DMI Expert flashcards
DMI Expert · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Porter's Five Forces
- Framework analyzing industry attractiveness through rivalry, supplier power, buyer power, threat of new entrants, and threat of substitutes; strong forces compress margins and lower profit potential.
- SMART goal criteria
- Goals must be Specific, Measurable, Achievable, Relevant, and Time-bound; without a metric, baseline, and deadline the goal cannot be evaluated or prioritized.
- OKR structure
- Objectives are qualitative, ambitious, and time-boxed; Key Results are three to five quantitative outcomes that prove progress. Tasks and deliverables are not Key Results.
- Ansoff growth matrix
- Four growth options by product and market: market penetration, product development, market development, and diversification. Diversification carries the highest risk because both dimensions are new.
- RACI responsibility matrix
- Assigns Responsible, Accountable, Consulted, and Informed roles per task. Exactly one person should be Accountable for each decision to avoid diffused ownership.
- LTV to CAC ratio
- Customer lifetime value divided by customer acquisition cost. A ratio of at least 3:1 with payback under twelve months is a common threshold before scaling paid acquisition.
- Digital maturity model
- Staged progression from ad hoc to optimized digital capability, typically through emerging, structured, and integrated stages. Governance and data foundations usually precede scale-up investment.
- Pareto prioritization
- Ranking initiatives by expected impact divided by effort, based on the observation that roughly 20 percent of levers often drive about 80 percent of incremental value.
- Ad Rank formula
- Ad Rank equals bid multiplied by Quality Score plus expected impact of extensions and formats. Higher quality can let a lower bid outrank a higher one.
- Quality Score components
- A zero-to-ten score built from expected click-through rate, ad relevance, and landing page experience. Higher scores usually reduce cost per click and improve position.
- Search match types
- Exact matches the keyword closely, phrase requires the meaning, and broad reaches related queries. Broad and phrase need negative keywords and audience signals to control spend.
- Breakeven ROAS formula
- Breakeven ROAS equals one divided by gross margin. A 25 percent margin gives a 4.0 breakeven, meaning revenue must be four times ad spend to avoid loss.
- Impression share diagnostics
- Impression share is impressions divided by eligible impressions. Lost share to budget suggests raising spend; lost share to rank suggests improving quality or bids.
- Crawl versus index control
- Robots.txt and meta robots control crawling; noindex and canonical tags control indexing. XML sitemaps aid discovery but do not guarantee ranking or indexing.
- Core Web Vitals thresholds
- At the 75th percentile: Largest Contentful Paint at or under 2.5 seconds, Interaction to Next Paint at or under 200 milliseconds, and Cumulative Layout Shift at or under 0.1.
- Search intent tiers
- Informational, navigational, commercial investigation, and transactional. Each maps to a page type such as guide, brand page, comparison, or product and checkout page.
- Negative keyword workflow
- Review search terms regularly and add high-spend, zero-conversion queries as negatives. This protects budget on broad and phrase match campaigns.
- SEO canonical tag
- An HTML link element that tells search engines which URL is the preferred version of a page, consolidating duplicate URLs and concentrating ranking signals.
- Engagement rate definition
- Interactions such as likes, comments, shares, and saves divided by reach, impressions, or followers. Always state the denominator because values are not comparable across definitions.
- Reach versus impressions
- Reach counts unique accounts that saw content; impressions count total displays. Frequency equals impressions divided by reach, and high frequency with flat reach signals creative fatigue.
- Social funnel mapping
- Awareness maps to reach and video views; consideration to saves, shares, and link clicks; conversion to leads, purchases, and return on ad spend.
- Share of voice formula
- Brand mentions divided by total category mentions. Used in social listening to compare competitive visibility and track earned conversation over time.
- Sentiment ratio formula
- Positive mentions divided by positive plus negative mentions. Neutral mentions are excluded so the ratio reflects polarized opinion rather than total volume.
- Creator disclosure requirement
- Paid partnerships must be clearly labeled with a platform paid partnership tag or hashtag such as ad. Briefs should also specify usage rights and exclusivity windows.
- Community management SLA
- A service-level agreement setting first-response times, commonly one to four business hours for public complaints, with escalation paths for crisis keywords to legal or public relations.
- Creative fatigue signal
- Rising frequency with flat or declining reach and falling engagement rate. Usually resolved by refreshing hooks, thumbnails, or formats rather than increasing budget.
- Paid social bid objectives
- Cost per mille for reach, cost per click for traffic, and cost per acquisition or return on ad spend for conversions. Optimize to the revenue-closest event with sufficient volume.
- Supervised versus unsupervised learning
- Supervised learning uses labeled data to predict outcomes; unsupervised learning finds structure such as clusters without labels. Reinforcement learning optimizes sequential decisions via reward signals.
- Train validation test split
- Data divided into training, validation, and test sets. Good training performance with poor held-out performance indicates overfitting rather than genuine learning.
- Precision and recall tradeoff
- Precision is true positives divided by predicted positives; recall is true positives divided by actual positives. The threshold should reflect the relative cost of false positives and false negatives.
- Retrieval-augmented generation
- A method that retrieves relevant documents and supplies them to a generative model before answering. It reduces hallucination but does not eliminate it, so sources should be cited.
- Prompt structure elements
- A strong prompt states role, task, context, constraints, and output format, optionally with few-shot examples. Specifying format such as JSON or a table reduces rework.