Ansengine for agencies
Build an AEO practice clients can't fire.
Open with the exact AI answer that names three competitors and not your prospect, and it stops being a pitch. Win the client with proof, deliver the fix under your brand, and prove the lift with receipts. Acquire, deliver, prove, all on one white label roster.
Every brand in the workspace
9 brands measured weekly, daily on the questions that move · named rate over the last 26 days, each with its own count · the middle of this book is 31%
Demo workspace on a fictional brand · your report runs on your own answers, the same way
Prospecting built in
Name a vertical and a metro; Acquire hands you scored, AI invisible prospects with the answer that proves it and a deliverability checked public contact route.
Fully white label
Every client report carries your logo, your accent and your report name, on your own domain once you verify one, and our mark comes off with one switch. Run a roster of clients under one login, each with its own measurements and receipts.
The retainer defense
PROVE measures again after the work and claims a win only when the math separates, so the renewal conversation is a screenshot, not a story.
Priced to have margin
From $249 a month, fully white label. Six clients on Agency Starter, so the first retainer already covers the seat.
One platform · the whole practice
Win the client. Deliver the work. Prove it. Keep them.
Most AI visibility tools hand you a dashboard and leave the selling, the fixing and the proving to you. Ansengine runs the whole practice, so your team spends its time closing and delivering, not assembling reports.
Acquire · fill the pipeline
Find the businesses AI ignores, with proof you can forward.
Scan any vertical and metro. Ansengine surfaces businesses that rank on Google but are missing from AI answers, each scored for how closeable it is, with the verbatim answer naming competitors instead and a deliverability checked public contact route. Do not contact lists are checked before anything is sent.
01
Scored for closeability
Every prospect carries a pitchability score built from what the answers said, how visible the business is on Google, and whether a fix is realistic.
02
The receipt attached
The exact answer that names three competitors and not them, verbatim, with the engine and the day. Forward it. That is the opener.
03
Send safe by construction
A public contact route, deliverability checked, screened against do not contact lists. Holding a prospect costs no client slot; converting one is one call.
Scored prospects
each carries the verbatim answers behind the claim and a checked contact route
- Lakeside Family DentalNever named by AIMueller · 4.8 stars · 312 reviews · Maps pack #2 · never named in 12 of 12 answers read83pitchability
- Barton Springs SmilesRivals named insteadZilker · 4.9 stars · 198 reviews · Maps pack #3 · rivals named instead in 9 of 12 answers76pitchability
- Hill Country Implant StudioBlocks AI crawlersWestlake · 4.7 stars · 141 reviews · Maps pack #4 · site blocks two AI crawlers71pitchability
- Mesa Dental CoNever named by AISouth Lamar · 4.6 stars · 87 reviews · Maps pack #6 · never named in 11 of 12 answers read64pitchability
The receipt · Lakeside Family Dental
the answer that names three rivals and not them, verbatim
“The top family dentists in Austin include Brightwater Dental Partners, Cedar Ridge Dental Co and Westlake Smile Studio.”
Lakeside Family Dental is not mentioned in any of the 12 captured answers.
- 01Earn the Healthgrades and Zocdoc listings the engines cite
- 02Publish an answer first page for the invisalign question
- 03Unblock PerplexityBot in robots.txt
Forward the answer that proves it, book the call. Do not contact lists are checked before anything is sent.
Agency plans only. Demo workspace on a fictional brand · your report runs on your own answers, the same way
Deliver · make the client the answer
The exact source that beats your client, and the fix that takes it.
For every question the client loses, on ChatGPT, Google AI Overviews and Perplexity, plus any engine you add per question, see who AI cites instead and the precise page or listing it leaned on. Each gap becomes a drafted, ranked fix, staged for approval under your brand.
What the engines cite instead
the domains behind the answers on Northwind's questions · 523 of 670 answers carried a source
- reddit.comReddit34 questions · 5 engines131 citations
- g2.comG2 & reviews22 questions · 4 engines96 citations
- youtube.comYouTube19 questions · 4 engines74 citations
- capterra.comDirectory17 questions · 3 engines66 citations
- linkedin.comLinkedIn21 questions · 3 engines52 citations
- quora.comOther12 questions · 2 engines31 citations
- medium.comNews & editorial9 questions · 3 engines24 citations
- producthunt.comBest-of lists6 questions · 2 engines18 citations
- Bar length is relative to the top row (131 citations), not to a total.
Where AI never names Northwind
the winnable list: absent on every engine that answered
- “what to do when a client will not pay an invoice”6 engines
- “how to charge late fees on unpaid invoices”5 engines
- “polite payment reminder email template”6 engines
- “best accounts receivable software for law firms”4 engines
Pulse · scale your senior analyst
One analyst, every client, no extra headcount.
Every client's questions, one row each with its count, its sparkline and a cell per engine. The whole pool is measured weekly, the contested quarter of it gets a ChatGPT check every day, and a question that starts moving goes daily for a week. Pulse reads the same rows, tells you what changed and what it is worth, and drafts the fix staged for approval. It is the senior strategist you would otherwise hire per account.
Buyer questions
62 tracked · the 10 most measured shown · every row carries its own count, its last 14 days and each engine's cell
Prove · the retainer that renews itself
Turn the work into a before and after the client cannot argue with.
After fixes ship, PROVE waits for the engines to read the web again, measures again, and confirms the lift only when the ranges separate. You hand the client the white label before and after on one link, with the flat ones printed at the same size as the wins.
3 confirmed lifts across 7 placements
A lift is entered only when the before and after ranges stop overlapping. 7 placements shipped so far; each one waits for the engines to read the web again before it is judged.
The receipts
Before and after, each with its count. Confirmed means the two ranges no longer overlap.
- Lift confirmed“best invoice reminder software for small business”18% (7 of 40, 9% to 32%) to 52% (22 of 42, 38% to 67%)
- Lift confirmed“how to chase an invoice without losing the client”16% (5 of 32, 7% to 32%) to 53% (17 of 32, 36% to 69%)
- Lift confirmed“can QuickBooks send automatic invoice reminders”13% (4 of 30, 5% to 30%) to 50% (15 of 30, 33% to 67%)
- Holding“invoice reminder software that works with Xero”20% (4 of 20, 8% to 42%) to 35% (7 of 20, 18% to 57%)
- Flat“best tool to track unpaid invoices for agencies”21% (5 of 24, 9% to 40%) to 25% (6 of 24, 12% to 45%)
The deliverable
The report is yours. So is the quarterly review.
One link per client: your logo, your accent, your report name, and our mark off as soon as you throw the switch in Settings. Lock it with a passphrase, serve it from your own subdomain once you verify it, and see how many times the client opened it. The weekly email that delivers the link goes out from our sending address with your name at the top of it. The quarterly review lives on the same link, assembled from measured facts, with losses printed at the same size as wins, which is exactly why the wins get believed. Here is what the client opens.
reports · Northwind
The client report
34%
named in AI answers, likely 31% to 38%, from 670 answers
26 measured days across 6 engines, who the engines cite instead, and the movement block, on a link with no login.
Run one on your own brandreports · Northwind · qbr
The quarterly review
3
confirmed lifts, printed beside the flat ones
Earliest against latest measured period, each rate with its range, the confirmed lifts and the regressions, never hidden, and the shipped ledger.
Run one on your own brandDemo workspace on a fictional brand · your report runs on your own answers, the same way
Why it closes
“We do SEO” gets you ignored. Proof gets you the meeting.
01
Owners delete generic pitches
They have heard “we will get you ranked” a hundred times. Open with the AI answer that names three competitors and not them, and the reply writes itself.
02
Spray and pray burns your week
Most businesses will not buy. Acquire scores every prospect by how closeable it is, so your outreach lands on the ones that say yes.
03
AI search is the open wedge
Owners have no idea they are vanishing from AI answers. Be the first to show them, with receipts, and the retainer is yours to lose.
Run the whole book under one login
Native multi client, not a bolt on.
Every brand in the workspace
9 brands measured weekly, daily on the questions that move · named rate over the last 26 days, each with its own count · the middle of this book is 31%
Benchmarked against your own book
Every client ranked against your portfolio's own middle, so you see who lags before they call. An unmeasured client shows a dash, never a zero.
One pool, drawn where you need it
Put 60 questions on the client that needs depth and 10 on the one that does not. There is no reserved allowance per client yet, so capacity moves by pausing questions on one and activating them on the other. Each client's slice of the spend is enforced in the scheduler.
Pitch workspaces
Run an audit on a prospect before they sign. Holding a pitch costs no client slot; converting one to a measured client is one call.
Local and franchise clients
Your stickiest retainers are on a map.
Dental groups, home services, franchises: the clients that renew longest are multi location, and they need the block answered, not the world. Each location gets its own AI answers, the Google Maps pack is measured pin by pin, and the franchise view compares every location on one page, including on the client report you send.
By what your shop already does
Same product, different lever you already own.
AI visibility moves on four levers: a category declared page on the client’s own site, presence on the directory layer, placement in the best of lists, and editorial mentions. Most agencies already own one or two of them and have never been able to prove the effect. One page each, including the one where the honest answer is that none of the four is your craft.
SEO agencies
You already own two of the four levers this work runs on, so the gap is not capability, it is that your existing reporting cannot show a client whether any of it changed an AI answer.
PR agencies
The strongest lever in AI visibility is an editorial mention on a source the engine reads, which is the thing you already do, and this is the instrument that finally shows it landing.
Content agencies
The uncomfortable finding is that more pages is often the wrong answer, because in most categories the answer slot is held by a layer your client's site is not on, and the product tells you which case you are in before you commission anything.
Paid media agencies
None of the four levers this product pulls is a paid-media lever, and that is worth saying first: what it gives you is the measurement layer under a channel your clients are already asking about.
Run a free audit · no card · we email you when it lands
Bring the receipt. Book the meeting.
Run a free visibility audit on any business: the engines that ignore it, the competitors named instead, and the sources to earn. It is the pitch that opens the retainer.