[Pandora 22] Boosting Game Design with Analytics - Nikola Vasiljevic
Download as PPTX, PDF0 likes32 views
Nikola Vasiljevic discusses the importance of analytics in game design, emphasizing that it aids designers by providing insights from player feedback to align their vision with player expectations. The document outlines various stages of game development where insights and analytics play crucial roles, including pre-production, production, and post-production. It also differentiates between game user research and game analytics, highlighting the need for a combination of qualitative and quantitative data to enhance player experience and decision-making.
[Pandora 22] Boosting Game Design with Analytics - Nikola Vasiljevic
- 1. Nikola Vasiljevic – Head of Insights & Analytics at Mad Head Games Boosting Game Design with Analytics
- 2. Why do we need analytics for game design? Analytics’ role is to aid game designers by providing insights from player feedback that lead to more valid and reliable decision making. In that way, it helps bridge the gap between game designers’ vision and players’ expectations. Creative problem solving doesn’t come with a predefined set of steps we can follow. Instead, we explore the space of challenges*, diverging and converging our focus until we reach a state we’re satisfied with. Analytics help navigate the space of challenges. The key to success is for that state to be matching players expectations, because success is relative, and it’s anchored in delivering on one’s expectations. B A ASKING QUESTIONS, IDEA GENERATION CONCRETIZATION AND PROTOTYPING DECISION MAKING 1. Opening up the problem (Divergence) 2. Analyzing the problem (Emergence) 3. Closing the problem (Convergence) CHALLENGE SPACE Initial state Desired state *Gamestorming (2010) – Gray, Brown, Macanufo
- 3. Role of insights in each game-dev stage STAGE CONTENT SCOPE KEY INSIGHTS UNDERTAKINGS Pre- production Concept It is centred on the ideation of the game Competitor evaluation Documentation The core plan for features and game scope Define overall UX vision and ideal player experience from game features Prototype Initial implementations, like whitebox and user flows Usability testing on basic interaction and core loop Production Production This stage encompasses the main cycle of the main cycle of development, where content is added and the level of polish increases Evaluation of usability, behavioural, and attitudinal aspects of the player experience at each milestone Alpha Milestone reached when core gameplay features are implemented Usability testing on features and player experience evaluation Beta Milestone reached when a complete version of the game with full content is implemented Usability on onboarding, full-playthrough testing Gold Milestone reached when a version of the game is at quality and is ready to be launched for public release Playtesting on final balancing and tuning Post-production/Live Service It covers all actions after the game has been released, including patches, expansions, and live content Ongoing usability, balancing, and player experience evaluation as more content is released GA GUR Games User Research – Drachen, Mirza-Babaei, Nacke
- 4. Game User Research (GUR) and Game Analytics (GA) as players’ direct and indirect feedback 1. GUR: Gives insight into players’ experience - how players feel and what they think about our game. Answers the WHYs 2. GA: Gives insight into player behaviour – show us how the players are playing our game. Answers the WHATs 3. Visual-Audio Analytics (gameplay recordings with players’ comments): Observational techniques used as an addition to previuos methods; usefil in showing us what players are doing at some exact moment; audio let’s us record “heats of the moment“
- 5. GUR – Playtesting as a way to get players’ opinions and experience Upon playing a game, players fill out a questionnaire, professionally designed so as to avoiding auto-suggestive and biased questions. This is how we get players’ opinion on important game matters: • Usability: can players use the specific mechanics easily? Mechanics need to be clearly stated • Appreciation: do players like their overall experience of the game/ specific game moments? • Originality: does the game feel original or as something already seen • Clarity: do the players find the mechanics easy to grasp and use, and understand what they have to do in general? • Fun: do the players find the game fun to play? • Challlenge difficulty: how hard is it to face the game’s challenges, such as combat, puzzles etc.? • Balance: is any of the game’s experiences prevailing or everything is in balance? • Satisfaction with game elements: Narrative, pacing, sound, setting & atmosphere, graphics, characters, mechanics, animations, controls, UI, tutorials, navigation • HLW – highlights, lowlights, wishlist • Image questions: We can test for some statements about our game, i.e. “This is a high quality game.“ • Recommendation and Purchase Intent: Would players purchase the game if available now? Would they recommend it to their friends?
- 6. Game Analytics – Telemetry signals from the game that uncover patterns in player behaviour Game analytics involves using quantitative measures, metrics, and tools that can be used to track events that occur over the course of a game, with the goal of capturing such data for statistical analysis. It is based on factual data! Game analytics looks at users from two perspectives: • As customers – tracking retention/churn, conversion, LTV • As players – tracking their behaviour and experience
- 7. Game Analytics – The road to insights is paved with Data Pipeline Ensures a datastream from your in-game telemetry events to your database, and analytics tools. You can build it yourself, or you can opt-in for a end-to-end cloud solution. MongoDB Tableau Prep BigQuery Tableau MySQL Rstudio Message Broker NoSQL Database ETL Tool SQL Warehouse Analytics/ Vizualization Machine Learning • Accepts events from game client • Provides data structures • Sends them to the next data node • Stores telemetry events • Requires no structure • Provides flexibility and speed • Data cleansing • Standardization • Deduplication • Verification • Sorting • Provides structure • Data integrity/ACID • Complex querying • Reporting • Insights • Decision making • Forecasting/Predictive analytics (churn, conversion) • Classification • Segmentation • Sequence mining • Decision trees COLLECTION INJECTION PREPARATION COMPUTATION & PRESENTATION
- 8. Game Analytics – Events as building blocks of analytics Events are one-off in-game actions that are tracked alongside with a set of parameters that define them more closely. In specific occasions, you can track event streams – contextual sets of many events. A good example is combat event, which consists of many attacks, reloads, dodges, ability usages, consumables etc. List of some trackable events: • Attack • Death • Combat • Reload • Dodge • Position ping • Tutorial_watched • Puzzle_solved • Purchase made • … event_id weapon_mod combat_id weapon_upgrades attack_type weapon_durability attack_subtype death attack_dealer death_id attack_receiver combo_attack_name boss_fight distance_from_enemy combat_phase enemy_in_cover attack_success player_in_cover hitspot stat_effect_dealer weakspot stat_effect_receiver dmg_amount armor_type ammo_left_total weapon_type ammo_left_magazine weapon_name session_data_block session_id, session_duration common_data_block game_id, build_id user_data_block user_id, control_type, difficulty, graphics_setting timestamp, gametime, area_name, mission_name, objective_name, player_location_xyz, health, stamina, armor_upgrades, weapon_upgrades, status_effects, consumables_left standard_parameter_block data blocks Event: ATTACK parameters
- 9. General use cases of Game Analytics in Game Design 1. Identifying pain points and bottlenecks – discover where and why players get stuck 2. Feature usage/usability – discover do player use game features and mechanics (in intended way) 3. Balancing – get familiar with how different mechanics work together and tweak the balance between them 4. Player segmentation – discover different player’ segments based on their behaviour 5. Level design optimization – navigation heatmaps that tell us where players move, where they fight, die, explore, find secret areas and items 6. System analytics – discover if your system design behaves in intended way, i.e. NPC AI
- 10. Use cases from Scars Above 1. Balancing game difficulty in general 2. Balancing a boss fight 3. Examining players’ strategy to defeating enemies and bosses and if it’s in line with game designers intention 4. Balancing weapons in terms of general and situational usage 5. Identifying ammo depletion hotspots that break the gameplay balance 6. Balancing weakspot accuracy 7. Tweaking stamina depletion 8. Tweaking the ability system 9. Tweaking consumables and gadgets system 10. Basically, looking at how everything of the above comes together ...
- 11. Example: Combat analysis with GA QUANTITY, FREQUENCY, OUTCOME FUN, FAIRNESS… DIFFICULTY, NPC ACCURACY, STRATEGY… LOCATION CAUSE TIMING PLAYER STATE PREFERENCE, USAGE, ACCURACY, AMMO DEPLETION… RELOADING, DODGING, CONSUMABLES, ENVIRONMENTAL 1. Define your approach: • Predefined questions • “Ammo depletion in combat“ • “Are players using the consumables in combat?“ • Explorative approach • Usually it’s a combination of both 2. Choose your metrics and start adding insight layers to your area of interest
- 12. GUR (left) and GA (right) combo for Boss Fight Analysis Success factors: • Recognized the pattern • Successful switch from defense to offense • Weakspot accuracy • Consumable usage • Avoiding special attack • Constant movement instead of dodging
- 13. For takeaway 1. Combine GA and GUR to a holy grail to get the full picture about: • Players behaviour – the skeleton (usage data) • Players experience – the meat (attitude data) 2. Understand your data – noise, limitations etc. 3. Know your stakeholders’ needs and the decisions they need to make 4. Provide actionability – answer the questions WHAT | WHY | SO WHAT | WHAT NOW Give actionable recommendations that contribute to better decision making
- 14. THANK YOU FOR YOUR ATTENTION! Q&A nikola.vasiljevic@madheadgames.com