Best of · researched 15 September 2026
By Advik Jain, Co-founder & CEO, Optivus Technologies · Published 11 September 2026 · Updated 15 September 2026
An AI visibility tool runs the questions your buyers ask through answer engines and records whether your brand is mentioned, cited or missing. We reviewed twelve from their own pages and list them in the order our published scoring rules put them, with Veritas, our product, scored by the same rules.
No single tool wins. Decide first whether you only need to watch AI answers or also want help writing the pages that win them, then compare the engines included at the price you would pay.
Ordered by the score our published rules calculate. Veritas is ours, is scored by the same rules, and sits where they put it.
| Tool | Best for | From | Engines | Score |
|---|---|---|---|---|
| SE Ranking | Agencies that want a full SEO suite with a large, clearly priced prompt allowance | $129/mo + $89 | 5 of 7 | 80 |
| Frase | Writers who want tracking attached to a content workflow | $49/mo | At least 1 of 7 | 74 |
| Writesonic | One brand wanting tracking, fixes and outreach on a card-free trial | $79/mo | 3 of 7 | 70 |
| VeritasOurs | Teams that want to track AI answers and write the cited pages that win them, in one workspace | ₹20,000/mo | 5 of 7 | 68 |
| Semrush | Teams already on Semrush who want AI tracking beside their SEO data | $199/mo | At least 5 of 7 | 68 |
| Scrunch | Six major engines plus Meta AI on every plan, with persona benchmarks | $300/mo | 6 of 7 | 67 |
| Otterly.AI | Low-cost daily monitoring with unlimited team members | $29/mo | 4 of 7, +3 add-ons | 61 |
| Surfer | Content teams who want tracking wired into their editor | $82/mo, yearly | At least 3 of 7 | 61 |
| AthenaHQ | Starting free and paying per AI response | Free plan | 5 of 7 | 57 |
| Peec AI | Daily multi-model tracking with unlimited users on every plan | Not published | Chosen per plan | 55 |
| Profound | Enterprise teams that want the deepest dedicated analytics | $99/mo | 1 of 7 | 48 |
| Ahrefs Brand Radar | Teams already paying for Ahrefs who want AI visibility without adding a vendor | $129/mo | Chosen per plan | 48 |
From is the cheapest published plan that tracks AI answers, with any add-on that tracking requires. Veritas bills in rupees, excluding GST; everything else is in US dollars as each vendor publishes it.
The one on this list that writes the fix
Veritas tracks five AI engines, then drafts each missing page from your own documents with every knowledge-based claim cited to its source. Try it on one domain.
On the cheapest plan that tracks AI answers. Most vendors list more engines than that plan includes, so this is what you get at the price shown, not the brochure maximum.
| Tool and plan | ChatGPT | Claude | Gemini | Perplexity | AI Overviews | AI Mode | Copilot |
|---|---|---|---|---|---|---|---|
SE Ranking Core with the AI Search Toolkit | Included | Not on this plan | Included | Included | Included | Included | Not on this plan |
Frase Starter | Included | Not on this plan | Not on this plan | Not on this plan | unclear | unclear | Not on this plan |
Writesonic Starter | Included | Not on this plan | Included | Not on this plan | Included | Not on this plan | Not on this plan |
VeritasOurs Content + SEO Premium | Included | Included | Included | Included | Included | Not on this plan | Not on this plan |
Semrush Starter | Included | Included | Included | Included | Included | unclear | unclear |
Scrunch Starter | Included | Included | Included | Included | Included | Included | Not on this plan |
Otterly.AI Lite | Included | add-on | add-on | Included | Included | add-on | Included |
Surfer AI Search Analytics | Included | unclear | Included | unclear | Included | unclear | unclear |
AthenaHQ Essential (free) | Included | Not on this plan | Included | Included | Included | Not on this plan | Included |
Peec AI Starter | pick | pick | pick | pick | pick | pick | pick |
Profound Starter | Included | Not on this plan | Not on this plan | Not on this plan | Not on this plan | Not on this plan | Not on this plan |
Ahrefs Brand Radar Lite | pick | pick | pick | pick | pick | pick | pick |
A tick is included. Add-on means an extra monthly fee. Pick means you choose a limited number of platforms. Unclear means the vendor’s wording does not say.
This decides more than the engine list does. A tracker tells you where you are missing; whether it also writes the fix decides how much work stays with your team.
Placed by what each product does on any plan, as its own pages describe it. The capability may need a higher plan than the cheapest one; each full review says which. Not described means we found no description either way.
The cheapest published plan that fits each team, over twelve months, with seat fees and required add-ons. Where no published plan fits, the note says why instead of guessing.
| Tool | One marketer, one brand1 seat, 50 tracked prompts, 1 brand | Team of five, one brand5 seats, 100 tracked prompts, 1 brand | Agency, three brands10 seats, 300 tracked prompts, 3 brands |
|---|---|---|---|
| Frase | $588lowest | $2,244lowest | $5,328lowest |
| Surfer | $984 | $3,588 | No published fit1 |
| Profound | $1,188 | No published fit1 | No published fit1 |
| Otterly.AI | $2,268 | $2,268 | $5,868 |
| Semrush | $2,388 | $5,748 | No published fit1 |
| SE Ranking | $2,616 | No published fit2 | No published fit1 |
| Scrunch | $3,600 | $4,200 | $5,700 |
| VeritasOurs | $3,780 | No published fit1 | No published fit1 |
| Ahrefs Brand Radar | $17,988off scale | No published fit1 | No published fit1 |
| AthenaHQ | No published plan fits any of these setups3 | ||
| Peec AI | No published plan fits any of these setups4 | ||
| Writesonic | No published plan fits any of these setups5 | ||
Bars share one scale across the whole table, so cost growth shows as well as rank. A figure far above the rest is marked off scale.
Standard monthly rates, without annual discounts unless a vendor publishes only a yearly price. Taxes, overage and negotiated terms are excluded. Indian rupee figures are converted at 95.2 INR to the US dollar, a rate we maintain by hand and review periodically. It is an approximation, not a live rate, and Veritas bills in rupees.
Veritas’s price includes keyword research, site audits and cited drafting, not only tracking.Start a 7-day trialSee pricing
Best for agencies that want a full SEO suite with a large, clearly priced prompt allowance.
The AI add-on is priced in the open and starts at 200 prompts, far more than most suites include.
AI search tracking is an add-on at $89 a month on top of the plan fee.
Best for writers who want tracking attached to a content workflow.
The most complete research-to-publish loop among content optimisers, at a low entry price.
Article generation is capped at 10, 40 or 100 a month by plan.
Best for one brand wanting tracking, fixes and outreach on a card-free trial.
Measurement, prioritised fixes, page rewrites and outreach in one product.
Starter to Growth track three platforms; the full ten are Enterprise-only.
Best for teams that want to track AI answers and write the cited pages that win them, in one workspace.
Every knowledge-based claim in a draft opens the passage it came from, so review is a check rather than a rewrite.
AI answers are re-checked every two weeks, with on-demand checks in between. Several dedicated trackers refresh daily.
Best for teams already on Semrush who want AI tracking beside their SEO data.
The deepest historical keyword and backlink data of any tool compared here.
Daily prompt tracking starts on the $199 Starter plan; the $139 SEO plan shows no tracked-prompt allowance.
Best for six major engines plus Meta AI on every plan, with persona benchmarks.
Seven engines on every plan, the broadest coverage at a self-serve price.
The entry price is $300 a month, or $250 billed annually.
Best for low-cost daily monitoring with unlimited team members.
The lowest paid entry point among dedicated trackers, with daily refresh on every plan.
Claude tracking costs $29, $109 or $439 a month on top of Lite, Standard or Premium.
Best for content teams who want tracking wired into their editor.
The most widely used content score, now covering AI Search as well as SEO.
The $49 Discovery plan does not include AI visibility tracking.
Best for starting free and paying per AI response.
A free plan, so tracking can start before a budget conversation.
Usage is metered in credits, where one credit is one AI response.
Best for daily multi-model tracking with unlimited users on every plan.
Unlimited users on every plan, with daily tracking and multi-project support.
Prices are not present in the pricing page HTML; they render only in a browser, so they cannot be quoted or compared here.
Best for enterprise teams that want the deepest dedicated analytics.
The deepest measurement of any dedicated tracker compared here, including real prompt volume data.
The $99 Starter plan tracks ChatGPT only, with 50 prompts and one seat.
Best for teams already paying for Ahrefs who want AI visibility without adding a vendor.
AI visibility is included in plans many SEO teams already pay for, at no extra cost for small prompt sets.
Prompt allowances are small until Enterprise: 5, 10 and 20 prompts on Lite, Standard and Advanced.
On prices read 11 September 2026, AthenaHQ has a free Essential plan with 300 credits, and Otterly.AI’s paid plans start at $29 a month for 15 prompts. Cheapest is rarely the right test: check which engines the entry plan actually includes.
At enterprise tiers, Profound lists up to nine engines and Writesonic lists ten platforms. At a self-serve price, Scrunch includes seven engines on every plan, and AthenaHQ Starter covers 11 models.
Often enough to see a trend, which is less often than it sounds. The same prompt returns a different answer on different runs, so one daily check says little on its own. Daily refresh helps during a launch or a crisis; weekly or fortnightly is enough for a content programme.
Treat any single reading as noise. In SparkToro’s research across 2,961 prompts, the same list of brands came back less than one time in a hundred. A mention rate measured across many prompts over several weeks is the signal worth tracking, whichever tool you use.
It includes one. Veritas tracks five engines on Content + SEO Premium and up, but it is priced and built as a content and SEO platform. If tracking is the only job, a dedicated tracker on this list is the simpler buy.
Every feature, price, pro and con, each fact linked to the vendor page it came from.
Best for: Agencies that want a full SEO suite with a large, clearly priced prompt allowance
The AI Search Toolkit is an add-on at $89 a month starting at 200 prompts, expandable to 450 or 1,000, covering AI Overviews, AI Mode, ChatGPT, Gemini and Perplexity with linked and unlinked brand mentions. Core is $129 a month with one manager seat, Growth $279 with three, and both include API and MCP access. Claude is not tracked and the refresh cadence is not published.
SE Ranking is a complete SEO platform: rank tracking, keyword research across a 5.5 billion keyword database, backlink analysis, site audits, on-page checks, local marketing, white-label reporting and an API, with MCP access for querying data inside AI assistants. It says it is trusted by more than 40,000 agencies.
Its AI Search Toolkit tracks AI Overviews, AI Mode, ChatGPT, Gemini and Perplexity, with brand mentions and links, unlimited competitor research and historical data. It is an add-on at $89 a month, or $71.20 billed annually, starting at 200 prompts and expanding to 450 or 1,000.
Trial: A free trial is offered; terms are not stated on the pricing page. source, 11 Sept 2026
Not the right fit for: Teams that need Claude tracked, or that want content grounded in their own documents.
seranking.comBest for: Writers who want tracking attached to a content workflow
AI visibility from the $49 Starter plan with ChatGPT and Google AI, widening to Perplexity on Professional and Claude and Gemini on Scale, alongside research, drafting, site audits and publishing, with monthly article caps.
Frase began as a SERP research and brief tool and now covers the loop from research to publishing: topic clusters, an AI agent that drafts with your brand in context, brand voice, content optimisation for Google, GEO optimisation for AI citation, site audits, backlinks with drafted outreach, Content Guard for ranking decay, and its own FraseCMS.
AI visibility tracking covers ChatGPT, Perplexity, Claude, Gemini and Google AI across its plans, widening as the price rises. It publishes to WordPress, Webflow, Sanity and Wix, and connects to Claude and Cursor through MCP and a CLI.
Trial: 7 days, no credit card. source, 11 Sept 2026
Not the right fit for: Teams that need every claim traced to their own documents, or high monthly volume without article caps.
frase.ioBest for: One brand wanting tracking, fixes and outreach on a card-free trial
Every self-serve plan tracks ChatGPT, Gemini and Google AI Overviews from $79 a month, with an Action Center, page rewrites for GEO and citation outreach. Perplexity and the other platforms are Enterprise-only, and plans allow one to three users.
Writesonic began as an AI writer and now presents itself as a generative engine optimisation platform. It tracks AI visibility, ranks opportunities in an Action Center by visibility and citation impact, and uses an AI agent to rewrite pages for GEO with FAQ blocks and comparison tables.
It also audits robots.txt, schema, indexing and broken pages from AI bot visits, surfaces the Reddit threads, YouTube videos and listicles citing competitors with drafted outreach, and connects to Google Analytics, Search Console, Looker Studio, WordPress, an API and MCP. It says more than 10,000 marketing teams use it.
Trial: Free trial, no credit card. source, 11 Sept 2026
Not the right fit for: Teams of more than three, or anyone who needs Perplexity or Claude tracked without an enterprise contract.
writesonic.comBest for: Teams that want to track AI answers and write the cited pages that win them, in one workspace
Tracks ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews on Content + SEO Premium and up, re-checked every two weeks with on-demand checks. Each gap opens into an editor that drafts from your documents with a citation on every claim, beside keyword research, site audits and backlinks. No daily refresh, no Copilot or AI Mode, and no API.
Veritas is an AI content and SEO platform built by Optivus Technologies. It organises a company’s own documents into a knowledge graph and drafts marketing content from it, attaching a numbered citation to every knowledge-based claim that opens the passage it came from. Claims it cannot trace to your material are left out rather than filled in.
Around the editor it bundles keyword research and gap analysis, competitor and SERP analysis, a live content score, site audits, backlinks, Google Analytics with Search Console, and AI visibility tracking across ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews. Plans are priced per workspace in INR with users included.
Trial: 7 days on one SEO domain; nothing is charged until it ends. source, 12 Sept 2026
Not the right fit for: Solo writers, teams that need market-wide historical SEO data or an API, and anyone who needs AI answers re-checked daily.
See Veritas pricingBest for: Teams already on Semrush who want AI tracking beside their SEO data
AI Visibility, Prompt Research and AI Brand Sentiment across ChatGPT, Perplexity, Gemini, Claude and AI Overviews, 50 to 200 prompts a day by tier, from the $199 Starter plan. The $139 SEO plan shows no tracked-prompt allowance, and extra users are $45 a month each.
Semrush is an all-in-one search marketing platform covering keyword research, competitor analysis, backlinks, rank tracking, local search visibility, technical site audits, digital PR and content creation. Its keyword database holds over 27.3 billion keywords, by its own count, which is the depth most teams buy it for.
It has moved hard into AI search. Its Research and Track toolkits now include AI Visibility, Prompt Research and AI Brand Sentiment across ChatGPT, Perplexity, Gemini, Claude and AI Overviews. Adobe completed its acquisition of Semrush on 28 April 2026, and says customers can expect an expanded roadmap as the two integrate.
Trial: 7 days on SEO, Starter and Pro+; none listed on Advanced. source, 11 Sept 2026
Not the right fit for: Small teams that mainly need content and AI visibility and do not want to pay per user.
semrush.comBest for: Six major engines plus Meta AI on every plan, with persona benchmarks
Every plan tracks ChatGPT, Claude, Gemini, Perplexity, Google AI Mode, AI Overviews and Meta, with persona benchmarks, AI bot traffic insight and AXP. Starter is $300 a month with 350 custom prompts and three seats.
Scrunch is an AI visibility platform for brands and agencies. It tracks performance for prompts, topics and entities with citations, competitors and rankings, benchmarked by persona, topic and geography, across ChatGPT, Claude, Gemini, Perplexity, Google AI Mode, AI Overviews and Meta on every plan.
Beyond monitoring it shows how AI bots crawl your site in a real-time feed, flags pages AI bots cannot crawl, analyses the sources AI relies on, and offers AXP, an agent experience platform that serves compressed, machine-readable versions of pages to AI agents while leaving the human site unchanged. It says more than 500 companies and agencies use it.
Trial: Trial of 7 days on Starter, no card required. source, 11 Sept 2026
Not the right fit for: Teams on a tight budget, or teams that want content drafted in the same tool.
scrunch.comBest for: Low-cost daily monitoring with unlimited team members
From $29 a month for 15 prompts, refreshed daily. Every plan tracks ChatGPT, Google AI Overviews, Perplexity and Microsoft Copilot; Claude, AI Mode and Gemini are add-ons. API and MCP access from Standard.
Otterly.AI is a monitoring-first platform for AI search. It tracks how brands are mentioned and websites cited across ChatGPT, Google AI Overviews, Perplexity and Microsoft Copilot on every plan, with Claude, Google AI Mode and Gemini as paid add-ons, refreshing daily with unlimited team members.
Around monitoring it adds AI Prompt Research, a Content Audit with crawlability checks and content briefs, GEO optimisation recommendations, Agent Analytics, ChatGPT Ads tracking, and API and MCP access on its higher plans. It says more than 40,000 marketing professionals use it, and was named a Gartner Cool Vendor for AI in Marketing in 2025.
Trial: Free trial offered from its homepage; terms not listed on the pricing page. source, 11 Sept 2026
Not the right fit for: Teams that want the tracker to draft and publish the fixes, or need Claude and Gemini without add-on fees.
otterly.aiBest for: Content teams who want tracking wired into their editor
AI Tracker covers ChatGPT, Gemini, AI Overviews, AI Mode and Perplexity, refreshing daily on Pro and above and weekly on Standard, and its recommendations now act on tracker findings inside the Content Editor. The $49 Discovery plan has no AI visibility, and prices are billed yearly.
Surfer describes itself as an AI Visibility Platform. Its Content Editor scores drafts in real time with a unified score that shows SEO on one side and AI Search on the other, fixed with Auto-Optimize, and Surfer AI generates branded articles, inserting internal and external links to sources as it writes rather than afterwards.
AI Tracker, launched in June 2025, measures mentions, position, sentiment and cited sources across ChatGPT, Gemini, Google AI Overviews, Google AI Mode and Perplexity, refreshing daily on Pro and above and weekly on Standard. Content Audit, Topical Map, an API and, since August 2026, an MCP server round out the product.
Trial: None listed on the pricing page. source, 11 Sept 2026
Not the right fit for: Teams that want monthly billing, a trial first, or backlinks and technical audits in the same tool.
surferseo.comBest for: Starting free and paying per AI response
A free Essential plan covers ChatGPT, Perplexity, AI Overviews, Gemini and Copilot. Starter at $295 a month gives 3,600 credits across 11 models, with sentiment analysis, hallucination detection and unlimited members.
AthenaHQ tracks brand visibility across 11+ AI models with sentiment analysis, real-time brand mention alerts and hallucination detection, and pairs it with a recommendation engine that suggests on-page and off-page actions.
Its Athena Citation Engine analyses citation sources for link building, a content optimisation agent and AI-friendly templates help produce content, and a Shopify integration ties AEO content to revenue. It integrates with GA4, Search Console, Webflow, Framer, Tableau, Power BI and Looker, with API access as a paid add-on.
Trial: Free Essential plan. source, 11 Sept 2026
Not the right fit for: Teams that want predictable flat pricing or SEO tooling in the same product.
athenahq.aiBest for: Daily multi-model tracking with unlimited users on every plan
Plans run at 50, 150 and 350 prompts with one, two and five projects, all with unlimited users, daily tracking and a choice of three models from up to thirteen, plus source analysis, CSV export, Looker Studio, an API and an MCP server. Its prices render only in a browser and are not in the page HTML, so we cannot quote or compare them here.
Peec AI is an AI search analytics product for marketing teams, tracking visibility, position and sentiment across AI platforms, with prompt setup, brand and competitor tracking, source analysis, CSV exports, a Looker Studio integration, an API, an MCP server and single sign-on. It says it is trusted by more than 3,000 brands and agencies.
Its plans are described by prompts and projects rather than seats: Starter at 50 prompts and one project, Pro at 150 prompts and two, Advanced at 350 prompts, five projects and multi-country tracking, all with unlimited users, daily tracking and a choice of three models from up to thirteen. It raised a $21 million Series A led by Singular in November 2025, after a €7 million seed.
Trial: A free trial is offered from its site. source, 11 Sept 2026
Not the right fit for: Buyers who need a published price before a sales conversation, or want content produced in the same tool.
peec.aiBest for: Enterprise teams that want the deepest dedicated analytics
Measurement is the core, with real prompt volume data and crawler analytics, and Agents now draft, refresh and publish content with approvals. Starter at $99 tracks ChatGPT only; three engines start at $399; up to nine on Enterprise.
Profound is a dedicated answer engine optimisation platform built for large brands. Its core modules are Answer Engine Insights for how brands appear in AI conversations, Prompt Volumes for what people actually ask AI systems, and Agent Analytics for how AI crawlers read your site, across engines including ChatGPT, Perplexity, Gemini, Claude, Grok, Microsoft Copilot, DeepSeek and Google AI Overviews.
It has added execution on top of measurement. Profound Agents create net-new content, refresh declining pages and generate FAQs using your brand kit, knowledge base, Answer Engine Insights and prompt data, with human approval before anything is published through a CMS integration. Profound raised a $96 million Series C at a $1 billion valuation in February 2026.
Trial: “Try for free” offered on Growth; none listed on Starter. source, 11 Sept 2026
Not the right fit for: Small teams tracking more than ChatGPT on a budget, or anyone who needs a full SEO toolkit alongside.
tryprofound.comBest for: Teams already paying for Ahrefs who want AI visibility without adding a vendor
Brand Radar checks AI Overviews, ChatGPT, Gemini, Perplexity, Copilot, AI Mode and Claude daily, and custom prompts come free in every paid plan. The allowance is a monthly check budget: Lite’s 150 checks cover its 5 prompts on one platform, so more platforms means fewer prompts. Each plan includes one user, and Brand Radar AI research across 475 million prompts is a separate $199 a month.
Brand Radar is Ahrefs’ AI visibility product, tracking how often a brand is mentioned or cited across AI Overviews, ChatGPT, Gemini, Perplexity, Copilot, AI Mode and Claude, with mentions, citations, AI share of voice and estimated impressions. Custom prompts are included free in every paid Ahrefs plan, refreshed daily, at 5, 10, 20 and 83 or more prompts by tier.
Its distinguishing asset is scale: Brand Radar AI, from $199 a month, researches any brand across more than 475 million prompts in Ahrefs’ database, and Ahrefs says its system processes more than 239 million prompts a month. It sits inside the Ahrefs suite, so the keyword, backlink and audit tooling comes with it, as does the Ahrefs bill.
Trial: No trial listed; a free Ahrefs plan is available. source, 11 Sept 2026
Not the right fit for: Teams that need many tracked prompts on a small plan, several seats, or content produced in the same tool.
ahrefs.com/brand-radar| Tool | Weighted score | Criteria scored | Strongest criterion |
|---|---|---|---|
| SE Ranking | 80 | 6 of 8 | SEO beyond AI answers |
| Frase | 74 | 7 of 8 | Refresh frequency |
| Writesonic | 70 | 5 of 8 | What happens after a gap |
| Veritas Our product | 68 | 8 of 8 | What happens after a gap |
| Semrush | 68 | 7 of 8 | Refresh frequency |
| Scrunch | 67 | 6 of 8 | Engine coverage at the entry price |
| Otterly.AI | 61 | 7 of 8 | Refresh frequency |
| Surfer | 61 | 8 of 8 | Refresh frequency |
| AthenaHQ | 57 | 5 of 8 | Team and data access |
| Peec AI | 55 | 5 of 8 | Refresh frequency |
| Profound | 48 | 6 of 8 | What happens after a gap |
| Ahrefs Brand Radar | 48 | 6 of 8 | Refresh frequency |
Eight rules, each applied to the published facts in the reviews and weighted to a total out of 100. The rules and weights are set out in our methodology.
Where a figure below is missing, the charts say so rather than estimate it.
Every fact about a tool other than Veritas comes from that vendor’s own pricing page, product pages, documentation or announcements. Each one is linked where it appears, with the date we read it.
We build Veritas and use it every day. We have not run paid accounts of every other tool on this page, so we describe what each vendor publishes rather than scores from tests we did not run. Where a capability was not described on a vendor’s pages, the table says exactly that, which is not the same as saying the product lacks it.
G2, Capterra and TrustRadius blocked our automated reads, so any rating shown is the figure the vendor publishes about itself, labelled as such.
Prices and plans change. We re-read every vendor page at least every 90 days and record what changed in the update log below.
The scores and cost figures on this page are calculated, not assigned. Our full methodology sets out every rule, what each criterion is worth, and where our evidence runs out.
Disclosure. Veritas is our product. It appears on this page because it competes in this category, and we have tried to state its limits as plainly as anyone else’s. If anything here is wrong or out of date, write to support@getveritas.io and we will correct it and note the change in the log.
How this page was made. Veritas is our product. This page was written with AI assistance and checked against the vendor pages linked beside each fact, on the dates shown. Scores and costs are calculated by our published rules.
7 days on one domain. Nothing is charged until the trial ends.
