|Setting SEO goals & forecasting traffic|
How do I forecast traffic from natural search?
I have been tasked with writing an SEO plan for my company. I have learned a great deal about what SEO is about and how to do it, but I am at a loss as to making forecasts about its effectiveness and how to set traffic goals related to natural search.
There are many professional SEO organizations out there and there are even more SEO managers inside of corporate enterprises. What are these people telling their clients about what they can deliver? How do they set goals for traffic growth? What type of metics are used to measure success in this arena?
This question must be on a lot of minds as companies are demanding see ROI for their SEO investment.
Please share your thoughts.
[edited by: msgraph at 1:04 pm (utc) on May 2, 2005]
[edit reason] no signatures please [/edit]
I have recently inherited SEO initiatives at my company and also find it difficult to create a forecasting model.
One thing I will say is that an integral, if not basic, part of creating your SEO plan is defining the goal/objectives of SEO for your company, and then identifying KPI's that are congruent with measuring said objective. For example, we are less concerned with setting traffic targets and more concerned with tieing back and increasing top line sales attributable to our SEO efforts.
As there are many SEO "dials" that can be tweaked, the relative cost/benefit of each, coupled with a deliverability schedule will be a focus for our plan as well. That way we can ensure we are getting the biggest bang for our SEO buck.
I apologize for not providing any tangible answers to your question. I hope we can both find the help we seek regarding forecasting.
It's just part of the corporate process. In every meeting that I have attended, each division reports their current results and projects future results.
It's a very rare occurrence to have a projection that 100% materializes. The truth is there are no guarantees with SEO. People who do this everyday know there are no gurantees. But, corporate policies and procedures dictate that something - hopefully positive - should be stated in a written report, even if it isn't guaranteed. Above all, it is a paper trail that leads back to someone.
I think it is enought if the CEO or your company understands the potencial of SEO, even if there is not guaranteed for it.
SEO is like selling your products illegaly on the street and not being sure if the police will come tomorrow and get you arrested - or in this case if Google is gonna change the algo and your forecast goes bananas.
The only available tool is empiric forecast. If you are targeting hundrets of keywords or keyphrases, group them by number of words and test them in one categorie, mess the visits and estimate your total.
The best forecast for us is not the total amount of visites comming from natural search but HOW MUCH we save in comparison with paid visits.
hope it helps.
on the matter of tangible and meaningful reports and predictions - i myself have to create them, not to hand in at a meeting, but to know how long i've got before i stop making enough money to eat.
if you have an individual page, say a free-info-page which has advertising content on it, you can easily...
1. get it into google for free, quickly
2. measure, over the course of a week, the average number of clicks to that page per day
3. measure, over the course of a month, the average number of advertising clicks per exposure
4. measure, over the course of 6 months, the average number of dollars per click
if instead of free content and ad revenue you have commercial content, you simply need to replace steps 3 and 4 with average number of penultimate-step-to-sale clicks in a month and average number of sales from those clicks in 6 months
like in most sciences, none of which is any more exact than any other, you actually get general rules which apply to 'normal' pages...
10 pages get 1 click per day between them
3000 ad exposures get 1 click
a few hundred adclicks gets 1 sale
and, for irregular verbs the following applies:
a few hundred product page exposures gets 1 sale
(all estimates, btw, are thoroughly conservative).
as with verbs also, the above laws are 1. definitely localised to an area of time - within another 5 years of this post the figures are likely to be v different; although they'll stabilise at a point when technology is very uniform (eg we've all got four-walled tv from the book) and 2. a matter of conjugation.
secondly, with regard to those figures i gave above...
10 pages get 1 click per day between them
3000 ad exposures get 1 click
a few hundred adclicks gets 1 sale
this pretty much applies to ALL pages which haven't been "beefed up" by "SEO" antics, something similar or amazing luck.
if, on the other hand, (to quote from dr strangelove), you take a single page, say for people who want to download free games, and you have 300 games listed on one single page - the likelihood is that for something like that, with so many different items all cross-referencing their connected keywords with each others', you'll end up, on a good day, with perhaps an average of 10 - 20 clicks on that page per day.
but either way, having to hand in these analyses to bosses or even clients is the bit i personally would not be able to deal with properly - sometimes, frequently in fact, days of agonising about what's not right are resolved by about 30 seconds of thinking things through at about 4am.
i don't think SEO can ever be a suit and tie affair.
if you were to get a 100% hold on it, why would
you keep your job, rather than going into business
and point 3: (there's always a point 3)
in the e-commerce world the only statistics which can be deemed to exist (at least as far as business planners and accountants can be concerned) are the ones where you have a list of how many dollars you got given in a month.
most web-marketeers cannot tell themselves how they're doing until 3 months after it's happened.
by then, of course, it's too late to do anything about it - so the rule we live by is "whatever you did yesterday to get today's hits must be done again today, for the sake of tomorrow".
the only sure thing you can ever say on monday morning is "on friday i'll still be doing this job, i won't be looking for employment yet" - but as for whether that holds to the following monday - it's a question of how many weeks of hell you can take all at once.
if you are estimating the natural traffic for many 10s of 1000s of pages, you'll also have to factor in the time it will take the engines to 'eat' the data - and for that experienced seo people will surely agree that a margin of error of about 1 to 3 years is always a good precaution against overestimating your luck.
Although you can (and should) measure your progress, it is always a "best efforts" affair.
I can not come up with a set number because 1)the numbers in increase of traffic is fairly rrelevant unless you are selling traffic and 2)it is a constant process of measuring, tuning and remeasuring to achieve the desired results.
Three things all senior executives understand.
Technology is not one of them.
The three are:
If you can show whatever you do will increase, protect, or maintain corporate assets, they will like it. Assets can be a piece of equipment, intellectual property, employees, etc.
This is the "good name" of the company in the public. If you can show that the project will maintain, protect, or increase the "good name" in public, you will find a strong champion in a hurry.
If your company is regulated (and despite what you think, ALL companies are regulated some way or shape), and you can show that your project will protect them from fines or even jail time, you are gold.
Of course, not all three will be available in every scenario.
From my experience, SEO goals should be measured through these measures:
1) Increased in qualified search engine traffic. (people that come to your site and spend time there looking for what you have to offer)
2) Increase in Click Through ratios for non-paid online services.
3) Online Saturation of your website
4) Online website promotion/ exposure to consumers searches
I hope this helps. It also depends if you are selling a service or a product. Different stratagies are required for both. All of these are measurable, and can be used to explain SEO goals.
I would definitley incorporate the importance of generating new content into your plan if your company seriously wants traffic. SEO can do a lot but it's a big pitfall to start giving people numbers for long term. You can almost always expect traffic if your pages are optimized and content rich. If you can talk your company into setting something up where everyone can help keep the content flowing, the traffic will always increase w/ the content. SEO ranking factors are always gonna change, but a search engine's goal to find the best content never will.
I am sorry guys but... to senior management CTR and visits per month, etc. means absolutely nothing...
Their eyes, will slowly glaze over, and roll back into their head to keep it moist... You will hear a slight noise... resembling a snore... Oh, wait.. it WOULD be snoring...
Yes, creating content, working on SEO/SEM is very important.
They don't give a monkey's behind.
They don't CARE! They care about the bottom line. Show that you are increasing sales by so many percentage. Show them you are increased brand awareness. You quantify that, and you are gold.
You will need a baseline, i.e. prior to SEO, and then an on-going report.
If you can correlate your SEO work to increased sales from the baseline, they will love you. Show them that prior to your work, the company was less known, but now is a "household" name. Start talking about click-throughs, and such, and they are gone...
the best policy is to not talk to them. or care for their money.
after all, how could you possibly keep abreast of all the disciplines and terminologies which actually count if you had to spend the best part of your day explaining it to people who will go to the grave ignorant?
your competitors, for the holy loot buried on the internet, are not, after all, these corporate wild animals - the people who could get the loot before you are not the investors but the programmers and scripters and coders, hacking away at so many keyboards.
just ignore the short-term assets of the corporate circus act and instead of being the circusmaster with the whip, live it rough for a few years and get the money for yourself. when the corporate lions and tigers have been fully drained by a combination of recession and ignorance, they'd have fired you anyway. but instead you'll have at least the capacity to carve out a minimal living and if your heart is really in it, you're bound to have a lot more.