Marketing decisioning glossary

Insika’s term for an AI system that holds the ongoing growth mandate a brand would traditionally give an agency of record: monitoring the market, planning, recommending and executing, with people approving each decision.

An approach to building AI products in which large language models are called only when they are needed, while purpose-built agents and statistical models handle most of the data processing and reasoning. It keeps answers grounded in real numbers and keeps compute costs down.

In Insika: Insika is built this way: frontier models handle language, and Insika’s own models handle the numbers.

How often AI assistants and AI search engines mention or cite a brand, which of its pages they draw from, and how that compares with competitors.

In Insika: Insika tracks AI citations for your brand and your competitors over time.

Total revenue divided by total marketing spend across all channels for a period. Because it does not depend on channel-level attribution, many direct-to-consumer brands use it as a top-line check on marketing efficiency.

In Insika: Insika alerts you when blended MER drops, traces the cause and recommends where to move spend.

A single score that summarizes the health of a brand, typically built from signals such as customer sentiment, reviews and search interest.

In Insika: Insika’s Brand Health Index breaks the score out by product, quality, pricing and service and tracks week-over-week movement.

A way of working in which the results of each decision are measured and fed back into planning, so the next decision is better informed than the last.

In Insika: Insika feeds results back into its models automatically.

Ongoing tracking of competitors’ campaigns, messaging, media investment, search presence and pricing, used to spot threats and openings early.

In Insika: Insika tracks competitors weekly by region and flags who is outspending you and which channels are still open.

A short document that sets out the objective, audience, message, deliverables and timeline for a piece of creative work, so everyone involved is working toward the same goal.

In Insika: Insika generates creative briefs informed by your historical performance and current customer sentiment.

Measuring how customers feel about a brand or product by analyzing what they say in reviews, social posts, forums and videos.

In Insika: Insika flags negative reviews and posts as they appear, with the source attached.

The research and influence that happens where analytics tools cannot see it, such as private messages, communities, podcasts, word of mouth and AI assistants. Buyers often arrive having already decided because of it.

Software that turns data into specific, ranked decisions rather than reports. A marketing decisioning engine connects data, models the likely outcome of each option and recommends the best move.

In Insika: Insika is a decisioning and execution engine: it can also carry the approved move into market.

The point at which each additional dollar spent on a channel produces less result than the dollar before it. It is the reason a channel’s best budget is rarely its largest possible budget.

The share of a result that would not have happened without a specific marketing activity. Measuring it separates the sales a channel truly caused from sales that would have happened anyway.

The three steps of Insika’s workflow. Ingest connects your data and market signals. Decide weighs every signal against your goals and recommends the best move. Act puts the approved move into market and keeps it optimized within your limits.

A statistical method that estimates how much each marketing channel, along with factors such as seasonality, pricing and promotions, contributes to sales or another outcome, using aggregated historical data. Because it does not rely on tracking individual people, it holds up as cookies and device IDs disappear. Teams use it to decide how to reallocate budget for a better return. Also called marketing mix modeling.

In Insika: Insika runs media mix models on your own data and updates them continuously.

The single most valuable move available right now, chosen by weighing the expected impact, cost and confidence of each option.

In Insika: Every Insika recommendation comes with its projected impact and model confidence.

Revenue attributed to advertising divided by what was spent on that advertising. Unlike blended MER, it depends on how revenue is attributed to each channel.

A curve showing how a channel’s results respond as spend increases. Early spend tends to produce strong returns; past a certain point each additional dollar produces less and the curve flattens. It shows where a channel is near its ceiling and where more budget would still pay off.

In Insika: Insika uses saturation curves to recommend moving spend from saturated channels to channels with room to grow.

Modeling the likely outcome of a decision before making it, for example how revenue would change if 20% of one channel’s budget moved to another.

In Insika: You can ask any what-if question and see the projected impact in seconds.

A brand’s share of the total conversation, advertising or search visibility in its category, compared with competitors.

A plan that sets marketing goals for a quarter or year and lays out the audiences, positioning, channels, budget and measures used to reach them.

In Insika: Insika builds strategic plans from your historical performance, market trends and sentiment.