finding: AI engines do not reward the most authoritative source or the most polished writing. They reward the most extractable and verifiable one, a clean, sourced, self-contained answer they can lift in the first sixty words and stand behind. That reshuffles who wins. Plain question-and-answer pages beat buried thought leadership, original data beats opinion, and a niche page with a checkable fact beats a famous brand's homepage. But it is not unanimous, and the disagreements are the most useful part. High domain authority still concentrates citations in some categories. Community sources, especially Reddit, are rising fast. And in trust-sensitive fields like health, legal, and finance, AI reaches for a narrow shortlist of credentialed and earned third-party sources, while brand content stays nearly invisible no matter how good it is. Read this as a field guide to what actually gets cited, organized by the forces that decide it.

  Read the report →
• 2026-06-14 · 35 contributors cited

  Share the machine, not the voice: what multi-brand operators build once, and the one thing that backfires every time they reuse it

  We asked operators who run marketing across more than one brand a single question: what is the one thing you built once that now pays off across every brand, and what backfired when you tried to share something that shouldn't be shared? More than fifty operators and agencies answered, and the responses converged on a split so consistent it reads like a law. The operators who scale across a portfolio build the **infrastructure** once (a shared measurement layer, a repeatable diagnostic, a launch-and-ops playbook, a tagged content and proof library) and reuse it hard. And almost every one of them, unprompted, named the same backfire: they tried to share voice, positioning, or creative across brands, and it blurred the brands, tanked engagement, or got flagged as duplicate content. The reusable thing is the structure. The thing you have to rebuild every time is the message. Read this as a field guide to what compounds across a portfolio and what quietly destroys it: the systems worth building once, organized by type, each paired with the operator's own account of what it replaced, and the cross-brand shortcut that failed.

  Read the report →
• 2026-06-11 · 26 contributors cited

  The road to $1M ARR: the channels operators built, the ones they killed — and the one variable that decided which won

  We asked 38 founders and operators a deliberately blunt question: which distribution channels actually drove revenue on the way to $1M ARR, and which did you try and drop? They answered through open source requests — a self-selected group, not a representative survey (see the methodology note). The clearest finding isn't a winning channel. It's that **there isn't one.** The same channel that printed money for one operator was a money-pit for another — SEO carried 80–90% of customers for some and converted *under 0.3%* for others, who dropped it. What decided the outcome was the *match* between channel and business: buyer intent, model, and sales-cycle length. High-intent search businesses compounded on SEO; complex or novel B2B got to revenue on outbound and trust-transfer while content went ignored; relationship and local-services businesses ran on referrals; DTC brands scaled on creator-affiliate and community, not paid-first. Read this as a segmented field guide — the mechanics operators used to build each channel, and the expensive lessons behind what they killed — rather than a leaderboard.

  Read the report →
• 2026-06-03

  What breaks first when you run marketing across multiple brands

  More than two dozen operators who run marketing and PR across multiple brand identities answered one question: what breaks first as you add brands, what you automated, and what stayed manual. They came from multifamily portfolios, hotel groups, DTC and supplement brands, e-commerce, SaaS, logistics, biotech, and agencies running dozens of client brands at once. They disagreed on almost nothing. Voice consistency breaks first. The thing that scales is the system, not the headcount. And the operators who last centralize one thing and decentralize everything else.

  Read the report →
• 2026-06-03 · 12 contributors cited

  The five patterns that landed B2B coverage in Q1 2026

  31 B2B operators across nine industries shared what pitch tactics actually landed coverage in Q1 2026. Five patterns emerged. Three counter-patterns stopped working. One signal worth flagging for the next cycle.

  Read the report →

What's next

The Q2 2026 research arc focuses on AI in journalism: how LLM citation patterns are shifting which sources get surfaced, what the spam-filter corpus says about AI-generated pitches getting through, and how practitioners are adapting their workflow since 2024. Open briefs feed into these reports.