Getting Started

Getting started with llmeknow

What this product is for

llmeknow measures AI Perception: how large language models describe, compare, recommend, or omit brands, organisations, people, and ideas when people ask everyday questions. You set up a market and a study, run a wave across chosen models and audience segments, then read the responses, entities, sources, and reports those waves produce.

Create your account with no card and explore for 30 days with no subscription fee. No subscription starts unless you choose one. You pay for the AI responses you generate: review the estimated cost and add funds before running a wave.

The core loop

  1. Market: the category and geography you care about (for example car insurance in South Africa, or municipal housing policy in a city).
  2. Study: the questions, entity classes, segments, models, and intent for that market.
  3. Wave: one full execution of the study. Repeat waves over time when you want trend, not a one-off snapshot.

Home walks you through this when the workspace is empty. Glossary in the sidebar defines the same terms in one place.

Choose a path by objective

You can use the same machinery for different jobs. Start from the outcome you need, then set study intent, questions, and analysis tools to match.

Brand and marketing work

Use this when you care how models talk about your brand versus competitors: who gets recommended, which attributes stick, and where you are missing from the answer set.

  • Set study intent to Representing your brand or Tracking competitors.
  • Keep questions neutral enough that you are not priming one brand; see avoiding context leak.
  • Lean on entity classes for brands and attributes, then clean entities before trusting Share of Voice.
  • Use segments when the interesting gap is by audience (for example first-time buyers versus switchers), not when every demographic dial is turned on at once; see designing segments.

PR and reputation work

Use this when the question is how models frame your organisation after news, crises, or campaigns: tone, associations, and which stories get repeated.

  • Prefer Neutral research or a short custom intent that names the reputation question you are answering.
  • Write questions the way a journalist or concerned citizen would ask them, not as brand slogans.
  • Watch sources and source quality as carefully as the answer text; reputation often rides on what the model cites.
  • Use co-occurrence and relation views to see what names travel with yours, and keep co-occurrence distinct from direct attribution.

Socio-political and public-affairs work

Use this when the subject is policy, institutions, public figures, or contested social questions rather than product brands.

  • Markets still need a clear geography and category (who is being asked about, and in which place).
  • Questions should stay open and parallel across actors; naming one party and not others tilts the wave the same way brand context leak does.
  • Entity classes often expand beyond brands into organisations, people, programmes, and issue labels. Treat cleanup as part of the method, not an afterthought.
  • Narrative tracking is usually the right extra axis here: you define statements you care about (for example a policy claim), then score how strongly responses agree or disagree with those statements across models and segments. Attach a taxonomy when you need stance over time, not only entity mention share.

Parts of the product you will use most

Segmentation

Segments are audience groups defined by attributes (demographics, attitudes, context). The platform generates synthetic individuals inside each segment and asks your questions as those people. Use few attributes that match the decision you need; overloaded personas overfit. Always keep a baseline (no segment) so you can see what the models say without audience context.

Entity classes and cleanup

After a wave, extraction pulls named things into classes (brands, attributes, and any custom classes you configured). Analytics join on those classes. Before you quote Share of Voice or rankings, merge duplicates, drop noise, and reclassify mis-typed entities. When you change the entity set, refresh the report after cleanup so charts and write-ups match the cleaned data.

Narrative tracking

Narrative tracking scores stance toward a curated set of statements (theme → topic → statement). It sits beside entity tracking when the question is agreement or disagreement with claims, not only whether a name appeared. Use it for policy debates, reputation claims, and message testing. Attach a taxonomy to the study, optionally backfill older responses, then read stance by model and segment on the Narratives views.

Questions, repetitions, and models

Question design drives almost everything downstream. Pair intent with careful phrasing (study intent, context leak). Use enough repetitions that a single lucky or unlucky answer does not define the result. Pick models that matter to your audience; compare them rather than assuming one voice equals the market.

Sources and reports

Models that search the web leave sources on responses. Treat those as evidence to inspect, not as proof by volume. The AI overview and exportable reports summarise a wave for stakeholders; they are only as good as the cleanup and intent behind them.

A practical first week

  1. Create one market with a real geography and category you already understand.
  2. Create one study with a clear intent and a short question set (five strong questions beat fifteen vague ones).
  3. Add one or two segments only if the decision depends on audience difference; otherwise run baseline alone first.
  4. Run a wave when you have wallet credits, then clean entities before sharing numbers.
  5. Add narrative tracking if your questions are about claims and stance, not only brand presence.
  6. Come back for a second wave when something in the world changes (launch, crisis, policy shift), so you can read movement rather than a single cut.

People, projects and AI usage

Each team member signs in with their own account and joins your organisation. Editing and admin roles can work on studies; viewers can read the results. Inviting a colleague gives them access to the shared workspace, rather than a separate copy of its studies. Check Team settings for your organisation's included seats and any added-seat charge before inviting someone.

A study slot is used when a study first runs. You can run further waves in that same study without using another study slot, although each wave still uses funds for AI responses. Archiving a study does not release its slot. Check Billing for your organisation's study allowance; the number of people and the number of studies are different limits.

Example: an agency's first client benchmark

Start with one client study, five questions, two selected models, no audience segments and three repetitions. This produces 30 baseline responses per wave: 5 questions × 2 models × 3 repetitions. Run the same study again next month to compare another 30 responses over time, using the same study slot.

If you add two audience segments with one synthetic individual in each, the baseline stays included. The same setup then produces 90 responses per wave: 5 questions × 2 models × 3 repetitions × 3 audience contexts (baseline plus two individuals).

These are response counts, not a promise of how many responses your balance will buy. Before running, review the estimate for the models you selected, your actual question text, web search, extraction and language settings. Those choices affect the cost. Your organisation's shared funds pay for usage across its studies; adding people does not create a separate balance for each person.

Related reading

Subscription, editing seats and research funds

Your subscription provides platform access and study capacity. Your organisation shares one research wallet. AI response costs depend on the configuration of each wave, so review the estimate before every run, including repeat waves. Included editing seats do not multiply study capacity.

The organisation-pricing terms below apply only when Billing offers them to your organisation. If no organisation-pricing quote is available, these terms are not available for your account: use the prices, seat allowance and renewal dates in Billing or your existing agreement. This guide does not enable organisation pricing or change an existing contract.

When this offer is available, Basic includes 3 editing/admin seats and Pro includes 10, counting the owner. Viewer seats are separate and free, with their own allowance. The owner reviews and accepts the organisation quote before starting checkout.

For an available standard organisation-pricing offer, base prices are Basic R1,875 per month or R18,000 per year, and Pro R3,749 per month or R35,988 per year. Annual checkout is available only when offered in Billing; its base fee is paid upfront. Under this offer, additional editing/admin seats renew monthly: Basic R937.50 and Pro R562.35 on monthly organisation plans; Basic R750 and Pro R449.85 on annual organisation plans. When Team offers a mid-cycle addition, its quote shows the prorated immediate charge and next monthly renewal before confirmation.

You pay the normal yearly price and receive your plan's full annual research budget (R3,600 on Basic, R6,000 on Pro) after confirmed payment. If you join on the 29th to 31st, your renewal falls on the 28th. Review the dates at checkout before paying.

Standard monthly subscription payments add R300 on Basic or R500 on Pro to the research wallet. Standard annual subscription payments add R3,600 on Basic or R6,000 on Pro upfront. Extra-seat payments do not add wallet funds. Complimentary, internal and invoiced arrangements follow their own agreement. The wallet ledger shows delivered funds; a scheduled plan change or card verification is not a wallet payment.

After an organisation-pricing quote is accepted, included editing seats activate when the first subscription payment is confirmed. If the accepted quote schedules Basic after no-fee access, it shows the start date and any R1 card verification separately. The R1 verification is added to the wallet; it does not activate the paid editing-seat allowance.

If Billing offers a tier change for an adopted organisation plan, it takes effect at the next organisation renewal. On an annual plan, that is the annual renewal, not a monthly extra-seat payment. If Billing offers annual cancellation, the quote preserves the paid annual term and shows the monthly extra-seat payments through that term. Review the effective dates before confirming. A pending request can lock team changes until its outcome is confirmed. Use a reversal only when Billing offers it; otherwise contact support for review. After a connection failure, resume the same saved request so its quote and reference are preserved.

If an extra-seat billing period crosses the annual term end, the annual quote shows a final prorated amount and its start and end dates. This replaces the full monthly extra-seat payment for that final period. Review the schedule before confirming, then check that the request is confirmed as scheduled. If the action is unavailable or needs review, contact support.

All help articles · Glossary