Tabellab

Hook Rate & Hold Rate: 2-Second Views, 6-Second Views and a Diagnosis for Every TikTok Ad

Lesson 1 of 20Lab

All lessons
  1. Hook Rate & Hold Rate: 2-Second Views, 6-Second Views and a Diagnosis for Every TikTok AdLabFree
  2. Watch Depth: Quartile Views, Completion Rate and Share of the Video Watched (Column Chart)Lab Locked
  3. The Price of Attention: CPM, CPV and Cost per 6-Second View by Ad GroupLab Locked
  4. From View to Purchase: Landing Rate, Purchase Rate and CPA per CampaignLab Locked
  5. What a Purchase May Cost: Contribution, Break-Even CPA, Break-Even ROAS and a Target CPALab Locked
  6. Creative Scorecard: Format Benchmarks with XLOOKUP, Hook Index, CPA Index and a Verdict (Conditional Formatting)Lab Locked
  7. Creative Fatigue: Daily Hook Rate, 3-Day CPA, Loss Against the Launch Baseline and a SLOPE Trend (Line Chart)Lab Locked
  8. Boosted Creator Posts: Paid CPV, All-In Cost per 1,000 Views and All-In CPA Against the Brand's Own Ads (Bar Chart)Lab Locked
  9. Bidding: Lowest Cost, Cost Cap and Bid Cap, Delivery, CPA Against the Goal and an Action per Ad GroupLab Locked
  10. Budgets and the Learning Phase: Expected Purchases per Week, Minimum Daily Budget and How Many Ad Groups You Can AffordLab Locked
  11. Budget Pacing: Cumulative Spend, an Even-Pace Plan with EOMONTH, Pace Index and a Month-End Projection (Line Chart)Lab Locked
  12. Attribution Windows: 1-Day Click vs 7-Day Click + 1-Day View, View-Through Share and What a Window Does to the CPALab Locked
  13. Conversion Lag: a Lag Curve, Maturity per Reporting Day, Projected Purchases and Projected CPA (Line Chart)Lab Locked
  14. Creative Split Tests: Lift in Purchase Rate, Pooled Standard Error, z-Score and a VerdictLab Locked
  15. Creative Testing Lane: Test Budget per Creative, Kill Rules, the Chance of Zero Purchases and Graduation (Conditional Formatting)Lab Locked
  16. Creative Rotation Plan: Days Live, Expected Retire Dates, Status and the Weekly Number of New CreativesLab Locked
  17. Audience Types and Countries: Spend, CPA, ROAS and Share of Spend with SUMIFS (Bar Chart)Lab Locked
  18. Frequency and Reach: Cost per 1,000 People, the Link Between Frequency and CTR with CORREL and a Forecast (Scatter Chart)Lab Locked
  19. What-If Grid: CPA by CPM and Purchase Rate with Mixed References, the Break-Even Zone and the Highest Affordable CPM (Conditional Formatting)Lab Locked
  20. Final Project: Account Dashboard with Hook Rate, Blended CPA, CPA by Type, the Best Ad and the Ads Above Break-Even (Column Chart)Lab Locked

You are not signed in — the lab works, but your progress is not saved. Sign up

Lesson 1 of 20 is free. Unlock all 20 hands-on labs, the final exam and your certificate.

Buy course — $9.90
Loading the spreadsheet…

Why the first two seconds decide a TikTok ad

On a short-video feed nobody has to sit through your ad. A thumb moves, and the video is gone. That makes the opening moments of a TikTok ad far more important than in most other formats: if people do not stop, the product, the offer and the call to action are never seen. Yet many advertisers still judge their videos only by the last number in the report, the cost per purchase, and then wonder why one video works and the next one does not.

Two simple ratios open that black box. The hook rate tells you how many people stopped scrolling. The hold rate tells you how many of those stayed for the next few seconds. Together they show where a video loses its audience, and that tells the creative team what to change. This free lesson explains both ratios, works through a complete example in Excel and ends with a diagnosis for every ad.

The example is Tindlewick Tea Co., a fictional online shop for loose-leaf tea, matcha and teaware that advertises with short videos. All ads, creators, numbers and results are invented for learning, and the benchmarks are illustrative values, not official platform data. The twelve ads below are the ones in the lab sheet of this lesson.

The two ratios: hook rate and hold rate

An ad report counts how often the video was shown (impressions) and how many of those impressions were watched for at least two seconds and at least six seconds. Those three numbers form a small funnel, and each ratio compares one step with the step before it.

  • Hook rate = 2-second views ÷ impressions. Of all the times the video appeared on a screen, how often did someone stay for two seconds? This is a verdict on the opening: the first frame, the first movement, the first words, the text on screen.
  • Hold rate = 6-second views ÷ 2-second views. Of the viewers who were hooked, how many were still there at second six? This is a verdict on what follows the opening: does the video deliver what the first second promised, and quickly enough?

Notice the denominators. The hook rate divides by impressions, the hold rate by 2-second views. Because the hold rate ignores everyone who swiped away immediately, it stays fair even when the hook is weak. A video can have a poor hook and an excellent hold: few people stop, but those who do are interested. It can also be the other way round, and that case is common with attention-grabbing openings that have little to do with the product.

Definitions differ between teams and tools. Some measure the hook at three seconds, some call the second ratio "retention" or measure it at a different point. None of that is wrong. What matters is that you pick one definition, write it next to the table and use it for every ad, so that the comparison is fair.

From impression to 6-second view: all twelve ads togetherImpressions481,7002-second views159,7636-second views66,166Hook rate = 159,763 ÷ 481,700 = 33.2 %Hold rate = 66,166 ÷ 159,763 = 41.4 %66.8 % of impressions are swiped away within two seconds
The attention funnel of the account. The hook rate compares the second bar with the first, the hold rate the third with the second.

A fully worked example: twelve Tindlewick Tea Co. ads

Here are the twelve video ads of the last 14 days with the three counts from the ad report and the three calculated columns of the lab. Tindlewick uses two house benchmarks: a hook rate of at least 30 % and a hold rate of at least 40 %. Both are illustrative; your own benchmarks should come from your own account history.

AdFormatImpressions2-second views6-second viewsHook rateHold rateDiagnosis
Iced matcha in 20 secondsCreator48,20017,4488,21836.2 %47.1 %Scale
Founder story: why loose leafFounder31,5006,7413,51921.4 %52.2 %Fix hook
Tea drawer tourCreator52,80017,8465,56833.8 %31.2 %Fix body
Cold brew bottle demoDemo44,10012,8335,83929.1 %45.5 %Fix hook
3 mistakes with green teaCreator61,30025,62311,17241.8 %43.6 %Scale
Unboxing the samplerUnboxing27,6008,4183,26630.5 %38.8 %Fix body
Chai latte at homeDemo39,40013,6726,72734.7 %49.2 %Scale
Customer review mashupReview22,9004,2821,46018.7 %34.1 %Fix hook
Morning routineCreator57,70018,8107,56232.6 %40.2 %Scale
Steep timer trickDemo35,20015,9104,56645.2 %28.7 %Fix body
Gift box revealUnboxing18,4004,9312,34726.8 %47.6 %Fix hook
Sleep blend night routineCreator42,60013,2495,92231.1 %44.7 %Scale
All twelve ads481,700159,76366,16633.2 %41.4 %

Altogether the ads were shown 481,700 times. 159,763 impressions lasted at least two seconds and 66,166 at least six. About two thirds of all impressions were over before the viewer could even read the first caption, which is completely normal for a feed and exactly why the opening deserves so much attention.

Calculating the hook rate step by step

In the lab sheet the impressions are in column C and the 2-second views in column D. The hook rate of the first ad goes into F4 as =D4/C4. Format the cell as a percentage and copy the formula down to F15. For the founder story the result is 6,741 ÷ 31,500 = 21.4 %: only about one impression in five survives two seconds. The best hook belongs to the steep timer trick with 45.2 %.

A quick plausibility check saves embarrassment later: a hook rate can never be above 100 %, because a 2-second view is always also an impression. If you see 4.67 or 467 % in the cell, the division is upside down.

Hook rate by adScaleFix hookFix bodyBenchmark 30 %Steep timer trick45.2 %3 mistakes with green tea41.8 %Iced matcha in 20 seconds36.2 %Chai latte at home34.7 %Tea drawer tour33.8 %Morning routine32.6 %Sleep blend night routine31.1 %Unboxing the sampler30.5 %Cold brew bottle demo29.1 %Gift box reveal26.8 %Founder story: why loose leaf21.4 %Customer review mashup18.7 %0 %10 %20 %30 %40 %50 %
Hook rate per ad against the 30 % benchmark. The colour shows the final diagnosis: an ad left of the dashed line is always a "Fix hook" case, whatever its hold rate.

Calculating the hold rate

The hold rate uses the next pair of columns: =E4/D4 in G4, copied down to G15. For the cold brew bottle demo it is 5,839 ÷ 12,833 = 45.5 %. Almost half of the hooked viewers stay, which is a good value for this account.

The most common mistake here is to divide the 6-second views by the impressions. For the same ad that gives 13.2 %. The number is not meaningless, it is the share of all impressions that reached six seconds, but it mixes the opening and the middle of the video into one figure. You would no longer know which part to repair. Keep the two ratios separate; if you want the combined figure, it is simply hook rate × hold rate.

Now look at the steep timer trick again. Its hook rate of 45.2 % is the best in the table, but its hold rate of 28.7 % is the worst. The opening works brilliantly and then more than two thirds of the hooked viewers leave. Without the second ratio this ad would look like the star of the account.

From two numbers to a diagnosis

Two ratios and two benchmarks give a clear decision path. Test the hook first. Only if the hook passes does it make sense to judge the hold rate, because a hold rate measured on a handful of accidental viewers says little about the video.

The diagnosis as a decision pathHook rateat least 30 %?Hold rateat least 40 %?Scale5 adsFix hook4 adsFix body3 adsyesyesnonoTest the hook first: a video nobody stops for cannot be judged on its middle part.
Two tests in a fixed order produce three labels. The counts are those of the worked example.

In Excel this is a nested IF. For row 4, with the benchmarks in K4 and K5: =IF(F4<$K$4,"Fix hook",IF(G4<$K$5,"Fix body","Scale")). The dollar signs lock the two benchmark cells so the formula can be copied down. If you prefer one formula without helper columns, replace F4 by D4/C4 and G4 by E4/D4.

  • Fix hook (4 ads). The opening does not stop people. Change the first frame, start with movement or with the result instead of the introduction, put the key phrase on screen in the first second. The founder story (21.4 %) and the customer review mashup (18.7 %) are the clearest cases.
  • Fix body (3 ads). People stop and then leave. The video is too slow after the hook, or the hook promised something the video does not deliver. Tighten the edit, bring the payoff forward, shorten the video. The tea drawer tour hooks 33.8 % but holds only 31.2 %.
  • Scale (5 ads). Both benchmarks are met. These ads have earned a closer look at their cost per purchase, which is where lessons 3 to 5 of the course continue.

Take the cold brew bottle demo as a borderline case. Its hook rate of 29.1 % misses the benchmark by less than one percentage point, so the label is "Fix hook" although its hold rate is well above the benchmark. That is the right reading: the body of this video already works, and a stronger first second could turn it into one of the best ads of the account. A label is a pointer to the cheapest improvement, not a grade.

The account hook rate: totals, not averages

What is the hook rate of the whole account? The tempting answer is the average of column F, which is 31.8 %. The correct answer divides the totals: =SUM(D4:D15)/SUM(C4:C15) = 159,763 ÷ 481,700 = 33.2 %.

The two results differ because the ads differ in size. "3 mistakes with green tea" has 61,300 impressions, "Gift box reveal" only 18,400. A plain average gives both the same weight. Dividing totals weights every ad by its impressions automatically, which is what actually happened in the account. The same rule applies to every ratio in this course: for a group, divide the sum of the numerators by the sum of the denominators.

Count the "Scale" ads with =COUNTIF(H4:H15,"Scale"). If you want the count without the helper column, SUMPRODUCT can test both conditions row by row: =SUMPRODUCT((D4:D15/C4:C15>=K4)*(E4:E15/D4:D15>=K5)).

Step by step in Excel

  1. Export impressions, 2-second views and 6-second views per ad for one fixed period and paste them next to the ad names. Use the same period for every ad.
  2. Enter the two benchmarks in their own cells (K4 and K5 in the lab). Never type them into formulas; you will want to change them.
  3. Hook rate in F4: =D4/C4. Hold rate in G4: =E4/D4. Format both as percentages with one decimal and copy down.
  4. Diagnosis in H4: =IF(F4<$K$4,"Fix hook",IF(G4<$K$5,"Fix body","Scale")), copied down.
  5. Account hook rate: =SUM(D4:D15)/SUM(C4:C15). Account hold rate, if you want it: =SUM(E4:E15)/SUM(D4:D15).
  6. Sort or filter by diagnosis and hand the two "Fix" lists to whoever edits the videos, with the ratio that failed next to each ad.

Typical errors and their warning signs

  • Dividing upside down. =C4/D4 returns values like 2.76. Any rate above 100 % is a signal to swap numerator and denominator.
  • Hold rate on impressions. =E4/C4 looks plausible but blends hook and hold. The hold rate divides by the 2-second views.
  • Unlocked benchmarks. Without dollar signs, K4 becomes K5, K6 and so on when copied down, and the comparison runs against empty cells. Write $K$4.
  • Tests in the wrong order. If the formula tests the hold rate first, an ad with a weak hook and a weak hold is told to fix its body. The hook comes first.
  • Averaging rates. =AVERAGE(F4:F15) is not the account hook rate. Divide totals.
  • Judging tiny ads. An ad with 300 impressions can show a hook rate of 50 % by luck. Wait for a few thousand impressions before you label anything.
  • Mixing definitions. Comparing a 2-second hook rate from one report with a 3-second rate from another makes one of the ads look worse for no reason.

How to read the result

Hook rate and hold rate are diagnostics. They explain why an ad performs the way it does, and they react within hours, long before there are enough purchases to judge a cost per acquisition. That makes them ideal for early decisions about new videos. They are not the goal, however. A video can hold attention superbly and sell nothing, and an unspectacular demonstration can sell very well to the few people who watch it.

Benchmarks deserve the same caution. Thirty and forty percent are Tindlewick's illustrative house values. Rates differ by product, audience, video length and placement, and platforms change how they count. The most useful benchmark is your own history: the median hook rate of your last fifty ads tells you more than any number from an article, including this one. Later in the course you will give each video format its own benchmark, because a founder talking to the camera and a fast creator clip should not be measured against the same line.

When a new ad has a strong hold rate but a weak hook, do not reshoot it. Cut three new openings for the same video and test them against each other. It is the cheapest creative test there is.

Quick reference: formulas in this lesson

Metric / functionWhat it doesExample
Hook rateShare of impressions watched for 2 seconds=D4/C4
Hold rateShare of 2-second viewers still watching at second 6=E4/D4
IF (nested)Two tests in order, three labels=IF(F4<$K$4,"Fix hook",IF(G4<$K$5,"Fix body","Scale"))
SUM ÷ SUMRate for a group of ads=SUM(D4:D15)/SUM(C4:C15)
COUNTIFCounts a label=COUNTIF(H4:H15,"Scale")
SUMPRODUCTCounts rows that pass two tests without a helper column=SUMPRODUCT((D4:D15/C4:C15>=K4)*(E4:E15/D4:D15>=K5))
$K$4Absolute reference: stays fixed when copiedBenchmark cells

Where you will use this

  • Weekly creative review. Sort all live ads by diagnosis. The meeting then discusses three short lists instead of thirty videos.
  • Briefs for creators and editors. "Hook rate 21 %, hold rate 52 %" is a precise brief: keep the video, replace the first two seconds.
  • Early reads on new videos. After a few thousand impressions the two rates already show whether a new video is worth more budget, days before the purchase data is reliable.
  • Boosted creator posts. The same ratios apply to creator videos you put budget behind, and they help decide which posts deserve a boost at all.

What you will build in the rest of this course

This opening lesson can be used without paying. Each of the nineteen lessons after it has its own live lab with a different Tindlewick data set and builds on the two ratios you calculated today, from watch depth and the price of attention to bidding, budgets, attribution and a final account dashboard.

  • Watch Depth: Quartile Views, Completion Rate and Share of the Video Watched (Column Chart). You read quartile views, calculate completion rate and share watched per video, find where the account loses its viewers and chart the completion rates.
  • The Price of Attention: CPM, CPV and Cost per 6-Second View by Ad Group. You calculate CPM, cost per 2-second view and cost per 6-second view for ten ad groups, the account figure from totals, and find the cheapest ad group with INDEX and MATCH.
  • From View to Purchase: Landing Rate, Purchase Rate and CPA per Campaign. You follow ten campaigns from clicks to landing page views to purchases, calculate landing rate, purchase rate and CPA, and the account figures from totals.
  • What a Purchase May Cost: Contribution, Break-Even CPA, Break-Even ROAS and a Target CPA. You calculate the contribution before ads for nine product bundles, turn it into a break-even CPA, a break-even ROAS and a target CPA, and give each product a verdict.
  • Creative Scorecard: Format Benchmarks with XLOOKUP, Hook Index, CPA Index and a Verdict (Conditional Formatting). You look up the hook benchmark and CPA goal of each creative's format, calculate two indexes, give every creative a verdict with IFS and colour the ones to retire.
  • Creative Fatigue: Daily Hook Rate, 3-Day CPA, Loss Against the Launch Baseline and a SLOPE Trend (Line Chart). You track one ad over 14 days, smooth its CPA over three days, measure the loss of hook rate against the launch baseline, fit a trend with SLOPE and find the day to refresh.
  • Boosted Creator Posts: Paid CPV, All-In Cost per 1,000 Views and All-In CPA Against the Brand's Own Ads (Bar Chart). You cost ten boosted creator posts: paid CPV, all-in cost per 1,000 views and all-in CPA including the creator fee, count the posts that beat the brand's own ads and chart them.
  • Bidding: Lowest Cost, Cost Cap and Bid Cap, Delivery, CPA Against the Goal and an Action per Ad Group. You check twelve ad groups with three bid strategies: delivery of the budget, CPA against the goal, an action per ad group and spend and CPA per strategy with SUMIFS.
  • Budgets and the Learning Phase: Expected Purchases per Week, Minimum Daily Budget and How Many Ad Groups You Can Afford. You test ten ad groups against a learning threshold, calculate the minimum daily budget with ROUNDUP, the budget gap with MAX and the number of ad groups the account budget can support.
  • Budget Pacing: Cumulative Spend, an Even-Pace Plan with EOMONTH, Pace Index and a Month-End Projection (Line Chart). You build a pacing sheet for one month: cumulative spend, an even-pace plan from dates, a pace index, a month-end projection and the daily spend needed to land on budget.
  • Attribution Windows: 1-Day Click vs 7-Day Click + 1-Day View, View-Through Share and What a Window Does to the CPA. You count the same purchases under two attribution windows, calculate both CPAs per campaign, the view-through share, the CPA change and how many click purchases arrive within seven days.
  • Conversion Lag: a Lag Curve, Maturity per Reporting Day, Projected Purchases and Projected CPA (Line Chart). You build a lag curve from finished data, look up how mature each recent day is, project its final purchases and CPA, and chart the curve.
  • Creative Split Tests: Lift in Purchase Rate, Pooled Standard Error, z-Score and a Verdict. You evaluate eight creative split tests: lift in purchase rate, pooled standard error, z-score and a verdict that separates real winners from noise.
  • Creative Testing Lane: Test Budget per Creative, Kill Rules, the Chance of Zero Purchases and Graduation (Conditional Formatting). You run a creative test lane: the chance that a creative shows no purchase by luck, a kill, keep or graduate decision with IFS, the test budget per creative and how many creatives a week can fund.
  • Creative Rotation Plan: Days Live, Expected Retire Dates, Status and the Weekly Number of New Creatives. You manage a library of 14 live creatives with dates: days live, expected retire date by format, a status per creative, the replacements due within a week and the weekly production need.
  • Audience Types and Countries: Spend, CPA, ROAS and Share of Spend with SUMIFS (Bar Chart). You summarise sixteen ad groups by audience type with SUMIFS: spend, CPA, ROAS and share of spend, calculate a CPA for one type in one country and chart the result.
  • Frequency and Reach: Cost per 1,000 People, the Link Between Frequency and CTR with CORREL and a Forecast (Scatter Chart). You calculate frequency, CTR and cost per 1,000 people reached for twelve weeks, measure how frequency and CTR move together with CORREL, forecast the CTR at a higher frequency and chart it.
  • What-If Grid: CPA by CPM and Purchase Rate with Mixed References, the Break-Even Zone and the Highest Affordable CPM (Conditional Formatting). You build a fifty-cell CPA grid from one formula with mixed references, count the scenarios below break-even, solve for the highest affordable CPM and colour the loss zone.
  • Final Project: Account Dashboard with Hook Rate, Blended CPA, CPA by Type, the Best Ad and the Ads Above Break-Even (Column Chart). You build the monthly dashboard for all ads and boosted creator posts: account hook rate, blended CPA, CPA per type, the best ad by name, the ads above break-even and a chart.

Summary

The hook rate, 2-second views ÷ impressions, measures whether the opening stops the scroll. The hold rate, 6-second views ÷ 2-second views, measures whether the next seconds keep the hooked viewers. A nested IF turns the two into a diagnosis: fix the hook, fix the body or scale. For the account, divide totals instead of averaging rates. At Tindlewick Tea Co., 5 of twelve ads pass both benchmarks, 4 need a new opening and 3 lose their viewers after it, and the account hook rate is 33.2 %.

The lab at the top of this page holds the same twelve ads with the three result columns left empty. Type the formulas, let the Coach confirm each one, and then try the structure on an export from your own ad account: impressions, 2-second views and 6-second views are all it needs.

Frequently asked questions

What is a hook rate in TikTok ads?

The hook rate is the share of impressions that were watched for at least two seconds: 2-second views divided by impressions. It shows how well the opening of a video stops people from scrolling on. Some teams measure it at three seconds; use one definition consistently.

How do I calculate the hold rate?

Divide the 6-second views by the 2-second views, for example =E4/D4 in Excel. The hold rate only looks at viewers who were already hooked and tells you how many of them kept watching, so it judges the part of the video that follows the opening.

What is a good hook rate?

There is no universal value. It depends on product, audience, video format and how the platform counts views. The course example uses an illustrative house benchmark of 30 percent. The most reliable benchmark is the median of your own recent ads.

Why should I not average the hook rates of my ads?

A plain average gives a small ad the same weight as a large one. For the hook rate of an account or campaign, divide the total 2-second views by the total impressions, in Excel =SUM(D4:D15)/SUM(C4:C15).

What should I change when the hook rate is good but the hold rate is low?

The opening works, but viewers leave right after it. Usually the video is too slow after the hook or does not deliver what the first second promised. Tighten the edit, bring the payoff forward or shorten the video, and keep the opening.

Do hook rate and hold rate predict sales?

Not on their own. They explain where a video loses attention and react quickly, which makes them useful for early creative decisions. Whether an ad is profitable is decided by its cost per purchase compared with the margin of the product, which later lessons cover.

Keep going: TikTok Ads & Short-Video Advertising: Hook Rate, Hold Rate, CPV, CPA, Creative Rotation, Bidding & Attribution Windows

This free lab is lesson 1 of 20. Unlock the other 19 hands-on labs, the final exam and your certificate for a one-time $9.90.

  1. Watch Depth: Quartile Views, Completion Rate and Share of the Video Watched (Column Chart)
  2. The Price of Attention: CPM, CPV and Cost per 6-Second View by Ad Group
  3. From View to Purchase: Landing Rate, Purchase Rate and CPA per Campaign
  4. What a Purchase May Cost: Contribution, Break-Even CPA, Break-Even ROAS and a Target CPA
  5. Creative Scorecard: Format Benchmarks with XLOOKUP, Hook Index, CPA Index and a Verdict (Conditional Formatting)
  6. Creative Fatigue: Daily Hook Rate, 3-Day CPA, Loss Against the Launch Baseline and a SLOPE Trend (Line Chart)
  7. Boosted Creator Posts: Paid CPV, All-In Cost per 1,000 Views and All-In CPA Against the Brand's Own Ads (Bar Chart)
  8. Bidding: Lowest Cost, Cost Cap and Bid Cap, Delivery, CPA Against the Goal and an Action per Ad Group
  9. Budgets and the Learning Phase: Expected Purchases per Week, Minimum Daily Budget and How Many Ad Groups You Can Afford
  10. Budget Pacing: Cumulative Spend, an Even-Pace Plan with EOMONTH, Pace Index and a Month-End Projection (Line Chart)
  11. Attribution Windows: 1-Day Click vs 7-Day Click + 1-Day View, View-Through Share and What a Window Does to the CPA
  12. Conversion Lag: a Lag Curve, Maturity per Reporting Day, Projected Purchases and Projected CPA (Line Chart)
  13. Creative Split Tests: Lift in Purchase Rate, Pooled Standard Error, z-Score and a Verdict
  14. Creative Testing Lane: Test Budget per Creative, Kill Rules, the Chance of Zero Purchases and Graduation (Conditional Formatting)
  15. Creative Rotation Plan: Days Live, Expected Retire Dates, Status and the Weekly Number of New Creatives
  16. Audience Types and Countries: Spend, CPA, ROAS and Share of Spend with SUMIFS (Bar Chart)
  17. Frequency and Reach: Cost per 1,000 People, the Link Between Frequency and CTR with CORREL and a Forecast (Scatter Chart)
  18. What-If Grid: CPA by CPM and Purchase Rate with Mixed References, the Break-Even Zone and the Highest Affordable CPM (Conditional Formatting)
  19. Final Project: Account Dashboard with Hook Rate, Blended CPA, CPA by Type, the Best Ad and the Ads Above Break-Even (Column Chart)

Buy course — $9.90 See the full course