UI

Disconnect your Google account from Topify

Goal: stop Topify from accessing your Google Search Console (GSC) data and revoke OAuth tokens. Topify connects to GSC in read-only mode (webmasters.readonly) to import and visualize your search data. You can disconnect at any time. What happens when you disconnect - Immediate effect: Topify revokes OAuth tokens and stops all data syncs, exports, and background jobs. - Dashboard: existing charts remain, but no new data will be pulled until you reconnect. - Data handling: we delete or anonymize associated Google user data within 30 days after disconnect (see our Privacy Policy). - You can reconnect later without losing your Topify account. Disconnect in Topify (recommended) 1. Open Topify in your Shopify admin. 2. Go to Settings → Connected Google account. 3. Click Disconnect and confirm. That’s it—tokens are revoked and sync stops immediately. Tip: Managing multiple storefronts? Disconnecting affects only the current store’s connection. Disconnect from your Google Account (alternative) You can also revoke access directly in Google: 1. Visit myaccount.google.com → Security. 2. Under Third-party access, open Manage third-party access. 3. Find Topify and click Remove access. If you remove access this way, open Topify once more— you’ll see a prompt to reconnect. Reconnect later Any time, go to Topify and click Connect Google Account. Choose the Google account that has access to your shop’s GSC property and follow the prompts. FAQ Will disconnecting affect my search rankings or site? No. Topify only had read-only access; disconnecting simply stops data imports. Do you keep my data after disconnect? We revoke tokens immediately and delete or anonymize related Google user data within 30 days. Can I disconnect one store but keep another connected? Yes. Connections are store-specific.

Dashboard overview

The Dashboard is your home base in Topify. It pulls full Google Search Console (GSC) data (up to 50,000 rows/day) and turns it into clear charts and tables so you can spot wins, catch drops, and act fast. 1) Filters bar (top) Use the filters to focus the dashboard on exactly what you need: - Date range — pick any range; compare to Previous period or Year over year. - Group by — switch aggregation for charts: Day / Week / Month. - Search type — Web (default). - Query / Page (URL) — filter by keyword or a specific page. - Country / Device — slice performance by market and device type. - Branded filter — include/exclude branded traffic (defined in Settings → Branded keywords). - Topics / Clusters — filter by your manual groups (set in Settings → Topics & Clusters). Tip: Click Reset filters anytime to return to the full view. 2) KPI cards (above the main chart) Four summary cards show totals for the selected filters and time range: - Impressions — how many times your pages were shown in Google results. - Clicks — visits from Google Search. - CTR — Click-through rate = Clicks ÷ Impressions. - Average Position — average rank of your highest listing (lower is better). Each card shows a ▲/▼ delta versus the previous comparable period (day-of-week matched). Hover any card to see a tooltip with definitions. Optional (if enabled): CCR (Crawl-to-Click Ratio) — estimated bot fetches per human click. Helpful to spot “answer-in-SERP” effects or snippet issues. 3) Main performance chart A combined time series that plots your core KPIs: - Lines: Clicks, Impressions, CTR, Average position (position uses an inverted scale so downward movement is an improvement). - Grouping: Day / Week / Month. - Legends & toggles: show/hide any metric; hover to inspect exact values. - Annotations lane: click on a date to add a note (e.g., “title tests”, “sale launched”, “theme update”). Annotations help explain changes later. Tip: Expand to full-width for presentations or reporting. 4) Countries & Devices Two compact tables for distribution by country and device: - Columns: Entity, Clicks, Impressions, CTR, Avg. position. - Views: Popular / Growing / Decaying. - Use these to spot markets or device types that need attention (e.g., mobile CTR drop). 5) Pages table Performance by URL: - Columns: Page (URL), Clicks, Impressions, CTR, Avg. position. - Views: Popular / Growing / Decaying. - Actions: click a page to drill down into its top queries, country/device split, and trend. Tip: Use this with branded = off to gauge true SEO gains. 6) Queries table A sortable table of search terms that triggered your pages: - Columns: Query, Clicks, Impressions, CTR, Avg. position. - Views: - Popular — sorted by impressions/clicks. - Growing — positive trend vs previous period. - Decaying — losing traffic or rank. - Filters: all top filters apply (date, page, country, device, branded, topics/clusters). - Actions: click a query to open the detail panel (top pages, trend sparkline, quick filters). - Pagination: use the pager or export (see below). 6) Countries & Devices Two compact tables for distribution by country and device: - Columns: Entity, Clicks, Impressions, CTR, Avg. position. - Views: Popular / Growing / Decaying. - Use these to spot markets or device types that need attention (e.g., mobile CTR drop). 7) Data freshness - On first connect, Topify fetches the last 16 months of GSC data (up to 50,000 rows per day). - Daily syncs run automatically; fresh data usually appears within minutes. - If you see gaps, check your filters (date, branded, search type) and the selected property. 8) How deltas and trends are calculated - Comparison uses a previous comparable period with day-of-week matching to avoid weekend/weekday skew. - Growing / Decaying views highlight statistically meaningful movement in clicks/CTR/position within the selected timeframe. 9) Common workflows - Find quick wins: Queries → Growing + filter positions 6-10 (Near-Top 5) → update titles/links. - Diagnose a drop: Pages → Decaying → open details → check queries, device, country → add an annotation for fixes. - Report progress: Set branded = off, choose Last 28 days vs previous, screenshot KPI cards + main chart, export the Queries table. FAQ Why are numbers different from the GSC UI? Topify ingests the full dataset (not the 1,000-row UI limit) with your filters, so totals can differ from sampled views in the UI. What does “Average position” mean? It’s the average rank of your highest listing per query; lower = better. It can move even when clicks stay flat. Can I exclude branded searches? Yes. Define branded terms in Settings → Branded keywords, then toggle Branded in filters. How do annotations work? Click any date on the main chart to add a note. Annotations are shared with your team and appear in reports.

Using filters in Topify

Filters let you focus the dashboard on exactly the traffic you care about—by date, query, page, market, device, branding, and your own topics/clusters. This guide explains each filter, how they combine, and a few quick recipes. Where filters apply - Everything on the Dashboard: KPI cards, the main time series, and the Queries / Pages / Countries / Devices tables. - Exports: CSV/Sheets exports respect the filters you set. - Share links: Copy link preserves current filters so teammates see the same view. Logic: Filters across different categories combine with AND (e.g., Date AND Country AND Device). Within a single multi-select (e.g., multiple countries) they combine with OR. Filter types 1) Date range Pick any range (e.g., Last 7 / 28 / 90 days, custom dates). - Group by controls chart aggregation: Day / Week / Month. - Compare to adds deltas vs Previous period or Year over year with day-of-week matching (fair weekend/weekday comparisons). Tip: If today looks low, remember GSC may delay the latest day’s data. 2) Country Slice performance by market. Supports multi-select (e.g., US OR CA). 3) Device Desktop, Mobile, Tablet. Multi-select allowed. 4) Page (URL) Focus on specific pages or URL patterns. - Type/paste a full URL or a prefix (e.g., /collections/ to target a section). - Click a row in Pages to auto-filter by that page. 3) Query Filter by keywords users searched. - Partial matches work (e.g., boots). - Use Branded filter (below) to include/exclude brand terms globally. - Click a row in Queries to auto-filter by that query. Common recipes - Find money keywords near the top: Date: Last 28 days → Branded: Off → Queries → view = Popular → sort by Avg. position 6–10 → pick candidates for title/UX tweaks. - Diagnose a mobile dip in the US: Date: Last 14 days vs previous → Country: US → Device: Mobile → check CTR and Pages → Decaying to find what slipped. - Assess a collection launch: Date: Since launch → Page: /collections/new-arrival → switch Group by: Week → add annotation; track CTR & position trend. Tips & best practices - Reset quickly: Use Reset filters to clear all and return to the full view. - Share context: Copy link to share a filtered view with teammates or clients. - Start broad, then narrow: Begin with date/country/device, then add query/page. - Branded first: Define branded terms early—your non-brand trends become much clearer. - Use annotations: When filters explain a change (sale, title tests), add an annotation so the story is obvious later. Troubleshooting - Empty state / missing rows: Check the date range, search type, and make sure the selected GSC property actually has data for those filters. - Numbers differ from GSC UI: Topify loads the full dataset (up to 50,000 rows/day) with your filters, while the GSC UI shows a sampled 1,000-row view. Totals can differ—Topify is more complete. - Brand filter not working as expected: Review Settings → Branded keywords. Add variations and misspellings; the filter uses “contains” logic. - Slow table with very wide ranges: Narrow the time window or apply a filter (country/device/page) to focus results; export for deeper offline analysis. FAQ Do filters affect exports? Yes—exports reflect your current filters. Can I save filter presets? You can use share links today; named presets are on our roadmap. Does Topics/Clusters change my raw data? No. They’re overlay groupings you control; raw GSC data stays intact.

Sorting data in tables

Topify tables (Queries, Pages, Countries, Devices) let you sort results so you can spot what matters first—top performers, rising opportunities, and items that need attention. Where sorting applies - All four tables on the Dashboard. - Detail drawers opened from a row. - Exports respect your filters; the sort is applied to the on-screen view (and used in CSV/Sheets when supported by your browser). Sort views: Popular, Growing, Decaying Each table has a view selector. It changes the default sort logic for that table: Popular (default) Surfaces high-impact rows first. - Queries/Pages: primarily by Clicks (desc), with Impressions as a tiebreaker, then Avg. position and CTR. - Countries/Devices: by Clicks (desc), then Impressions. Use it for: weekly reporting, “top of the pile,” and sanity checks. Decaying Prioritizes rows with significant negative movement vs the previous comparable period. - Mirrors Growing but for declines (drops in clicks/CTR or worse position), with the same minimum-volume guardrails. Use it for: early-warning triage, pages to refresh, queries losing intent match, device-specific regressions. Growing Highlights rows with meaningful positive movement vs the previous comparable period (with day-of-week matching). - We compute a Growth Score that blends: - Δ Clicks (relative and absolute), - Δ CTR, - Δ Avg. position (improvements count more when volume is non-trivial), - light smoothing to reduce noise. - Rows below a small volume threshold are down-weighted to avoid “one-click spikes.” Use it for: quick wins, content to double-down on, titles that started working, new markets/devices gaining traction. Column sorting (manual) You can override the view’s default by clicking a column header: - First click: sort descending by that column. - Second click: flip to ascending. - Third click (or “Reset view”): return to the view’s default logic. Common useful sorts: - Avg. position (asc) → find near-top opportunities (positions 6–10). - CTR (desc) → pages/queries with compelling snippets; compare to impressions. - Impressions (desc) → “visibility without clicks” → improve titles/intent. Tip: Combine a view + column sort. Example: Growing + Avg. position (6–15) finds rising keywords that are close to page one. How comparisons work When a view shows ▲/▼ changes or uses Growing/Decaying: - Comparison window: Previous period of the same length, with day-of-week matching to avoid weekend/weekday bias. - Smoothing: minor smoothing for CTR/position signals at low volume. - Minimum volume: we require a small base of impressions/clicks before flagging growth/decay. Interactions with filters Sorting runs after filters: - Date range, Query, Page, Country, Device, Branded, and Topics/Clusters all constrain the dataset first. - Branded = Off often makes Growing/Decaying more actionable (no brand noise). - For large ranges, narrow by Country or Device to cut noise before sorting. Practical recipes - Find money terms near the top: Filters: Branded Off → View: Popular → Sort: Avg. position (asc) → focus on positions 6–10. - Double-down on winners: View: Growing → Sort: Clicks (desc) → open rows with strong ΔCTR or better position. - Triage losses fast: View: Decaying → Sort: Clicks (desc) → check pages/queries with worse position and falling CTR. - Market/device issues: Filters: choose Country/Device → View: Decaying → scan CTR and position to isolate intent/snip issues. FAQ Why does the sorted list differ from GSC UI? Topify loads the full dataset (up to 50,000 rows/day), while the GSC UI shows a sampled 1,000-row view. Totals and ordering can differ—Topify is more complete. Do exports keep my sort? Exports always respect your filters. The row order follows the table’s active sort when supported; otherwise, re-apply the sort in your spreadsheet. Why isn’t a tiny keyword flagged as Growing? We apply a small volume threshold and smoothing to avoid false positives from one-off clicks or very low impressions. Position got “worse” but CTR went up—what wins? In Growing/Decaying, multiple signals are blended. If CTR gains outweigh small position drift (and volume is there), the row can still rank as Growing.