31:T91c,defintions = md`## Data source and definitions The v0.1 iteration of the dashboard uses Github data to identify and classify our users as follows: - __User__: A Github repo that is using our protocols in some way. We consider a user of either IPFS or Filecoin a user of our stack, though in future iterations this definition might be more strict. IPFS and Filecoin data are graphed separately (data is stored separately, and we do not deduplicate and combine for performance reasons). The repo meets at least one of the following criteria - Has at least one dependency on IPFS or Filecoin packages (direct or transitive) - Has at least 20 search hits in events across the last 90 days for either IPFS or Filecoin and is part of one of our collaborator orgs ([IPFS](https://ipfs.ecosystem-dashboard.com/collabs), [Filecon](https://filecoin.ecosystem-dashboard.com/collabs) - __Low-value (_high-value_)__: A repo that has at least 1 (_10_) associated [Github event(_s_)](https://docs.github.com/en/developers/webhooks-and-events/github-event-types) over a 14 day window, stars and member events excluded These definitions have been calibrated to optimize the capture of true positives while limiting noise ([research here](https://github.com/dchoi27/libin-ecosystem-dash-calibration/blob/master/ipfs-calibration.ipynb)). However, __you can use your own definitions via the same APIs this notebook is built on - more information [here](https://observablehq.com/@protocol/pmf-endpoints-repo-based-experiment)__! ## What this data is missing It is __important to contextualize this data__ via its shortcomings so it is appropriately interpreted when measuring PMF. The data: - Excludes all closed source repos that are dependent on our stack - Excludes incorporation of our technologies in projects that do not call our libraries as dependent (e.g., Fleek static website deployment) - Is not a measure of actual usage of our technologies, just development work that might lead to future usage (gated on these users achieving their own PMF) - Contains a good amount of noise, especially in the high-value user count (where a developer's side project they spend a few hours on for a week could appear as a "high-value" user) Future efforts will be spent improving this data as well as supplementing it to cover its blind spots!`32:[] 2f:{"title":"W3DT Product-Market Fit Dashboard v1 (IPFS and Filecoin data separate)","cells":[{"id":0,"value":"md`# W3DT Product-Market Fit Dashboard v1 (IPFS and Filecoin data separate)\n\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":870,"value":"md`\nThe trends and progress for our protocols' [product-market fit](#define_pmf). The metrics shown here are lagging indicators, showing progress as a result of our efforts.\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1016,"value":"md`\nThe table and diagram below show the number of users in each state of the [Libin model](https://www.youtube.com/watch?v=3XumZ_LT_ZE) of user activity over the last **two weeks**. We show the user volumes for Filecoin, IPFS, and the two combined, though the combined numbers contain some duplicates (generally <2% of IPFS users and ~40% of Filecoin users overlap). \n\nIn other diagrams below, we keep these numbers separate. __For a combined view of the data, use [this dashboard](https://observablehq.com/@protocol/w3dt-product-market-fit-dashboard-v1-beta-shared-data)__.\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":668,"value":"md`\n${tableMd(ipfsTransitions, filecoinTransitions, 1)}\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1174,"value":"md`In the current implementation _Users_ are Github repos that depend on 1 or more Filecoin, IPFS, IPLD or libp2p modules. _High-Value_ are repos with 10 or more [GitHub events](https://docs.github.com/en/developers/webhooks-and-events/github-event-types) over a 14 day window. _Low-Value_ are those with less than 10 but more than 0. See the [Definitions](#defintions) at the end for more detail.`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1170,"value":"md`### _The incredible state machine that answers all startup questions_`","pinned":false,"hidden":false,"mode":"ojs"},{"id":692,"value":"viewof showEmoji = Radio([\"IPFS\", \"Filecoin\"], {value: 'IPFS', format: x => x === 'IPFS' ? '🧊 IPFS' : '⨎ Filecoin'})","pinned":false,"hidden":false,"mode":"ojs"},{"id":490,"value":"emojies = {\n if(showEmoji === 'IPFS') return emoji(1, ipfsTransitions, ipfsStates, {legend: true})\n \n return emoji(1, filecoinTransitions, filecoinStates, {legend: true})\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":912,"value":"md`\nHover over a transition to see the percentage of the outflow that the transition represents from its previous value, which let's us see numbers like churn from high-value to inactive and bounce from first-time to inactive.\n\n__Change the period these visualizations are looking at [here](https://observablehq.com/@protocol/w3dt-transitions-focused-dashboard)!__\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":706,"value":"md`---`","pinned":false,"hidden":false,"mode":"ojs"},{"id":926,"value":"md`\n## All users by transition\nThe total number of active users over time, split by the volume in each transition. These transitions are colored by the destination, so all high-value users are colored a shade of green, inactive red, etc. Inactive to inactive transitions are omitted.\n\nWe use color to denote where the state the user is ends up in. All the greens are users flowing to High-Value. yellows for all paths to Low-Value, and red for all paths to Inactive.\n\n__Additional visualizations drilling into the overall Active User population can be found [here](https://observablehq.com/@protocol/pmf-dashboard-active-population)__!\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1058,"value":"allTransitions(ipfsTransitions, 'IPFS Transitions')","pinned":false,"hidden":false,"mode":"ojs"},{"id":1061,"value":"allTransitions(filecoinTransitions, 'Filecoin Transitions')","pinned":false,"hidden":false,"mode":"ojs"},{"id":919,"value":"md`\n## High-Value focus: State volume and growth\n\nHigh-Value User volume over time is shown by the line, with % change from the previous two week period shown by the bars. Ideally, we'd see exponential growth in users (hockey stick-shaped line), with green bars that are level to rising over time. \n\nOur current target for end of Q2'20 is a >5% week-over-week growth in high-value users (which translates to a >~10% growth rate across two week periods).\n\n__Additional visualizations drilling into High-Value Users can be found [here](https://observablehq.com/@protocol/pmf-dashboard-high-value-users)__!\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1035,"value":"ipfsHighValue = highValueUsers(ipfsStates, 'IPFS High Value User Volume & Growth Rate', {showTrailing: true})","pinned":false,"hidden":false,"mode":"ojs"},{"id":1226,"value":"filecoinHighValue = highValueUsers(filecoinStates, 'Filecoin High Value User Volume & Growth Rate', { showTrailing: true })","pinned":true,"hidden":false,"mode":"ojs"},{"id":822,"value":"md`---`","pinned":false,"hidden":false,"mode":"ojs"},{"id":925,"value":"md`\n## New User focus: Bounce rate over time\nShows bounce rate over time (new user to inactive), as well as the rate of new users following the \"golden path\" (new users to high-value). New user volume is shown for the period \\`\\ n-1\\`\\ to put the rates in context.\n\nOur current target for end of Q2'20 is a <75% bounce rate.\n\n__Additional visualizations drilling into New Users can be found [here](https://observablehq.com/@protocol/pmf-dashboard-new-users)__!\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1116,"value":"bounceRate(ipfsTransitions, 'IPFS Bounce Rate', { showTrailing: true })","pinned":false,"hidden":false,"mode":"ojs"},{"id":1118,"value":"bounceRate(filecoinTransitions, 'Filecoin Bounce Rate', { showTrailing: true })","pinned":false,"hidden":false,"mode":"ojs"},{"id":846,"value":"md`---`","pinned":false,"hidden":false,"mode":"ojs"},{"id":928,"value":"md`\n## Users going inactive\nShows the source of inactive users over time. This just shows the inflows - a small proportion of users reactivate (e.g., if a collab took some time off development of a specific repo) so the total volume under the graph doesn't exactly show total inactive popualtion, but it's a useful way to see the sources of inactive users and how this changes over time.\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1046,"value":"viewof showSideBySide = Radio([\"IPFS\", \"Filecoin\", \"Both\"], {value: 'Both', format: x => x === 'IPFS' ? '🧊 IPFS' : x === 'Filecoin' ? '⨎ Filecoin' : 'Both'})","pinned":false,"hidden":false,"mode":"ojs"},{"id":849,"value":"sideBySide(inflowsToInactive(ipfsTransitions, 'IPFS users going inactive'), inflowsToInactive(filecoinTransitions, 'Filecoin users going inactive'))","pinned":false,"hidden":false,"mode":"ojs"},{"id":206,"value":"md`---`","pinned":false,"hidden":false,"mode":"ojs"},{"id":977,"value":"define_pmf = md`\n## What is _Product Market Fit_ (PMF)\nPMF can be described as _“accelerating growth in the number of high-value users of a product (aka those that get significant value from the product)”_. __We model this user cycle using [Phil Libin's user state machine](https://www.youtube.com/watch?v=3XumZ_LT_ZE)__. In this dashboard, “high-value users” are developers actively building on top of our stack. Successfully achieving PMF will look like accelerating growth of high-value users: both more users getting high value from the stack, and fewer users attriting (becoming inactive). More discussion on this can be found [here](https://observablehq.com/d/536c37787c312370).\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1148,"value":"md`_Individual Project Teams who are looking to improve PMF in focused ways should supplement this data with PMF leading indicators that are specific to tracking progress of the specific area they are trying to improve (e.g., bounce rate of docs websites for Developer Onboarding)._`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1098,"value":"$31","pinned":false,"hidden":false,"mode":"ojs"},{"id":1214,"value":"md`\n---\nPeriod:\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1218,"value":"md`---`","pinned":false,"hidden":false,"mode":"ojs"},{"id":902,"value":"viewof historicalWeeks = Range([1, totalWeeks - 1], {value: 0, label: \"weeks ago\", step: 1})","pinned":false,"hidden":false,"mode":"ojs"},{"id":731,"value":"sideBySide = (..._graphs) => {\n const graphs = showSideBySide === 'Both' ? _graphs : showSideBySide === 'IPFS' ? [_graphs[0]] : [_graphs[1]];\n const container = html`

`;\n container.innerHTML = graphs.map(_ => '

').join('')\n graphs.forEach((graph, index) => container.children[index].appendChild(graph))\n return container;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":550,"value":"import { churnRate } from '@protocol/churn-rate'","pinned":false,"hidden":false,"mode":"ojs"},{"id":956,"value":"import { emoji, graphStyle } from '@protocol/transitions-v2';","pinned":true,"hidden":false,"mode":"ojs"},{"id":582,"value":"import { bounceRate } from '@protocol/bounce-rate'","pinned":false,"hidden":false,"mode":"ojs"},{"id":601,"value":"import { ipfsActiveUsers, filecoinActiveUsers } from '@protocol/active-user-volume-growth'","pinned":false,"hidden":false,"mode":"ojs"},{"id":842,"value":"import { inflowsToInactive } from '@protocol/inflows-to-inactive'","pinned":true,"hidden":false,"mode":"ojs"},{"id":829,"value":"import { allTransitions } from '@protocol/all-transitions';","pinned":false,"hidden":false,"mode":"ojs"},{"id":1223,"value":"import { highValueUsers } from '@protocol/high-value-user-volume-growth-rate'","pinned":true,"hidden":false,"mode":"ojs"},{"id":665,"value":"import { tableMd } from '@protocol/weekly-user-model';","pinned":false,"hidden":false,"mode":"ojs"},{"id":522,"value":"import {Range, Radio} from \"@observablehq/inputs\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":764,"value":"import {ipfsGetTransitionsUrl, ipfsGetStatesUrl, filecoinGetTransitionsUrl, filecoinGetStatesUrl} from '@protocol/pmf-endpoints-repo-based-experiment';","pinned":false,"hidden":false,"mode":"ojs"},{"id":783,"value":"startDate = {\n const t = new Date()\n t.setDate(t.getDate() - (7 * (totalWeeks)))\n return t.toISOString().split('T')[0];\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":898,"value":"totalWeeks = 40","pinned":true,"hidden":false,"mode":"ojs"},{"id":762,"value":"ipfsTransitions = (await fetch(`${ipfsGetTransitionsUrl}?start_date=${startDate}`)).json()","pinned":false,"hidden":false,"mode":"ojs"},{"id":797,"value":"filecoinTransitions = (await fetch(`${filecoinGetTransitionsUrl}?start_date=${startDate}`)).json()","pinned":false,"hidden":false,"mode":"ojs"},{"id":778,"value":"ipfsStates = (await fetch(`${ipfsGetStatesUrl}?start_date=${startDate}`)).json()","pinned":false,"hidden":false,"mode":"ojs"},{"id":793,"value":"filecoinStates = (await fetch(`${filecoinGetStatesUrl}?start_date=${startDate}`)).json()","pinned":true,"hidden":false,"mode":"ojs"},{"id":400,"value":"md`### Styles`","pinned":false,"hidden":false,"mode":"ojs"},{"id":493,"value":"graphStyle","pinned":true,"hidden":false,"mode":"ojs"},{"id":643,"value":"customCss = html``","pinned":false,"hidden":false,"mode":"ojs"},{"id":402,"value":"html`\n`","pinned":false,"hidden":false,"mode":"ojs"}],"files":"$W32","stdlib":"1","titling":"auto"}