PPC Projections & Forecasting

This week Matt Umbro (@Matt_Umbro) hosts PPCChat with another great question set titled “PPC Projections & Forecasting.” The following is the transcribed Streamcap from the live chat:

Q1: Are metric projections (i.e.: budget, revenue, traffic, etc) necessary when running PPC accounts? Why or why not?

  • Yes. All of them are intertwined. How can you project traffic w/o budget? How can you project rev./conversion w/o traffic? – Brooke Townsend (@btownsend13)
    • Even then I think it’s more about setting times of the year when you are focusing more on efficiency vs volume. – Matt Umbro
  • Metrics are often very important to the client, but difficult to nail down. – Kyle Crocker (@kacrocker)
  • They are necessary, but sometimes I think they are given too much weight – at the end of the day they are estimates. Estimates with educated guesses, but still estimates. But for the client, I understand why they are important to share with their teams. – Matt Umbro
  • On a grand level, yes, but getting too caught up in granular predictions can become unnecessarily distracting. – Timothy Jensen (@timothyjjensen)
  • Estimated clicks & CPC. Knowing what type of bid range you should expect for keywords helps to keep you on budget. – Nate Velazquez (@NaceVela)
  • Yes, unless you’re insane. – Steve Gibson (@stevegibsonppc)
  • I think projections are needed. I’m new to AdWords, so starting with budget. But the more metrics the better! – Terry Porter (@TPorter2)
  • I find budget projections are essential for me in managing well, & IMO KPI projections are less important, but clients like them. – Kirk Williams (@PPCKirk)
  • I usually provide ballpark projections as goals to work towards in that month/quarter/year. – Davis Baker (@davisbaker)
  • Projections should have a purpose. You’re asking for budget or you’re prepping them for meaningful changes in the coming Q. – Mark Irvine (@MarkIrvine89)
    • Yes true, projections as a means to an end are helpful. Projections for the sake of projections waste time. – Kirk Williams
    • Similar to Analytics, I see projections as not absolute, but rather trends. – Matt Umbro
  • It’s easy to get wrapped up in projections. Project what is valuable for the acct & if you can act on those projections. – Erika Schmidt (@erikapdx)
  • There is always an implicit projection or estimate whenever you decide to work on one task over another. – Richard Fergie (@RichardFergie)
  • Yes, sets expectations and then when things don’t go accordingly (which they often won’t), launches thoughtful conversation. – Aviva Downing (@AvivCuriously)
  • I think they are a great guideline for assuming what you can expect v. what you’re asking for. – Amy Valleskey (@amy_valleskey)
  • So only those who never make the decision on what tasks to work on work without projections. – Richard Fergie
  • We look at overall business goals, and work backwards to see what’s obtainable/realistic but still seeing growth. – Emma Franks (@akaEmmaLouise)
    • Exactly – they are educated guesses based upon your data and vertical. – Matt Umbro

Q2: What factors go into determining your budget projections? Why?

  • We look at overall business goals, and work backwards to see what’s obtainable/realistic but still seeing growth. – Brooke Townsend
  • Client’s pre-defined budget, anticipated search volume, historical data if available. – Timothy Jensen
  • There could be many factors, but at core, it’s Seasonal impr volume, CPC and conversion type. – Glenn Schmelzle (@heyglenns)
  • I use historical data as the biggest variable when determining projections – gotta use what account history has showed. – Matt Umbro
  • Product margins and profit per order paired with sales history for individual brands. – David Cox (@dcoxdesigns)
  • Project scope, search volume, potential impact to business goals, est target CPA. – Maria Corcoran (@mariacorcoran)
  • Cashflow. Because if client has cashflow sorted, their marketing budget should be infinite.  (Assuming, of course, there’s no limit to the number of clients they can handle.) – Steve Gibson
  • Seasonality, growth expectation, historical data when applicaple. – Amy Valleskey
  • I have a handy excel sheet to calculate budget. You grab like 5 metrics & it calculates budget for you. Major nerdery that I love. – Erika Schmidt
  • Many a time, competitors’ budgets matter too. – Rohan Ayyar (@searchrook)
  • Total budget, how much they want to spend per day, CPC. – Terry Porter
  • I also follow the 70/20/10 rule, 70% of budget is for tried and true, 20% is for safe bets, 10% is for new initiatives. – Matt Umbro
  • In yearly growth projections, def use historical data in addition to calc imprvt of performance based on CPA or rev goals. – Maria Corcoran
  • Current performance (Impr, CTR, IS, CPL, Cost) as well as lost IS for potential expansion. – Emma Franks
  • Seasonality, growth expectation, CVR & CPA, CPC expectation. In-house so I dont consider budgetary limitation, rather needs. – Rachel King (@rachelking237)

Q3: Do you tend to set 1 revenue/ROAS/ROI/CPA goal per account or do you breakout by channel? Why or why not?

  • Depends on the account. For accounts with complicated products, breaking out ROAS is very important IMO for accurate analysis. I had client struggling to meet ROAS goals. We broke out their “complicated” brands, and remaining ROAS was beating goals. – Kirk Williams
  • I am running an AdWords account for my employer, so just the 1 account. Will be watching to see how others do it. – Terry Porter
  • Was a great signal to both of us to treat them separately to get more accurate idea of ROAS. Now I report them completely separate. – Kirk Williams
  • Hard to answer as in-house, but we have an overall marketing ROAS, with specific ROAS goals for PPC by campaign. – Brooke Townsend
  • Depends on the overall goal. Sometimes you have to use the high margin channel to make up for the high volume, break even ones. – Jesse Semchuck (@jessesem)
  • Goals often are dependent on budget. I can’t honestly pitch an assisted conversion path to someone with a tiny budget. – Mark Irvine
  • By channel; in most cases, clients don’t want to ignore other channels in favor of the best performing one. – Rohan Ayyar
  • 100% by act for opt. Products will have different REV value. Also audience can vary w/LTV. Brand goals should be set diff too. – Maria Corcoran
  • I will break out goals by campaign/category, but generally, clients want to see an exec summary overview. – Matt Umbro
  • I run ROAS by brand to consider AOV of product grouping under each then manage P/L by channel. Team runs CPA by channel by brand. – Rachel King
  • I like breaking PPC ROI goals out by brand terms & non-brand terms & have account structure that supports that measurement. – Tim Mayer (@timmayer)
  • If channels have diff purposes, then breakout goals. Awareness channels can’t have the same rev as a purchase-level channel. – Erika Schmidt
  • Also depends on time of month – measure Net ROI to campaign by device monthly but CPA daily. I don’t work off assumed ROI. – Rachel King

Q4: Do you provide multiple projections based upon varying goals (i.e.: efficiency, volume, brand awareness)? Why or why not?

  • I do different projections based on scenarios. IE: growth of 3x, efficiency gain of X, etc. A good way to look at all angles. – Brooke Townsend
  • Try to align projections (and reporting) to goals to close the loop and fit with client’s priorities. – Kyle Crocker
  • I think you can project based on a campaign’s purpose, is it awareness only, purchase intent, competitive? – Terry Porter
  • I run 5 What-ifs based on changes to KPIs, being in-house. In prior agency life would provide qtrly scenarios for growth. – Rachel King
  • I’ll provide projections based upon potential budget increases/decrease. – Matt Umbro
  • No. I believe a campaign should have one main goal. – Steve Gibson
  • Not necessarily based on varying goals, but based on varying factors that could influence outcome. – Emma Franks
  • I really prefer “What-ifs” to projections. Change in KPIs is typically a result of change in the account. – Mark Irvine
  • Choose one primary goal and project for that. Secondary goals are ok to project for but must be clear that those are secondary. – Aviva Downing
  • TBH, not a whole lot. I see slight shifts, but we see more shifts in consumer behavior than Google updates. – Rohan Ayyar

Q5: How much do industry factors play into your projections (i.e.: algorithm changes, right side ad update, etc)?

  • TBH, not a whole lot. I see slight shifts, but we see more shifts in consumer behavior than Google updates. – Brooke Townsend
  • If there’s a significant new feature making a new channel worthwhile investing in, allocate new budget. – Timothy Jensen
  • Google constantly intros uncertainty into the mix. Vastly impactful and impossible to predict from historics. Hence What-ifs! – Rachel King
  • The only constant is change – create your projections with the knowledge and information you have. – Matt Umbro
  • When AdWords changes things, it makes a big difference. Text ads to enhanced ads, etc. Especially when you are new to ppc! – Terry Porter
  • If you’re seeing changes in other accounts, it’s sensical to see it affecting your projections. If not, then no. – Mark Irvine
  • Projections might lay a base for understanding industry (and other) changes – especially if they are made from historical data. – Kyle Crocker
  • More in presentation of data: this is the story of what COULD happen, but industry is always changing so be willing to adjust. – Emma Franks
  • Current team mantra: what we knew yesterday Google may change tomorrow. I.e. Yellow Ad labels running again today. – Rachel King
  • Some make more difference than others, of course. Difficult to predict & take into account nature & scale of algo changes. – Rohan Ayyar
  • They can have an indirect impact on ROI which then has a knock-on effect on budget projections. – Bhavek Rughani (@bhavekrughani)

Q6: How closely should you be held responsible for the projections you create? Why?

  • It’ll always be a chase to optimize to the newest update – adapt and adjust as necessary. – Amy Valleskey
  • Assuming you set proper expectations (including saying that projections are estimated best guesses). I would say that you should be held responsible to come within at least 10% of projections. – Matt Umbro
  • Everything is a product of expectations. Set proper ones & use appropriate disclaimers/contextualizing to be in best spot. – Julie Bacchini
  • Like it or not, you’re going to be held to these numbers to an extent. This is your chance to under promise & over deliver. – Mark Irvine
  • Certain clients I KNOW will use it as a fact, others use it as a goal. Know your clients and communicate your hesitations. – Tim Halloran (@timmhalloran)
  • As a data-driven educated guess that is never 100% accurate, I try to have 85%+ confidence. Im never held to them internally! – Rachel King
  • In-house answer: you’re always going to be held responsible for projections, while basing it off of overall business goals. – Brooke Townsend
  • Again, as long as you set proper expectations with projections, you should be held responsible. – Matt Umbro
  • Gives you a chance to prove yourself though. I’ve had to prove that some things just aren’t possible given the goals. – Brooke Townsend
  • Never, ever over promise. Even when the CEO tries to force a concrete number in front of an audience. – Rachel King
  • Which is where it depends on analysts confidence in prediction and how this was communicated. Say “this will 100% happen” and then it doesn’t – line up and put your head on the block. – Richard Fergie
  • I am new at this, so I hope that I will be given a chance to show improvement. I projected X and we got Y. Make that better! – Terry Porter
  • You should always be held responsible but it does help to have flexibility otherwise you can over obsess. – Bhavek Rughani
  • ALWAYS specify (include in writing) assumptions used for any projections! – Julie Bacchini (@NeptuneMoon)
  • At least in-house there is more subjectivity. Ie volume 10% short but AOV was up 15%, Net P/L grew 1% after we cut X expense. – Rachel King

PPCChat Sponsored by

CallRail LogoCallRail is a call tracking platform that brings enterprise-level call analytics to businesses and agencies. CallRail makes it easy to track which marketing sources and keywords make your phone ring. We provide call tracking, recording, and analytics for PPC, SEO, web, and offline marketing campaigns. With CallRail, you can create tracking phone numbers instantly, get reports in real time, and increase your advertising ROI by learning which campaigns and keywords deliver valuable phone leads.

CallRail

Resources

More PPCChats

Don’t forget to stay tuned for the next #PPCchat on Tuesday at 12 noon Eastern, 9 am Pacific and 5pm in the UK. Same Chat time, same Chat channel.

Participants

Check out the PPCChat Twitter list to see and connect with all current and prior participants.

• Matt Umbro (@Matt_Umbro)
• Paul Kragthorpe (@PaulKragthorpe)
• Amy Valleskey (@amy_valleskey)
• Aviva Downing (@AvivCuriously)
• Bhavek Rughani (@bhavekrughani)
• Brooke Townsend (@btownsend13)
• David Cox (@dcoxdesigns)
• Davis Baker (@davisbaker)
• Emma Franks (@akaEmmaLouise)
• Erika Schmidt (@erikapdx)
• Glenn Schmelzle (@heyglenns)
• Jesse Semchuck (@jessesem)
• Julie Bacchini (@NeptuneMoon)
• Kirk Williams (@PPCKirk)
• Kyle Crocker (@kacrocker)
• Maria Corcoran (@mariacorcoran)
• Mark Irvine (@MarkIrvine89)
• Nate Velazquez (@NaceVela)
• Rachel King (@rachelking237)
• Richard Fergie (@RichardFergie)
• Rohan Ayyar (@searchrook)
• Steve Gibson (@stevegibsonppc)
• Terry Porter (@TPorter2)
• Tim Halloran (@timmhalloran)
• Tim Mayer (@timmayer)
• Timothy Jensen (@timothyjjensen)
 

Streamcap Projection, Forecasting, and Creation

This is a guest post by Paul Kragthorpe; works at WebRanking in Minneapolis, Minnesota.
Connect with Paul @PaulKragthorpe, and Google Plus.

Be Sociable, Share!

Tags: , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *

Sponsored By

Recent Tweets



I am speaking at SMX East
PPCChat.co was rated one of the Best PPC Blogs by Boost CTR