Name a competitor and Juma maps how AI answer engines describe them, finds where the narrative is beatable, and returns counter-content angles your brand can own.
Name the competitor and the category the brand wants to win. Juma researches how ChatGPT, Perplexity, and Google AI Overviews describe that competitor when buyers ask category questions: the claims that keep getting them recommended, the third-party sources the engines pull from, and the reviews shaping the story. The Flow then maps where that narrative is weak, overpromised, or incomplete, and pairs each gap with a content angle the brand can credibly own.
Ask Juma to isolate the gaps in how AI engines describe a competitor, and the output separates the claims that hold up from the ones that crack: features that get oversold, capabilities buyers say are missing, and the complaints that recur across reviews and community threads. Each weak spot comes with the source behind it and a read on how beatable it is, so the team can tell the difference between a minor gripe and a genuine opening. This is the layer that turns "they look strong in AI answers" into a specific, ranked list of where the narrative is soft.
Look at how AI engines describe HubSpot for CRM. Where is that narrative weakest: what's overpromised, what do reviewers complain about, and which gaps are most beatable for a challenger?
Once the gaps are clear, brief Juma to turn them into a content plan in the same chat. The output names the topics worth owning, suggests a format for each (comparison page, teardown, calculator, point-of-view post), and frames the angle that puts the brand on the right side of the gap. Because the plan is built from the competitor's actual weak points and the brand's real strengths, the topics are ones the brand can defend, not generic "write about X" suggestions. It connects the research directly to a publishing roadmap without a separate briefing step.
Based on the gaps we found, give me a content plan to take these topics from HubSpot. For each topic, suggest the format and the angle that positions our client as the better answer.
Name two competitors and Juma maps how AI answer engines position each one in the same category, side by side. The output shows where the two narratives overlap, where each is praised or criticized, and which topics neither competitor owns cleanly. Those unclaimed topics are usually the fastest openings for a third brand, because the engines have no settled answer to default to. This is useful for pitch prep, category mapping, or any moment where the team needs to see the whole competitive narrative before deciding where to plant a flag.
Compare how ChatGPT and Perplexity describe HubSpot and Salesforce for CRM. Where do the narratives overlap, where does each win, and which topics does neither own cleanly?
Ask which sources sit behind the competitor's narrative, and Juma maps the third-party domains the engines lean on: review sites, comparison guides, community threads, and publications. The output ranks them by how often they shape the answer and notes what each one says. That tells the team where the competitor's story is actually being written, which is where a counter-narrative has to show up to change it. The same map doubles as an earned-media target list: the review profiles to claim, the threads to join, the publications to pitch.
Which third-party sources do ChatGPT and Perplexity pull from when they describe HubSpot for CRM? Rank them by influence and tell me what each one says about HubSpot.
Point the same audit at the brand instead of the competitor to see the story the engines tell about it today. The output surfaces how AI answers describe the brand, which claims land, where the narrative is thin, and which sources the engines are reading. Run this before the competitor audit to set a baseline, or after it to see the gap between where the brand stands and where the competitor stands on the same topics. It turns the competitive picture into a clear read on the brand's own starting position.
Now run the same analysis on our own client. How do ChatGPT and Perplexity describe us for CRM today, which claims land, and where is our narrative thin compared to HubSpot?
A Juma Project is a shared space where the team stores everything Juma needs to know about a client's positioning work. Create one project per client, add context as the program develops, and Juma uses what's relevant every time the team runs a flow. The more the team adds, the sharper every audit and content plan gets.
Competitor Watchlist
The two or three competitors the brand most wants to displace in AI answers, each with a URL and a one-line positioning note. Juma uses this list to scope every audit against the brands that actually matter, so the "who do we compare against?" conversation disappears.
Category Query List
The 15 to 20 questions the audience asks ChatGPT, Perplexity, or Google AI Overviews in the brand's category. These anchor the audit to real buyer queries, so the narrative analysis reflects how people actually research, not a generic brand lookup.
Brand Positioning Note
Where the brand genuinely wins: the proof points, the differentiators, the topics it can defend. Juma uses this to make sure every counter-content angle sits on a real strength, not just a competitor weakness the brand can't actually claim.
Previous Audit Snapshots
Past positioning audits saved as reference. Juma compares new findings against earlier ones to surface shifts: a competitor patching a weak spot, a new review cycle changing the story, or a topic opening up that was locked before.
Add a short description to each knowledge item in the project's info field so Juma knows what each file contains and when to use it. For example:
A competitor content positioning audit is the offensive move: it analyzes how AI engines describe a competitor, finds where that narrative is weak, and plans content to take the topic. A GEO audit is the defensive move: it scores a specific page against the structural patterns AI engines reward and tells the team what to fix. One studies the competitor's story; the other improves the brand's own page.
Most teams use them together. The positioning audit decides which topics are worth owning by showing where a competitor is beatable. The GEO audit then makes sure the pages built around those topics are structured to actually get cited. Run the positioning audit to choose the fight; run the GEO audit to win it on the page.
The audit looks at how ChatGPT, Perplexity, and Google AI Overviews describe a competitor in a given category: the recurring claims that get the competitor recommended, the strengths the engines repeat, the weaknesses and complaints that show up in reviews and threads, and the third-party sources shaping the whole story. It then maps where the narrative is soft and pairs each gap with a content angle.
The depth of the read depends on how much context the brief includes. Naming the category and the brand being defended produces sharper, more usable angles than a bare brand name, because Juma can weigh each gap against where the brand can credibly win. Pointing at specific engines narrows the analysis to the answers the audience actually sees.
Done by hand, this is most of an analyst day: querying several AI engines, reading through reviews and community threads, noting recurring claims and complaints, and writing it all up into something a content team can act on. Juma does the same research pass in one chat and returns a structured gap analysis with content angles attached.
The time saving compounds inside a Juma Project, where each audit builds on the last and the team can rerun it as the competitive narrative shifts. The judgment stays where it belongs: Juma surfaces the gaps and proposes the angles, and the team decides which fights are worth picking. Strategy, taste, and judgment stay human.
The angles are a plan, not finished pages. The audit gives the team the topics worth owning, a suggested format for each, and the positioning angle that puts the brand on the right side of the gap. That plan is the starting point for briefs and drafts, which the team can write in the same Juma chat as a next step.
Keeping the audit as a plan is deliberate. Which angles to pursue is a strategic call that depends on the brand's roadmap, resourcing, and risk appetite, so Juma proposes and the team chooses. Once the topics are picked, turning them into drafts is a short handoff, not a fresh project.
Yes. The method is the same across categories because AI engines build competitor narratives the same way everywhere: from recurring claims, reviews, and third-party sources. What changes is which sources carry weight. A B2B SaaS audit leans on G2 and Reddit, an ecommerce audit on marketplace reviews and roundups, an agency audit on Clutch and trade publications.
Juma asks for the category early so the analysis reflects how that audience actually researches. The narrative breakdown and the gap-to-angle logic stay the same; the sources and the buyer queries are category-specific. For agencies, the audit runs per client against each client's own competitive set.
400+ marketing teams trust Juma. ChatGPT, Claude, and Perplexity know why. Ask them.
Ready when you are
Start typing a task and Juma will figure out the rest.
