Blog · 24 September 2026 · By HeardOf

HeardOf vs Otterly: what each one measures, and what Otterly does better

What does Otterly cost?

$29 a month for its Lite plan on monthly billing, or $25 a month on annual billing, on Otterly's own pricing page, read 23 September 2026. Lite carries 15 prompts, four AI engines — ChatGPT, Google AI Overviews, Perplexity and Microsoft Copilot — one workspace, daily tracking, unlimited team members and unlimited brand reports, per the same page.

The other plans, same page and day: Standard at $189 a month, or $160 on annual billing, with 100 prompts, API and MCP access, unlimited workspaces and a Looker Studio connector; Premium at $489, or $422 annual, with 400 prompts; Enterprise on request, from 1,000 prompts. Google AI Mode, Gemini and Claude are add-ons on every plan, priced by plan: $9, $9 and $29 a month on Lite, $59, $59 and $109 on Standard, $149, $149 and $439 on Premium, on monthly billing, with annual figures on the same page. Packs of 100 extra prompts are $99 a month on Standard and Premium and are not offered on Lite. The trial is seven days, with no credit card and 50 prompts, by its help article on the free trial, dated 4 August 2026 on the page; the pricing page's own FAQ says a trial exists and does not say how long; five of its pages read the same day — the home page, Features, AI Search Analytics, Prompt Research and geo-tools — carry a button reading "Start Your Free 14-Day Trial", and the home page's own visible FAQ says "a 7-day free trial for new users". Two lengths on one site, and we print both.

These are the same Lite figures we read on 5 September and again on 23 September for our post on what nineteen AI-visibility tools charge, which lists Otterly among the fifteen vendors that show a price on their own site. What HeardOf charges is on our own site and not in this post.

What does each one actually measure?

Otterly checks a prompt list you keep, every day, on four engines and up to seven, and reports brand coverage, share of voice, sentiment, domain citations and a visibility index built from coverage and average position — its help center, read 23 September 2026. HeardOf counts whether you and up to three competitors you name were named or cited, per engine, on 40 prompts we write for your category, on four engines as of 23 September 2026 — our own settings, as described in our post on which engines we check. The table puts the two side by side; Otterly's column is its own pages, read 23 September 2026, with each help article's own date; ours is that post, same date.

What does Otterly do better?

Five things: seven engines on its pricing page where we run four, as of 23 September 2026 per our post on which engines we check; 68 countries in its help table where, per the same post and date, our ChatGPT and Perplexity calls are pinned to the United States, the AI Overviews page is fetched as a United States search, and no country is stated for Gemini; a list you keep, checked every day, with the history kept; sentiment and position, which the same post does not describe as of the same date; and a published study of its own data at a scale we have not matched. All five of Otterly's halves are read from its pages on 23 September 2026, and none is a matter of opinion.

Engines. Otterly's pricing page puts ChatGPT, Google AI Overviews, Perplexity and Microsoft Copilot on every plan and sells Google AI Mode, Gemini and Claude as add-ons; its API reference page lists seven engine values. Copilot, AI Mode and Claude are on our list of what we do not check, as of 23 September 2026 and per the same post, and so are Grok, DeepSeek, Meta and Mistral.

Countries. Its help article on countries, dated 28 July 2026 on the page, says it monitors "as if a real user were searching from within each country" and lists 68 of them in a table, with four engine-country cells marked as not supported; the same article says "65+" and the pricing page says "50+", so we print all three. Our ChatGPT and Perplexity calls are pinned to the United States and the AI Overviews page is fetched as a United States, English search, per our post on which engines we check as of 23 September 2026.

Daily, on a list you keep. Otterly runs every tracked prompt on every engine once a day, at no fixed hour, and accumulates from the day a prompt is added, by its monitoring-interval and historical-data articles; its KPI definitions, dated 4 September 2026 on the page, print the counting rule, and its brand-detection article prints the matching rule. Competitors are unlimited, and brands that showed up without being added are listed for you to add, by its track-competitors article. Our audit asks each prompt once per engine per UTC day and scores four brands, you and up to three competitors you name, and a brand you did not name is stored and not counted — our own rules as read on 19 September 2026, per our post on which engines we check as it stood on 23 September 2026; if a fifth competitor is the one taking your answers, the audit will not tell you.

Sentiment and position. Otterly scores sentiment from −100 to +100 and builds its Brand Visibility Index from coverage and average position, by its KPIs and sentiment articles. Our post on which engines we check describes named and cited, describes no measure of tone, and says that as of 19 September 2026 our scoring stores whether you were named or cited, not where: an earlier version of the tool did record position, the product dropped it, and as of that date we have not put it back. The same post, as of 23 September 2026, cites the SparkToro and Gumshoe study of 28 January 2026 and says we have no measurement that contradicts it.

The study. Otterly's post on answer stability, dated 11 September 2026 on the page — 520 prompts, three industries, seven engines, 40 to 91 days, 252,407 answers, one run per prompt per engine per day, collected in the United States — is Otterly's data on its own product; the run our post on which engines we check describes, as of 23 September 2026, is 30 B2B software brands on two engines, ChatGPT and Perplexity, 25 prompts, on 4 September 2026. One of its findings bears on prompt counts in general: in its 320-prompt AI-search-monitoring dataset, "moving from 10 to 100 prompts reduced variation in the reported Brand Coverage by about 73%", and with 10 prompts "90% of the Brand Coverage results fell between 39% and 71%", narrowing to 49%–62% at 50. Otterly's Lite plan carries 15 prompts. Our audit asks 40 per engine, as of 23 September 2026 per our post on which engines we check, and we do not place that number on Otterly's curve: its subsets were drawn with a 30-day window of daily runs held fixed, and one HeardOf audit holds one answer per prompt per engine, not thirty days of them, so the two variances are not the same measurement. The rest of its blog is 180 posts by our census of 23 September 2026, 36 of them with a study-shaped slug, more than any of the other four vendors we counted, and its llms.txt experiment of 5 February 2026 is the one measurement among the four vendor posts on that file.

Method is a difference rather than a ranking. Otterly's help article on how it collects data, dated 17 July 2026, says it posts prompts "through the website of the AI search (like chatgpt.com etc)" so that the responses it captures are "identical to what your customers experience", and uses an API only for Claude; its blog post of 19 May 2025 on API versus interface says the two can differ. We reach ChatGPT, Perplexity and Gemini through their vendors' APIs and print which model answers, per our post on which engines we check, as of 23 September 2026. If what you want measured is what that article calls what your customers experience, that is Otterly's claim about its method, and ours, as of the same date, does not make it. Whether an answer came from a model's training or from a web search is a separate question from how it was collected, and training data or live search takes it up.

Is there anything HeardOf measures that Otterly does not?

On the evidence we have, no, and we would rather say so than invent one. Otterly's help center, read 23 September 2026, counts mentions and citations, coverage, share of voice, sentiment and position, per engine, per day; the two halves of the named-versus-cited split we report — per our post on which engines we check, as of 23 September 2026 — appear in its prompt-monitoring article, dated 17 July 2026, as Brand Mentions and Domain Citations. This is an Otterly AI review against the tool we build, read from Otterly's pages, not from inside its reports.

What differs is what each one prints. Otterly's help center prints its daily cadence, its 1-or-0 counting rule, its exact-match detection and its no-backfill rule, each in a dated article. Ours, in our post on which engines we check as of 23 September 2026, prints the model behind each of ChatGPT, Perplexity and Gemini, reached through their vendors' APIs, what Google AI Overviews is read from, what an empty AI Overview counts as, the 90% rule under which an engine leaves the headline, and a list of what we do not measure. On one rule the two print opposite answers. Otterly's KPI definitions, dated 4 September 2026 on the page, say of Google AI Overviews that "When AIO are not triggered, they do not count towards that formula": in the worked example on that page, a prompt on which Google showed no overview leaves the coverage denominator. Ours, per the same post of ours as of 23 September 2026, counts that prompt as an answer in which nobody was named, which lowers the rate. Both rules are printed, and a buyer comparing an AI Overviews figure from each should know they are not the same fraction. One thing none of the 144 pages in its help-center sitemap says, and its API specification does not either, as of 23 September 2026: how a page that failed to load is counted. The specification gives each AI response a state field and a flag saying whether an overview was available, and defines no counting rule for either. That is not a claim that Otterly has no rule; it is a statement about the pages we read.

The other difference is shape. Otterly measures a list you keep — 15 prompts on Lite, checked daily, in a country you choose — and grows the history from there. We write 40 prompts from your category and the competitors you name and ask them on four engines, as of 23 September 2026 per the same post; a list you keep can be wrong in ways you chose, a set written for you can be wrong in ways you did not, and neither is the other's superset.

Which one should you buy?

If you came here for an Otterly alternative, the answer turns on what you want measured. Buy Otterly Lite if you want a prompt list you keep, checked daily on four engines including Copilot, in one of 68 countries, with sentiment and position, and 15 prompts is enough — $29 a month, or $25 on annual billing, its figures read 23 September 2026. Buy HeardOf if you want one category, you and up to three competitors you name, 40 prompts written for you on four engines with the method printed, as of 23 September 2026 per our post on which engines we check, and a report that says who was named and who was cited.

If your market is outside the United States, Otterly's 68-country table settles it and nothing here argues otherwise. If you want an API, workspaces for clients and a Looker Studio connector, those start at Otterly's Standard plan, $189 a month on its pricing page, and our post on which engines we check, as of 23 September 2026, describes none of them. If the tool you are weighing is Peec AI rather than Otterly — Starter at $95 a month on monthly billing, as Peec's pricing page draws it in a browser on 24 September 2026, in US dollars as drawn for us, a figure the page's served HTML does not carry — that comparison, with what Peec does better, is in HeardOf vs Peec AI.

The price of what we do is on our site, and it is not in this post, because a comparison page that carries a number we can change is a page that goes stale on our account.

How was this read?

With curl and a browser user-agent, on 23 September 2026, one fetch per page, no JavaScript executed. Otterly's pricing page serves both its monthly and its annual figures in the page and both are reported. Of the 32 help-center pages we quote, the 29 articles carry a date at the top, which we quote as the page's own date, and the three category pages carry none; the two blog posts carry theirs. For the one negative in this post we also fetched all 144 pages in the help center's sitemap and its API specification, the same day. Otterly's own alternatives page, read the same day, compares it with nine tools, and HeardOf is not one of them.

Two third-party claims about Otterly circulate, and we could reach the source of one. GetLatka lists Otterly at an estimated $770K in annual revenue for 2025 and approximately 7 employees as of 2026, on a page last updated 14 September 2026 that says the revenue is its own estimate; the same page puts the headquarters in the United States, and Otterly's llms-full.txt, generated 14 August 2025 by its own header, puts it in Persenbeug, Austria. Otterly's about page says the product launched in October 2024 and that 40,000 marketing professionals were using it as of August 2026; its home page says 40,000+ in one place and 30,000+ in two; its Features page says 40,000+, its geo-tools page 30,000+ and its llms.txt 30,000+ — two figures across five of its pages, all Otterly's, read the same day. A second claim, that Otterly's founders grew it without paid advertising, is attributed in our research to growthmaestros.com, which answered HTTP 500 on every address we tried on 23 September 2026, so it is not in this post.

Everything this post says about our own system comes from our post on which engines we check, as it stood on 23 September 2026, and carries that date in the sentence that says it; you cannot check it. Everything else carries a page name and a day.

Training data or live search: why ChatGPT can know your brand in one mode and not the other